From 04e143db96f4b52c48d7d2f3ebe736fcf2ecddea Mon Sep 17 00:00:00 2001 From: "Kyle M. Lang" Date: Thu, 12 Oct 2023 09:09:47 +0200 Subject: [PATCH 1/4] starting work on week 6 practical materials --- docs/404.html | 11 +- docs/a3_components.html | 11 +- docs/acknowledgement.html | 11 +- docs/assignment-1-path-analysis.html | 11 +- ...gnment-2-confirmatory-factor-analysis.html | 11 +- docs/assignments.html | 11 +- docs/at-home-exercises-1.html | 11 +- docs/at-home-exercises-2.html | 11 +- docs/at-home-exercises-3.html | 11 +- docs/at-home-exercises-4.html | 11 +- docs/at-home-exercises-5.html | 11 +- docs/at-home-exercises-6.html | 11 +- docs/at-home-exercises.html | 11 +- docs/attendance.html | 11 +- docs/cfa.html | 11 +- docs/course-data.html | 11 +- docs/course-overview.html | 11 +- docs/data_options.html | 11 +- docs/efa.html | 11 +- docs/elaboration-tips.html | 11 +- docs/full-sem.html | 11 +- docs/grading-1.html | 11 +- docs/grading.html | 11 +- docs/in-class-exercises-1.html | 11 +- docs/in-class-exercises-2.html | 11 +- docs/in-class-exercises-3.html | 11 +- docs/in-class-exercises-4.html | 11 +- docs/in-class-exercises-5.html | 1486 ++++++++++++++++- docs/in-class-exercises-6.html | 11 +- docs/in-class-exercises.html | 11 +- docs/index.html | 13 +- docs/installing-software.html | 11 +- docs/instructors.html | 11 +- docs/introduction-to-r.html | 11 +- docs/learning-goals.html | 11 +- docs/lecture-1.html | 11 +- docs/lecture-2.html | 11 +- docs/lecture-3.html | 11 +- docs/lecture-4.html | 11 +- docs/lecture-5.html | 11 +- docs/lecture-6.html | 11 +- docs/lecture.html | 11 +- docs/mediation-moderation.html | 11 +- docs/multiple-group-models.html | 11 +- docs/note-on-data-updates.html | 11 +- docs/procedures.html | 11 +- docs/reading-1.html | 11 +- docs/reading-2.html | 11 +- docs/reading-3.html | 11 +- docs/reading-4.html | 11 +- docs/reading-5.html | 11 +- docs/reading-6.html | 11 +- docs/reading-questions.html | 11 +- docs/reading.html | 11 +- docs/reference-keys.txt | 7 + docs/resources.html | 11 +- docs/rules.html | 11 +- docs/schedule.html | 11 +- docs/search_index.json | 2 +- docs/software-setup.html | 11 +- docs/statistical-modeling-path-analysis.html | 11 +- docs/typographic-conventions.html | 11 +- docs/weekly-preparation.html | 11 +- .../figure-html/unnamed-chunk-130-1.png | Bin 67452 -> 66993 bytes .../figure-html/unnamed-chunk-132-1.png | Bin 72013 -> 71495 bytes docs/wrap-up.html | 11 +- sections/week6/{w7_home.Rmd => class.Rmd} | 10 +- week6.Rmd | 3 +- 68 files changed, 2111 insertions(+), 70 deletions(-) rename sections/week6/{w7_home.Rmd => class.Rmd} (95%) diff --git a/docs/404.html b/docs/404.html index ad7fe8f4..a012c0ee 100644 --- a/docs/404.html +++ b/docs/404.html @@ -410,7 +410,16 @@
  • Questions
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
  • 6.3 At-Home Exercises
    • -
  • 6.4 In-Class Exercises
    • +
  • 6.4 In-Class Exercises +
  • 7 Multiple Group Models
      diff --git a/sections/week6/w7_home.Rmd b/sections/week6/class.Rmd similarity index 95% rename from sections/week6/w7_home.Rmd rename to sections/week6/class.Rmd index 9e628a31..866d3c01 100644 --- a/sections/week6/w7_home.Rmd +++ b/sections/week6/class.Rmd @@ -1,4 +1,4 @@ -## At-Home Exercises +## In-Class Exercises In these exercises, you will use full structural equation modeling (SEM) to evaluate the *Theory of Reasoned Action* (TORA), which is a popular psychological @@ -44,7 +44,7 @@ condom <- read.csv("toradata.csv", stringsAsFactors = TRUE) ``` ```{r, echo = FALSE} -#dataDir <- "../../data/" +dataDir <- "../../../../data/" condom <- read.csv(paste0(dataDir, "toradata.csv"), stringsAsFactors = TRUE) ``` @@ -405,6 +405,8 @@ predictor of *condom* use. + + --- End of At-Home Exercises @@ -542,4 +546,4 @@ End of At-Home Exercises [mi_tutorial]: https://www.lavaan.ugent.be/tutorial/groups.html [van_de_schoot_et_al]: https://doi.org/10.1080/17405629.2012.686740 -[putnick_bornstein]: https://doi.org/10.1016/j.dr.2016.06.004 \ No newline at end of file +[putnick_bornstein]: https://doi.org/10.1016/j.dr.2016.06.004 diff --git a/week6.Rmd b/week6.Rmd index 89a2542c..9b3f1fd0 100644 --- a/week6.Rmd +++ b/week6.Rmd @@ -27,6 +27,5 @@ knit_child(paste0(partDir, "lecture.Rmd"), quiet = TRUE) %>% cat(sep = "\n") knit_child(paste0(partDir, "reading.Rmd"), quiet = TRUE) %>% cat(sep = "\n") knit_child(paste0(partDir, "../home-placeholder.Rmd"), quiet = TRUE) %>% cat(sep = "\n") -knit_child(paste0(partDir, "../class-placeholder.Rmd"), quiet = TRUE) %>% - cat(sep = "\n") +knit_child(paste0(partDir, "class.Rmd"), quiet = TRUE) %>% cat(sep = "\n") ``` From ba285fa0589fab9f1e1a344a9483966d297728d8 Mon Sep 17 00:00:00 2001 From: "Kyle M. Lang" Date: Thu, 12 Oct 2023 13:54:04 +0200 Subject: [PATCH 2/4] more work on week 6 practicals --- docs/404.html | 23 +- docs/a3_components.html | 23 +- docs/acknowledgement.html | 23 +- docs/assignment-1-path-analysis.html | 23 +- ...gnment-2-confirmatory-factor-analysis.html | 23 +- docs/assignments.html | 23 +- docs/at-home-exercises-1.html | 23 +- docs/at-home-exercises-2.html | 23 +- docs/at-home-exercises-3.html | 23 +- docs/at-home-exercises-4.html | 23 +- docs/at-home-exercises-5.html | 457 +++++++++++++++++- docs/at-home-exercises-6.html | 23 +- docs/at-home-exercises.html | 23 +- docs/attendance.html | 23 +- docs/cfa.html | 23 +- docs/course-data.html | 23 +- docs/course-overview.html | 23 +- docs/data_options.html | 23 +- docs/efa.html | 23 +- docs/elaboration-tips.html | 23 +- docs/full-sem.html | 23 +- docs/grading-1.html | 23 +- docs/grading.html | 23 +- docs/in-class-exercises-1.html | 23 +- docs/in-class-exercises-2.html | 23 +- docs/in-class-exercises-3.html | 23 +- docs/in-class-exercises-4.html | 23 +- docs/in-class-exercises-5.html | 245 +++++----- docs/in-class-exercises-6.html | 23 +- docs/in-class-exercises.html | 23 +- docs/index.html | 23 +- docs/installing-software.html | 23 +- docs/instructors.html | 23 +- docs/introduction-to-r.html | 23 +- docs/learning-goals.html | 23 +- docs/lecture-1.html | 23 +- docs/lecture-2.html | 23 +- docs/lecture-3.html | 23 +- docs/lecture-4.html | 23 +- docs/lecture-5.html | 23 +- docs/lecture-6.html | 23 +- docs/lecture.html | 23 +- docs/mediation-moderation.html | 23 +- docs/multiple-group-models.html | 23 +- docs/note-on-data-updates.html | 23 +- docs/procedures.html | 23 +- docs/reading-1.html | 23 +- docs/reading-2.html | 23 +- docs/reading-3.html | 23 +- docs/reading-4.html | 23 +- docs/reading-5.html | 23 +- docs/reading-6.html | 23 +- docs/reading-questions.html | 23 +- docs/reading.html | 23 +- docs/reference-keys.txt | 11 +- docs/resources.html | 23 +- docs/rules.html | 23 +- docs/schedule.html | 23 +- docs/search_index.json | 2 +- docs/software-setup.html | 23 +- docs/statistical-modeling-path-analysis.html | 23 +- docs/typographic-conventions.html | 23 +- docs/weekly-preparation.html | 23 +- .../figure-html/unnamed-chunk-130-1.png | Bin 66993 -> 66832 bytes .../figure-html/unnamed-chunk-132-1.png | Bin 71495 -> 71801 bytes docs/wrap-up.html | 23 +- sections/week6/home.Rmd | 296 ++++++++++++ week6.Rmd | 5 +- 68 files changed, 1844 insertions(+), 552 deletions(-) create mode 100644 sections/week6/home.Rmd diff --git a/docs/404.html b/docs/404.html index a012c0ee..516341c3 100644 --- a/docs/404.html +++ b/docs/404.html @@ -409,16 +409,25 @@
    • Reference
    • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/a3_components.html b/docs/a3_components.html index 8e47dcf6..597a384d 100644 --- a/docs/a3_components.html +++ b/docs/a3_components.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/acknowledgement.html b/docs/acknowledgement.html index 052ecaef..7f4bc630 100644 --- a/docs/acknowledgement.html +++ b/docs/acknowledgement.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/assignment-1-path-analysis.html b/docs/assignment-1-path-analysis.html index 9416248b..2c7cf3da 100644 --- a/docs/assignment-1-path-analysis.html +++ b/docs/assignment-1-path-analysis.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/assignment-2-confirmatory-factor-analysis.html b/docs/assignment-2-confirmatory-factor-analysis.html index 77ecf756..d0fb758b 100644 --- a/docs/assignment-2-confirmatory-factor-analysis.html +++ b/docs/assignment-2-confirmatory-factor-analysis.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/assignments.html b/docs/assignments.html index 2fd83a88..f9aed89d 100644 --- a/docs/assignments.html +++ b/docs/assignments.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/at-home-exercises-1.html b/docs/at-home-exercises-1.html index 76d85812..6500b6b3 100644 --- a/docs/at-home-exercises-1.html +++ b/docs/at-home-exercises-1.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/at-home-exercises-2.html b/docs/at-home-exercises-2.html index be3d9b47..d60ffecf 100644 --- a/docs/at-home-exercises-2.html +++ b/docs/at-home-exercises-2.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/at-home-exercises-3.html b/docs/at-home-exercises-3.html index 207e87ea..6d8f65b5 100644 --- a/docs/at-home-exercises-3.html +++ b/docs/at-home-exercises-3.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/at-home-exercises-4.html b/docs/at-home-exercises-4.html index 91bac4e9..ac9e3df5 100644 --- a/docs/at-home-exercises-4.html +++ b/docs/at-home-exercises-4.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/at-home-exercises-5.html b/docs/at-home-exercises-5.html index 41d7e4d4..472b2aac 100644 --- a/docs/at-home-exercises-5.html +++ b/docs/at-home-exercises-5.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models @@ -452,7 +461,439 @@

    6.3 At-Home Exercises

    -

    Coming soon to a GitBook near you!

    +

    This week, wrap up our re-analysis of the Kestilä (2006) results. During +this practical, you will conduct a CFA of the Trust in Politics items and +compare the results to those obtained from your previous EFA- and PCA-based +replications of Kestilä (2006).

    +
    +
    +

    6.3.1

    +

    Load the ESS data.

    +
      +
    • The relevant data are contained in the ess_round1_trust.rds file. +
        +
      • This file is in R Data Set (RDS) format.
        • +
      • The dataset is already stored as a data frame with the processing and +cleaning that you should have done for previous practicals completed.
        • +
      • Check the documentation for the readRDS() function to see how you can +load data stored in an RDS file.
        • +
      • +
    +
    + +Click for explanation + +
    ess <- readRDS("ess_round1_trust.rds")
    +
    +
    +

    Unfortunately, we need to do a little data processing before we can fit the +regression model. At the moment, lavaan will not automatically convert a factor +variable into dummy codes. So, we need to create explicit dummy codes for the +two factors we’ll use as predictors in our regression analysis: sex and +political orientation.

    +
    +
    +
    +

    6.3.2

    +

    Convert the sex and political interest factors into dummy codes.

    +
    + +Click for explanation + +

    In R, we have several ways of converting a factor into an appropriate set of +dummy codes.

    +
      +
    • We could use the dplyr::recode() function as we did last week.
      • +
    • We can use the model.matrix() function to define a design matrix based on +the inherent contrast attribute of the factor. +
        +
      • Missing data will cause problems here.
        • +
      • +
    • We can us as.numeric() to revert the factor to its underlying numeric +representation {Male = 1, Female = 2} and use arithmetic to convert {1, 2} +\(\rightarrow\) {0, 1}.
      • +
    +

    When our factor only has two levels, though, the ifelse() function is the +simplest way.

    +
    library(dplyr)
+
+## Create a dummy codes by broadcasting a logical test on the factor levels:
+ess <- mutate(ess,
+              female = ifelse(sex == "Female", 1, 0),
+              hi_pol_interest = ifelse(polintr_bin == "High Interest", 1, 0)
+             )
+
+## Check the results:
+with(ess, table(dummy = female, factor = sex))
    +
    ##      factor
+## dummy Female Male
+##     0      0 8841
+##     1   9309    0
    +
    with(ess, table(dummy = hi_pol_interest, factor = polintr_bin))
    +
    ##      factor
+## dummy Low Interest High Interest
+##     0         8099             0
+##     1            0         10042
    +
    +
    +
    +

    6.3.3

    +

    Finally, subset the data to only Finnish participants.

    +
    + +Click for explanation + +
    ess_fin <- filter(ess, cntry == "Finland")
    +
    +
    +

    We are now ready to estimate our latent regression model. Specifically, we want +to implement the following regression as an SEM in lavaan.

    +

    \[ +Y_{trust\_inst} = \beta_1 X_{age} + \beta_2 X_{female} + \beta_3 X_{edu\_years} + + \beta_4 X_{hi\_pol\_interest} + \beta_5 X_{lrscale} + + \varepsilon_Y +\]

    +
    +
    +
    +

    6.3.4

    +

    Define the lavaan model syntax for the regression shown above.

    +
      +
    • Use the definition of the institutions factor from 5.3.2 to +define the DV.
      • +
    +

    Hint: You can simply copy the line of syntax that defines the latent factor +and add a line to define the latent regression model.

    +
    + +Click for explanation + +
    mod_sem <- '
+## Define the latent DV:
+institutions =~ trstlgl + trstplc + trstun
+
+## Specify the structural relations:
+institutions ~ female + age + eduyrs + hi_pol_interest + lrscale
+'
    +
    +
    +
    +
    +

    6.3.5

    +

    Estimate the SEM, and summarize the results.

    +
      +
    • Fit the model to the processed Finnish subsample from above.
      • +
    • Estimate the model using lavaan::sem().
      • +
    • Use the default settings in the sem() function.
      • +
    • Request the standardized parameter estimates with the summary.
      • +
    • Request the \(R^2\) estimates with the summary by supplying the rsquare = TRUE +argument to summary().
      • +
    +
    + +Click for explanation + +
    library(lavaan)
+
+## Fit the SEM:
+fit_sem <- sem(mod_sem, data = ess_fin)
+
+## Summarize the results:
+summary(fit_sem, fit.measures = TRUE, standardized = TRUE, rsquare = TRUE)
    +
    ## lavaan 0.6.16 ended normally after 43 iterations
+## 
+##   Estimator                                         ML
+##   Optimization method                           NLMINB
+##   Number of model parameters                        11
+## 
+##                                                   Used       Total
+##   Number of observations                          1835        2000
+## 
+## Model Test User Model:
+##                                                       
+##   Test statistic                               101.932
+##   Degrees of freedom                                10
+##   P-value (Chi-square)                           0.000
+## 
+## Model Test Baseline Model:
+## 
+##   Test statistic                              1221.668
+##   Degrees of freedom                                18
+##   P-value                                        0.000
+## 
+## User Model versus Baseline Model:
+## 
+##   Comparative Fit Index (CFI)                    0.924
+##   Tucker-Lewis Index (TLI)                       0.863
+## 
+## Loglikelihood and Information Criteria:
+## 
+##   Loglikelihood user model (H0)             -10940.642
+##   Loglikelihood unrestricted model (H1)     -10889.676
+##                                                       
+##   Akaike (AIC)                               21903.284
+##   Bayesian (BIC)                             21963.947
+##   Sample-size adjusted Bayesian (SABIC)      21929.001
+## 
+## Root Mean Square Error of Approximation:
+## 
+##   RMSEA                                          0.071
+##   90 Percent confidence interval - lower         0.059
+##   90 Percent confidence interval - upper         0.084
+##   P-value H_0: RMSEA <= 0.050                    0.003
+##   P-value H_0: RMSEA >= 0.080                    0.121
+## 
+## Standardized Root Mean Square Residual:
+## 
+##   SRMR                                           0.030
+## 
+## Parameter Estimates:
+## 
+##   Standard errors                             Standard
+##   Information                                 Expected
+##   Information saturated (h1) model          Structured
+## 
+## Latent Variables:
+##                    Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
+##   institutions =~                                                       
+##     trstlgl           1.000                               1.783    0.835
+##     trstplc           0.615    0.035   17.401    0.000    1.097    0.640
+##     trstun            0.603    0.037   16.355    0.000    1.076    0.528
+## 
+## Regressions:
+##                    Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
+##   institutions ~                                                        
+##     female           -0.025    0.094   -0.264    0.792   -0.014   -0.007
+##     age              -0.002    0.003   -0.534    0.594   -0.001   -0.015
+##     eduyrs            0.072    0.013    5.405    0.000    0.041    0.157
+##     hi_pol_interst    0.178    0.096    1.850    0.064    0.100    0.050
+##     lrscale           0.110    0.023    4.727    0.000    0.062    0.125
+## 
+## Variances:
+##                    Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
+##    .trstlgl           1.379    0.168    8.199    0.000    1.379    0.302
+##    .trstplc           1.739    0.084   20.599    0.000    1.739    0.591
+##    .trstun            3.000    0.116   25.792    0.000    3.000    0.722
+##    .institutions      3.019    0.209   14.451    0.000    0.949    0.949
+## 
+## R-Square:
+##                    Estimate
+##     trstlgl           0.698
+##     trstplc           0.409
+##     trstun            0.278
+##     institutions      0.051
    +
    +
    +
    +
    +

    6.3.6

    +

    Finally, we will rerun the latent regression model from above as a path model +with the factor score for Trust in Institutions as the DV.

    +
      +
    • Add the Trust in Institutions factor score that you estimated in +4.4.10 to the ESS data. +
        +
      • If you did not save the factors scores last week, you’ll need to rerun the +relevant EFA.
        • +
      • Don’t forget to subset the data to the Finish participants before fitting +the path model.
        • +
      • +
    • Rerun the above SEM with the EFA-derived Trust in Institutions factor score +taking the place of the analagous latent variable as DV. +
        +
      • Request the standardized parameter estimates with the summary.
        • +
      • Request the \(R^2\) estimates with the summary.
        • +
      • +
    +
    + +Click for explanation + +

    First, we’ll quickly reproduce the Trust in Institutions factor score that we +estimated last week.

    +
      +
    • Note that psych::fa() returns the factor scores in a different order than it +did for the Week 2 analyses.
      • +
    +
    ## Load the psych library:
+library(psych)
+
+## Rerun the three-factor EFA from last week:
+fit_efa <- fa(ess[7:19], 
+              nfactors = 3,          
+              rotate   = "promax",   
+              scores   = "Bartlett")
+
+## View the factor loadings:
+print(fit_efa$loadings, cut = 0.3)
    +
    ## 
+## Loadings:
+##         MR3    MR1    MR2   
+## pltcare         0.815       
+## pltinvt         0.806       
+## trstprl  0.405  0.381       
+## trstlgl  0.843              
+## trstplc  0.776              
+## trstplt         0.571       
+## trstep   0.503              
+## trstun   0.543              
+## stfeco                 0.703
+## stfgov                 0.594
+## stfdem                 0.462
+## stfedu                 0.701
+## stfhlth                0.664
+## 
+##                  MR3   MR1   MR2
+## SS loadings    2.178 2.101 2.023
+## Proportion Var 0.168 0.162 0.156
+## Cumulative Var 0.168 0.329 0.485
    +
    ## Reproduce the factor score from last week:
+ess$trust_inst_efa <- fit_efa$scores[ , 1]
+
+## Subset the data again:
+ess_fin <- ess[ess$cntry == "Finland", ]
    +

    Now, we’ll rerun our regression as a path analysis with the EFA-derived factor +score as DV.

    +
    ## Define the model syntax for the path analysis:
+mod_pa <- 'trust_inst_efa ~ female + age + eduyrs + hi_pol_interest + lrscale'
+
+## Estimate the path model:
+fit_pa <- sem(mod_pa, data = ess_fin)
+
+## Summarize the results:
+summary(fit_pa, fit.measures = TRUE, standardized = TRUE)
    +
    ## lavaan 0.6.16 ended normally after 1 iteration
+## 
+##   Estimator                                         ML
+##   Optimization method                           NLMINB
+##   Number of model parameters                         6
+## 
+##                                                   Used       Total
+##   Number of observations                          1740        2000
+## 
+## Model Test User Model:
+##                                                       
+##   Test statistic                                 0.000
+##   Degrees of freedom                                 0
+## 
+## Model Test Baseline Model:
+## 
+##   Test statistic                                71.266
+##   Degrees of freedom                                 5
+##   P-value                                        0.000
+## 
+## User Model versus Baseline Model:
+## 
+##   Comparative Fit Index (CFI)                    1.000
+##   Tucker-Lewis Index (TLI)                       1.000
+## 
+## Loglikelihood and Information Criteria:
+## 
+##   Loglikelihood user model (H0)              -2352.608
+##   Loglikelihood unrestricted model (H1)      -2352.608
+##                                                       
+##   Akaike (AIC)                                4717.216
+##   Bayesian (BIC)                              4749.985
+##   Sample-size adjusted Bayesian (SABIC)       4730.924
+## 
+## Root Mean Square Error of Approximation:
+## 
+##   RMSEA                                          0.000
+##   90 Percent confidence interval - lower         0.000
+##   90 Percent confidence interval - upper         0.000
+##   P-value H_0: RMSEA <= 0.050                       NA
+##   P-value H_0: RMSEA >= 0.080                       NA
+## 
+## Standardized Root Mean Square Residual:
+## 
+##   SRMR                                           0.000
+## 
+## Parameter Estimates:
+## 
+##   Standard errors                             Standard
+##   Information                                 Expected
+##   Information saturated (h1) model          Structured
+## 
+## Regressions:
+##                    Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
+##   trust_inst_efa ~                                                      
+##     female            0.002    0.045    0.036    0.971    0.002    0.001
+##     age              -0.003    0.001   -1.977    0.048   -0.003   -0.051
+##     eduyrs            0.024    0.006    3.820    0.000    0.024    0.099
+##     hi_pol_interst    0.154    0.047    3.313    0.001    0.154    0.081
+##     lrscale           0.058    0.011    5.159    0.000    0.058    0.122
+## 
+## Variances:
+##                    Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
+##    .trust_inst_efa    0.875    0.030   29.496    0.000    0.875    0.960
    +
    ## Extract R^2:
+inspect(fit_pa, "r2")
    +
    ## trust_inst_efa 
+##           0.04
    +
    +
    +
    +
    +

    6.3.7

    +

    Compare the results from the path analysis to the SEM-based results.

    +
      +
    • Does it matter whether we use a latent variable or a factor score to define +the DV?
      • +
    +

    Hint: When comparing parameter estimates, use the fully standardized estimates +(i.e., the values in the column labeled Std.all).

    +
    + +Click for explanation + +

    First, we’ll source a script that defines a bunch of convenience functions. One +of these functions, partSummary(), will allow us to print only the interesting +pieces of the model summary.

    +
    ## Source a script of convenience function definitions:
+source("supportFunctions.R")
    +

    Now, we’ll compare the results. Specifically, we’re interested in differences in +the regression coefficients and the \(R^2\).

    +
    ## View the regression estimates from the SEM:
+partSummary(fit_sem, 8, standardized = TRUE)
    +
    ## Regressions:
+##                    Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
+##   institutions ~                                                        
+##     female           -0.025    0.094   -0.264    0.792   -0.014   -0.007
+##     age              -0.002    0.003   -0.534    0.594   -0.001   -0.015
+##     eduyrs            0.072    0.013    5.405    0.000    0.041    0.157
+##     hi_pol_interst    0.178    0.096    1.850    0.064    0.100    0.050
+##     lrscale           0.110    0.023    4.727    0.000    0.062    0.125
    +
    ## View the regression estimates from the path analysis:
+partSummary(fit_pa, 7, standardized = TRUE)
    +
    ## Regressions:
+##                    Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
+##   trust_inst_efa ~                                                      
+##     female            0.002    0.045    0.036    0.971    0.002    0.001
+##     age              -0.003    0.001   -1.977    0.048   -0.003   -0.051
+##     eduyrs            0.024    0.006    3.820    0.000    0.024    0.099
+##     hi_pol_interst    0.154    0.047    3.313    0.001    0.154    0.081
+##     lrscale           0.058    0.011    5.159    0.000    0.058    0.122
    +
    ## View the R-squared estimates from the SEM:
+partSummary(fit_sem, 10, rsquare = TRUE)
    +
    ## R-Square:
+##                    Estimate
+##     trstlgl           0.698
+##     trstplc           0.409
+##     trstun            0.278
+##     institutions      0.051
    +
    ## View the R-squared estimate from the path analysis:
+#partSummary(fit_pa, 9, rsquare = TRUE)
+inspect(fit_pa, "r2")
    +
    ## trust_inst_efa 
+##           0.04
    +

    It certainly looks like the way we define the DV has a meaningful impact. The +patterns of significance differ between the two sets of regression slopes, and +the \(R^2\) is 26.7% larger in the SEM.

    +
    +
    +

    End of In-Class Exercises

    +
    +
    diff --git a/docs/at-home-exercises-6.html b/docs/at-home-exercises-6.html index 7eed9420..f4e30a03 100644 --- a/docs/at-home-exercises-6.html +++ b/docs/at-home-exercises-6.html @@ -409,16 +409,25 @@
  • Reference
  • Questions

    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/at-home-exercises.html b/docs/at-home-exercises.html index 8e44fc76..fa128d06 100644 --- a/docs/at-home-exercises.html +++ b/docs/at-home-exercises.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/attendance.html b/docs/attendance.html index 8de19699..f91f356c 100644 --- a/docs/attendance.html +++ b/docs/attendance.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/cfa.html b/docs/cfa.html index baf45925..e350ca20 100644 --- a/docs/cfa.html +++ b/docs/cfa.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/course-data.html b/docs/course-data.html index c5e6d4b1..12d8e9d3 100644 --- a/docs/course-data.html +++ b/docs/course-data.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/course-overview.html b/docs/course-overview.html index a77f895d..c57359bd 100644 --- a/docs/course-overview.html +++ b/docs/course-overview.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/data_options.html b/docs/data_options.html index bcac48a4..dbc359de 100644 --- a/docs/data_options.html +++ b/docs/data_options.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/efa.html b/docs/efa.html index de24a4e5..d86721ea 100644 --- a/docs/efa.html +++ b/docs/efa.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/elaboration-tips.html b/docs/elaboration-tips.html index 9a5b621c..c55fbf6c 100644 --- a/docs/elaboration-tips.html +++ b/docs/elaboration-tips.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/full-sem.html b/docs/full-sem.html index b92bcad0..fd7fb573 100644 --- a/docs/full-sem.html +++ b/docs/full-sem.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/grading-1.html b/docs/grading-1.html index 4ea11dba..a1099f31 100644 --- a/docs/grading-1.html +++ b/docs/grading-1.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/grading.html b/docs/grading.html index d8361d02..474fe11b 100644 --- a/docs/grading.html +++ b/docs/grading.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/in-class-exercises-1.html b/docs/in-class-exercises-1.html index d7ccbfa7..0a21871b 100644 --- a/docs/in-class-exercises-1.html +++ b/docs/in-class-exercises-1.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/in-class-exercises-2.html b/docs/in-class-exercises-2.html index b8aca572..03d7e89f 100644 --- a/docs/in-class-exercises-2.html +++ b/docs/in-class-exercises-2.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/in-class-exercises-3.html b/docs/in-class-exercises-3.html index c5ce3fbc..64f47b12 100644 --- a/docs/in-class-exercises-3.html +++ b/docs/in-class-exercises-3.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/in-class-exercises-4.html b/docs/in-class-exercises-4.html index 8a596a5a..282827f6 100644 --- a/docs/in-class-exercises-4.html +++ b/docs/in-class-exercises-4.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/in-class-exercises-5.html b/docs/in-class-exercises-5.html index c1ea949e..b983cab9 100644 --- a/docs/in-class-exercises-5.html +++ b/docs/in-class-exercises-5.html @@ -409,16 +409,25 @@
  • Reference
  • Questions
    • -
  • 6.3 At-Home Exercises
    • +
  • 6.3 At-Home Exercises +
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models @@ -491,18 +500,18 @@

    6.4 In-Class Exercises +

    6.4.1

    Load the data contained in the toradata.csv file.
    Click for explanation -
    condom <- read.csv("toradata.csv", stringsAsFactors = TRUE)
    +
    condom <- read.csv("toradata.csv", stringsAsFactors = TRUE)
    -
    +

    6.4.2

    The data contain multiple indicators of attitudes, norms, and control. Run a CFA for these three latent variables.

    @@ -516,17 +525,17 @@

    6.4.2

    Click for explanation -
    library(lavaan)
-
-mod_cfa <- '
-attitudes =~ attit_1   + attit_2   + attit_3
-norms     =~ norm_1    + norm_2    + norm_3
-control   =~ control_1 + control_2 + control_3
-'
-
-fit <- cfa(mod_cfa, data = condom)
-
-summary(fit, fit.measures = TRUE)
    +
    library(lavaan)
+
+mod_cfa <- '
+attitudes =~ attit_1   + attit_2   + attit_3
+norms     =~ norm_1    + norm_2    + norm_3
+control   =~ control_1 + control_2 + control_3
+'
+
+fit <- cfa(mod_cfa, data = condom)
+
+summary(fit, fit.measures = TRUE)
    ## lavaan 0.6.16 ended normally after 29 iterations
 ## 
 ##   Estimator                                         ML
@@ -633,7 +642,7 @@ 
    6.4.2
    -
    +

    6.4.3

    Estimate the basic TORA model as an SEM.

      @@ -649,19 +658,19 @@

      6.4.3

      Click for explanation -
      mod <- '
-## Define the latent variables:
-attitudes =~ attit_1 + attit_2 + attit_3
-norms     =~ norm_1  + norm_2  + norm_3
-
-## Define the structural model:         
-intent   ~ attitudes + norms
-behavior ~ intent
-'
-
-fit <- sem(mod, data = condom)
-
-summary(fit, fit.measures = TRUE, rsquare = TRUE)
      +
      mod <- '
+## Define the latent variables:
+attitudes =~ attit_1 + attit_2 + attit_3
+norms     =~ norm_1  + norm_2  + norm_3
+
+## Define the structural model:         
+intent   ~ attitudes + norms
+behavior ~ intent
+'
+
+fit <- sem(mod, data = condom)
+
+summary(fit, fit.measures = TRUE, rsquare = TRUE)
      ## lavaan 0.6.16 ended normally after 24 iterations
 ## 
 ##   Estimator                                         ML
@@ -789,18 +798,18 @@ 
      6.4.4 
      
 
 Click for explanation
 
-
      mod_tora <- '
-attitudes =~ attit_1   + attit_2   + attit_3
-norms     =~ norm_1    + norm_2    + norm_3
-control   =~ control_1 + control_2 + control_3
-             
-intent   ~ attitudes + norms
-behavior ~ intent + control
-'
-
-fit_tora <- sem(mod_tora, data = condom)
-
-summary(fit_tora, fit.measures = TRUE, rsquare = TRUE)
      
+
      mod_tora <- '
+attitudes =~ attit_1   + attit_2   + attit_3
+norms     =~ norm_1    + norm_2    + norm_3
+control   =~ control_1 + control_2 + control_3
+             
+intent   ~ attitudes + norms
+behavior ~ intent + control
+'
+
+fit_tora <- sem(mod_tora, data = condom)
+
+summary(fit_tora, fit.measures = TRUE, rsquare = TRUE)
      
 
      ## lavaan 0.6.16 ended normally after 31 iterations
 ## 
 ##   Estimator                                         ML
@@ -933,7 +942,7 @@ 
      6.4.4 
      
 effect may be (partially) mediated by intention.
    -
    +

    6.4.5

    Evaluate the hypothesized indirect effects of attitudes and norms.

      @@ -950,23 +959,23 @@

      6.4.5

      Click for explanation -
      mod <- '
-attitudes =~ attit_1   + attit_2   + attit_3
-norms     =~ norm_1    + norm_2    + norm_3
-control   =~ control_1 + control_2 + control_3
-             
-intent   ~ a1 * attitudes + a2 * norms
-behavior ~ b * intent + control + attitudes + norms
-
-ie_att  := a1 * b
-ie_norm := a2 * b
-'
-
-set.seed(235711)
-
-fit <- sem(mod, data = condom, se = "bootstrap", bootstrap = 1000)
-
-summary(fit, ci = TRUE)
      +
      mod <- '
+attitudes =~ attit_1   + attit_2   + attit_3
+norms     =~ norm_1    + norm_2    + norm_3
+control   =~ control_1 + control_2 + control_3
+             
+intent   ~ a1 * attitudes + a2 * norms
+behavior ~ b * intent + control + attitudes + norms
+
+ie_att  := a1 * b
+ie_norm := a2 * b
+'
+
+set.seed(235711)
+
+fit <- sem(mod, data = condom, se = "bootstrap", bootstrap = 1000)
+
+summary(fit, ci = TRUE)
      ## Length  Class   Mode 
 ##      1 lavaan     S4
        @@ -986,7 +995,7 @@

        6.4.5

        accurately represent the data, or are they “dead weight”.

    -
    +

    6.4.6

    Use a \(\Delta \chi^2\) test to evaluate the necessity of including the direct effects of attitudes and norms on condom use in the model.

    @@ -1001,7 +1010,7 @@

    6.4.6

    to the fit of the model without the direct effects. We’ve already estimated both models, so we can simply submit the fitted lavaan objects to the anova() function.

    -
    anova(fit, fit_tora)
    +
    anova(fit, fit_tora)
    -
    anova(fit1, fit3)
    +
    anova(fit1, fit3)
    -
    BIC(fit1, fit2, fit3)
    +
    BIC(fit1, fit2, fit3)
    -
    ##      name              essround   edition            proddate        
-##  Length:42359       Min.   :1   Length:42359       Length:42359      
-##  Class :character   1st Qu.:1   Class :character   Class :character  
-##  Mode  :character   Median :1   Mode  :character   Mode  :character  
-##                     Mean   :1                                        
-##                     3rd Qu.:1                                        
-##                     Max.   :1                                        
-##                                                                      
-##      cntry            idno              trstlgl          trstplc      
-##  Min.   : 1.00   Min.   :        1   Min.   : 0.000   Min.   : 0.000  
-##  1st Qu.: 6.00   1st Qu.:     1647   1st Qu.: 4.000   1st Qu.: 5.000  
-##  Median :11.00   Median :     4636   Median : 6.000   Median : 7.000  
-##  Mean   :11.31   Mean   :  8096164   Mean   : 5.479   Mean   : 6.201  
-##  3rd Qu.:17.00   3rd Qu.:   201732   3rd Qu.: 8.000   3rd Qu.: 8.000  
-##  Max.   :22.00   Max.   :109801018   Max.   :10.000   Max.   :10.000  
-##                                      NA's   :1207     NA's   :523     
-##      trstun           trstep         trstprl          stfhlth      
-##  Min.   : 0.000   Min.   : 0.00   Min.   : 0.000   Min.   : 0.000  
-##  1st Qu.: 4.000   1st Qu.: 3.00   1st Qu.: 3.000   1st Qu.: 3.000  
-##  Median : 5.000   Median : 5.00   Median : 5.000   Median : 5.000  
-##  Mean   : 5.363   Mean   : 4.71   Mean   : 4.896   Mean   : 5.288  
-##  3rd Qu.: 7.000   3rd Qu.: 6.00   3rd Qu.: 7.000   3rd Qu.: 7.000  
-##  Max.   :10.000   Max.   :10.00   Max.   :10.000   Max.   :10.000  
-##  NA's   :4415     NA's   :6015    NA's   :1582     NA's   :570     
-##      stfedu           stfeco           stfgov           stfdem      
-##  Min.   : 0.000   Min.   : 0.000   Min.   : 0.000   Min.   : 0.000  
-##  1st Qu.: 4.000   1st Qu.: 3.000   1st Qu.: 3.000   1st Qu.: 4.000  
-##  Median : 6.000   Median : 5.000   Median : 5.000   Median : 6.000  
-##  Mean   : 5.531   Mean   : 4.492   Mean   : 4.378   Mean   : 5.468  
-##  3rd Qu.: 7.000   3rd Qu.: 6.000   3rd Qu.: 6.000   3rd Qu.: 7.000  
-##  Max.   :10.000   Max.   :10.000   Max.   :10.000   Max.   :10.000  
-##  NA's   :2396     NA's   :1254     NA's   :3585     NA's   :1857    
-##     pltinvt         pltcare         trstplt          imsmetn     
-##  Min.   :1.000   Min.   :1.000   Min.   : 0.000   Min.   :1.000  
-##  1st Qu.:1.000   1st Qu.:2.000   1st Qu.: 2.000   1st Qu.:2.000  
-##  Median :2.000   Median :2.000   Median : 4.000   Median :2.000  
-##  Mean   :2.224   Mean   :2.362   Mean   : 3.899   Mean   :2.244  
-##  3rd Qu.:3.000   3rd Qu.:3.000   3rd Qu.: 5.000   3rd Qu.:3.000  
-##  Max.   :5.000   Max.   :5.000   Max.   :10.000   Max.   :4.000  
-##  NA's   :2255    NA's   :879     NA's   :914      NA's   :2042   
-##     imdfetn        eimrcnt         eimpcnt         imrcntr         impcntr     
-##  Min.   :1.00   Min.   :1.000   Min.   :1.000   Min.   :1.000   Min.   :1.000  
-##  1st Qu.:2.00   1st Qu.:2.000   1st Qu.:2.000   1st Qu.:2.000   1st Qu.:2.000  
-##  Median :3.00   Median :2.000   Median :2.000   Median :2.000   Median :2.000  
-##  Mean   :2.53   Mean   :2.362   Mean   :2.434   Mean   :2.451   Mean   :2.505  
-##  3rd Qu.:3.00   3rd Qu.:3.000   3rd Qu.:3.000   3rd Qu.:3.000   3rd Qu.:3.000  
-##  Max.   :4.00   Max.   :4.000   Max.   :4.000   Max.   :4.000   Max.   :4.000  
-##  NA's   :2096   NA's   :2262    NA's   :2005    NA's   :2332    NA's   :2180   
-##     qfimchr          qfimwht          imwgdwn         imhecop     
-##  Min.   : 0.000   Min.   : 0.000   Min.   :1.000   Min.   :1.000  
-##  1st Qu.: 0.000   1st Qu.: 0.000   1st Qu.:2.000   1st Qu.:2.000  
-##  Median : 3.000   Median : 1.000   Median :3.000   Median :2.000  
-##  Mean   : 5.788   Mean   : 4.212   Mean   :2.934   Mean   :2.677  
-##  3rd Qu.: 7.000   3rd Qu.: 5.000   3rd Qu.:4.000   3rd Qu.:4.000  
-##  Max.   :99.000   Max.   :99.000   Max.   :5.000   Max.   :5.000  
-##                                    NA's   :2531    NA's   :2382   
-##     imtcjob          imbleco          imbgeco          imueclt      
-##  Min.   : 0.000   Min.   : 0.000   Min.   : 0.000   Min.   : 0.000  
-##  1st Qu.: 3.000   1st Qu.: 3.000   1st Qu.: 3.000   1st Qu.: 4.000  
-##  Median : 5.000   Median : 5.000   Median : 5.000   Median : 6.000  
-##  Mean   : 4.468   Mean   : 4.184   Mean   : 4.959   Mean   : 5.765  
-##  3rd Qu.: 6.000   3rd Qu.: 5.000   3rd Qu.: 7.000   3rd Qu.: 8.000  
-##  Max.   :10.000   Max.   :10.000   Max.   :10.000   Max.   :10.000  
-##  NA's   :1904     NA's   :2808     NA's   :2590     NA's   :2390    
-##     imwbcnt          imwbcrm          imrsprc         pplstrd     
-##  Min.   : 0.000   Min.   : 0.000   Min.   :1.000   Min.   :1.000  
-##  1st Qu.: 3.000   1st Qu.: 2.000   1st Qu.:2.000   1st Qu.:2.000  
-##  Median : 5.000   Median : 3.000   Median :2.000   Median :2.000  
-##  Mean   : 4.772   Mean   : 3.093   Mean   :2.591   Mean   :2.626  
-##  3rd Qu.: 6.000   3rd Qu.: 5.000   3rd Qu.:3.000   3rd Qu.:4.000  
-##  Max.   :10.000   Max.   :10.000   Max.   :5.000   Max.   :5.000  
-##  NA's   :2195     NA's   :1643     NA's   :1411    NA's   :721    
-##      vrtrlg          shrrfg         rfgawrk         gvrfgap         rfgfrpc    
-##  Min.   :1.000   Min.   :1.000   Min.   :1.000   Min.   :1.000   Min.   :1.00  
-##  1st Qu.:2.000   1st Qu.:2.000   1st Qu.:2.000   1st Qu.:2.000   1st Qu.:2.00  
-##  Median :3.000   Median :2.000   Median :2.000   Median :3.000   Median :3.00  
-##  Mean   :2.975   Mean   :2.387   Mean   :2.491   Mean   :3.141   Mean   :2.74  
-##  3rd Qu.:4.000   3rd Qu.:3.000   3rd Qu.:3.000   3rd Qu.:4.000   3rd Qu.:3.00  
-##  Max.   :5.000   Max.   :5.000   Max.   :5.000   Max.   :5.000   Max.   :5.00  
-##  NA's   :1444    NA's   :4756    NA's   :2192    NA's   :2185    NA's   :5927  
-##     rfggvfn         rfgbfml           gndr           yrbrn          edulvl     
-##  Min.   :1.000   Min.   :1.000   Min.   :1.000   Min.   :1893   Min.   :0.000  
-##  1st Qu.:2.000   1st Qu.:2.000   1st Qu.:1.000   1st Qu.:1942   1st Qu.:2.000  
-##  Median :3.000   Median :3.000   Median :2.000   Median :1957   Median :3.000  
-##  Mean   :2.899   Mean   :2.851   Mean   :1.526   Mean   :1956   Mean   :2.837  
-##  3rd Qu.:4.000   3rd Qu.:4.000   3rd Qu.:2.000   3rd Qu.:1971   3rd Qu.:3.000  
-##  Max.   :5.000   Max.   :5.000   Max.   :2.000   Max.   :1989   Max.   :6.000  
-##  NA's   :2059    NA's   :1968    NA's   :55      NA's   :258    NA's   :2559   
-##      eduyrs         polintr         lrscale      
-##  Min.   : 0.00   Min.   :1.000   Min.   : 0.000  
-##  1st Qu.: 9.00   1st Qu.:2.000   1st Qu.: 4.000  
-##  Median :12.00   Median :3.000   Median : 5.000  
-##  Mean   :11.85   Mean   :2.574   Mean   : 5.068  
-##  3rd Qu.:14.00   3rd Qu.:3.000   3rd Qu.: 6.000  
-##  Max.   :40.00   Max.   :4.000   Max.   :10.000  
-##  NA's   :642     NA's   :148     NA's   :5215
    +
    + +
    +
    Click here for a description of the variables. @@ -1013,6 +930,39 @@

    4.3.1

    +

    One thing you might notice when inspecting the ess data is that most of the +variables are stored as labelled vectors. When loading SPSS data, haven +will use these labelled vectors to preserve the metadata associated with SPSS +scale variables (i.e., variable labels and value labels).

    +

    While it’s good to have this metadata available, we want to analyze these items +as numeric variables and factors, so the value labels are only going to make our +lives harder. Thankfully, the labelled package contains many routines for +manipulating labelled vectors.

    +

    We’ll deal with the numeric variables in just a bit, but our first task will be +to covert grouping variables to factors.

    +
    +
    +
    +

    4.3.2

    +

    Convert the cntry, gndr, edulvl, and polintr variables into factors.

    +
      +
    • Use the as_factor() function to do the conversion.
      • +
    • Convert edulvl and polintr to ordered factors.
      • +
    +
    + +Click to see code + +
    library(dplyr)
+
+ess <- mutate(ess,
+              country = as_factor(cntry),
+              sex = as_factor(gndr),
+              edulvl = as_factor(edulvl, ordered = TRUE),
+              polintr = as_factor(polintr, ordered = TRUE)
+              )
    +
    +

    The ess dataset contains much more information than Kestilä (2006) used. Kestilä only analyzed data from the following ten countries:

      @@ -1027,14 +977,14 @@

      4.3.1

    • Norway
    • Sweden
    -

    So, our first task is to subset the data to only the relevant population.

    +

    So, our next task is to subset the data to only the relevant population.

    When we apply logical subsetting, we can select rows from a dataset based on logical conditions. In this case, we want to select only rows from the 10 countries listed above.

    -
    -

    4.3.2

    +
    +

    4.3.3

    Subset the data to include only the 10 countries analyzed by Kestilä (2006).

    • Inspect the subsetted data to check that everything went well.
      • @@ -1043,37 +993,32 @@

      4.3.2

      • Use the %in% operator to create a logical vector that indicates which elements of the cntry variable are in the set of target counties.
        • -
      • You may find it helpful to convert cntry to a factor using as_factor().
        • -
      • You can clean up empty factor levels using droplevels().
        • +
      • Use the droplevels() levels function to clean up empty factor levels.
      Click to show code -
      library(dplyr)
-
-## Create a character vector naming the target countries:
-targets <- c("Austria", 
-             "Belgium", 
-             "Denmark", 
-             "Finland", 
-             "France", 
-             "Germany", 
-             "Italy", 
-             "Netherlands", 
-             "Norway", 
-             "Sweden")
-
-## Select only those rows that come from a target country:
-ess <- ess %>%
-  mutate(cntry = as_factor(cntry)) %>% # Convert cntry to a factor
-  filter(cntry %in% targets) %>%       # Subset rows
-  mutate(cntry = droplevels(cntry))    # Drop empty factor levels
-
-## Inspect the result:
-dim(ess)
      -
      ## [1] 19690    50
      -
      table(ess$cntry)
      +
      ## Create a character vector naming the target countries:
+targets <- c("Austria", 
+             "Belgium", 
+             "Denmark", 
+             "Finland", 
+             "France", 
+             "Germany", 
+             "Italy", 
+             "Netherlands", 
+             "Norway", 
+             "Sweden")
+
+## Select only those rows that come from a target country:
+ess <- filter(ess, country %in% targets) %>% # Subset rows
+  droplevels()                               # Drop empty factor levels
+
+## Inspect the result:
+dim(ess)
      +
      ## [1] 19690    52
      +
      table(ess$country)
      ## 
 ##     Austria     Belgium     Germany     Denmark     Finland      France 
 ##        2257        1899        2919        1506        2000        1503 
@@ -1081,350 +1026,33 @@ 
      4.3.2 
      
 ##        1207        2364        2036        1999
      -

      Before we can analyze the data, we need to screen them for problems. At this -point, we won’t get into the nitty-gritty of outlier analysis and data -cleaning. We must, however, ensure that the variables we want to analyze are -formatted correctly (e.g., interval and ratio variables should be numeric, -nominal variables should be factors).

      -
      -
    -
    -

    4.3.3

    -

    Screen the data to see if the variables are formatted correctly.

    -
    - -Click to show code - -
    library(tidySEM)
-
-## Check the data structure:
-str(ess)
    -
    ## tibble [19,690 × 50] (S3: tbl_df/tbl/data.frame)
-##  $ name    : chr [1:19690] "ESS1e06_1" "ESS1e06_1" "ESS1e06_1" "ESS1e06_1" ...
-##   ..- attr(*, "label")= chr "Title of dataset"
-##   ..- attr(*, "format.spss")= chr "A9"
-##   ..- attr(*, "display_width")= int 14
-##  $ essround: num [1:19690] 1 1 1 1 1 1 1 1 1 1 ...
-##   ..- attr(*, "label")= chr "ESS round"
-##   ..- attr(*, "format.spss")= chr "F2.0"
-##   ..- attr(*, "display_width")= int 10
-##  $ edition : chr [1:19690] "6.1" "6.1" "6.1" "6.1" ...
-##   ..- attr(*, "label")= chr "Edition"
-##   ..- attr(*, "format.spss")= chr "A3"
-##   ..- attr(*, "display_width")= int 9
-##  $ proddate: chr [1:19690] "03.10.2008" "03.10.2008" "03.10.2008" "03.10.2008" ...
-##   ..- attr(*, "label")= chr "Production date"
-##   ..- attr(*, "format.spss")= chr "A10"
-##   ..- attr(*, "display_width")= int 12
-##  $ cntry   : Factor w/ 10 levels "Austria","Belgium",..: 1 9 1 1 9 1 2 9 1 9 ...
-##  $ idno    : num [1:19690] 1 1 2 3 3 4 4 4 6 6 ...
-##   ..- attr(*, "label")= chr "Respondent's identification number"
-##   ..- attr(*, "format.spss")= chr "F9.0"
-##   ..- attr(*, "display_width")= int 11
-##  $ trstlgl : dbl+lbl [1:19690] 10,  6,  8,  4,  8, 10,  9,  7,  7,  7,  5,  6,  5, ...
-##    ..@ label        : chr "Trust in the legal system"
-##    ..@ format.spss  : chr "F2.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:14] 0 1 2 3 4 5 6 7 8 9 ...
-##    .. ..- attr(*, "names")= chr [1:14] "No trust at all" "1" "2" "3" ...
-##  $ trstplc : dbl+lbl [1:19690] 10,  8,  5,  8,  8,  9,  8,  9,  4,  9,  6,  6,  8, ...
-##    ..@ label        : chr "Trust in the police"
-##    ..@ format.spss  : chr "F2.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:14] 0 1 2 3 4 5 6 7 8 9 ...
-##    .. ..- attr(*, "names")= chr [1:14] "No trust at all" "1" "2" "3" ...
-##  $ trstun  : dbl+lbl [1:19690]  9,  8,  6, NA,  5,  8, NA,  7,  5,  7, NA, NA,  8, ...
-##    ..@ label      : chr "Trust in the United Nations"
-##    ..@ format.spss: chr "F2.0"
-##    ..@ labels     : Named num [1:14] 0 1 2 3 4 5 6 7 8 9 ...
-##    .. ..- attr(*, "names")= chr [1:14] "No trust at all" "1" "2" "3" ...
-##  $ trstep  : dbl+lbl [1:19690] NA,  3,  0,  7,  3,  7,  0,  3,  4,  6,  2,  5,  3, ...
-##    ..@ label      : chr "Trust in the European Parliament"
-##    ..@ format.spss: chr "F2.0"
-##    ..@ labels     : Named num [1:14] 0 1 2 3 4 5 6 7 8 9 ...
-##    .. ..- attr(*, "names")= chr [1:14] "No trust at all" "1" "2" "3" ...
-##  $ trstprl : dbl+lbl [1:19690]  9,  7,  0,  6,  8,  8, 10,  2,  6,  8,  0,  6,  0, ...
-##    ..@ label        : chr "Trust in country's parliament"
-##    ..@ format.spss  : chr "F2.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:14] 0 1 2 3 4 5 6 7 8 9 ...
-##    .. ..- attr(*, "names")= chr [1:14] "No trust at all" "1" "2" "3" ...
-##  $ stfhlth : dbl+lbl [1:19690] 10,  4,  0,  7,  6,  8, NA,  6,  3,  5,  6,  9,  2, ...
-##    ..@ label        : chr "State of health services in country nowadays"
-##    ..@ format.spss  : chr "F2.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:14] 0 1 2 3 4 5 6 7 8 9 ...
-##    .. ..- attr(*, "names")= chr [1:14] "Extremely bad" "1" "2" "3" ...
-##  $ stfedu  : dbl+lbl [1:19690]  8,  7,  7,  5,  8,  7, NA,  7,  6,  7,  3,  8,  7, ...
-##    ..@ label      : chr "State of education in country nowadays"
-##    ..@ format.spss: chr "F2.0"
-##    ..@ labels     : Named num [1:14] 0 1 2 3 4 5 6 7 8 9 ...
-##    .. ..- attr(*, "names")= chr [1:14] "Extremely bad" "1" "2" "3" ...
-##  $ stfeco  : dbl+lbl [1:19690]  7,  6,  0,  7,  8,  6, NA,  9,  8,  9,  0,  8,  2, ...
-##    ..@ label      : chr "How satisfied with present state of economy in country"
-##    ..@ format.spss: chr "F2.0"
-##    ..@ labels     : Named num [1:14] 0 1 2 3 4 5 6 7 8 9 ...
-##    .. ..- attr(*, "names")= chr [1:14] "Extremely dissatisfied" "1" "2" "3" ...
-##  $ stfgov  : dbl+lbl [1:19690]  7,  7,  0,  7,  6,  3, NA,  5,  5,  7,  0,  5,  0, ...
-##    ..@ label      : chr "How satisfied with the national government"
-##    ..@ format.spss: chr "F2.0"
-##    ..@ labels     : Named num [1:14] 0 1 2 3 4 5 6 7 8 9 ...
-##    .. ..- attr(*, "names")= chr [1:14] "Extremely dissatisfied" "1" "2" "3" ...
-##  $ stfdem  : dbl+lbl [1:19690]  8,  5,  5,  5,  7,  7, NA,  7,  7,  9,  0,  8,  7, ...
-##    ..@ label      : chr "How satisfied with the way democracy works in country"
-##    ..@ format.spss: chr "F2.0"
-##    ..@ labels     : Named num [1:14] 0 1 2 3 4 5 6 7 8 9 ...
-##    .. ..- attr(*, "names")= chr [1:14] "Extremely dissatisfied" "1" "2" "3" ...
-##  $ pltinvt : dbl+lbl [1:19690] 1, 3, 1, 1, 4, 1, 1, 3, 2, 3, 1, 2, 1, 3, 4, 2, 3, 4...
-##    ..@ label        : chr "Politicians interested in votes rather than peoples opinions"
-##    ..@ format.spss  : chr "F1.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:8] 1 2 3 4 5 7 8 9
-##    .. ..- attr(*, "names")= chr [1:8] "Nearly all just interested in votes" "Most just interested in votes" "Some just interested in votes" "Most interested in opinions" ...
-##  $ pltcare : dbl+lbl [1:19690] 1, 4, 1, 1, 4, 3, 2, 5, 2, 3, 1, 2, 2, 3, 5, 3, 2, 3...
-##    ..@ label        : chr "Politicians in general care what people like respondent think"
-##    ..@ format.spss  : chr "F1.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:8] 1 2 3 4 5 7 8 9
-##    .. ..- attr(*, "names")= chr [1:8] "Hardly any politicians care" "Very few care" "Some care" "Many care" ...
-##  $ trstplt : dbl+lbl [1:19690]  0,  5,  0,  2,  5,  4,  8,  2,  4,  6,  0,  5,  0, ...
-##    ..@ label        : chr "Trust in politicians"
-##    ..@ format.spss  : chr "F2.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:14] 0 1 2 3 4 5 6 7 8 9 ...
-##    .. ..- attr(*, "names")= chr [1:14] "No trust at all" "1" "2" "3" ...
-##  $ imsmetn : dbl+lbl [1:19690]  4,  3,  2,  3,  2,  1, NA,  2, NA,  1,  1, NA,  3, ...
-##    ..@ label        : chr "Allow many/few immigrants of same race/ethnic group as majority"
-##    ..@ format.spss  : chr "F1.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:7] 1 2 3 4 7 8 9
-##    .. ..- attr(*, "names")= chr [1:7] "Allow many to come and live here" "Allow some" "Allow a few" "Allow none" ...
-##  $ imdfetn : dbl+lbl [1:19690]  3,  3,  2,  3,  2,  2, NA,  2, NA,  1,  1, NA,  3, ...
-##    ..@ label        : chr "Allow many/few immigrants of different race/ethnic group from majority"
-##    ..@ format.spss  : chr "F1.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:7] 1 2 3 4 7 8 9
-##    .. ..- attr(*, "names")= chr [1:7] "Allow many to come and live here" "Allow some" "Allow a few" "Allow none" ...
-##  $ eimrcnt : dbl+lbl [1:19690]  4,  2,  2,  2,  3,  1, NA,  2, NA,  1,  1,  3,  3, ...
-##    ..@ label        : chr "Allow many/few immigrants from richer countries in Europe"
-##    ..@ format.spss  : chr "F1.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:7] 1 2 3 4 7 8 9
-##    .. ..- attr(*, "names")= chr [1:7] "Allow many to come and live here" "Allow some" "Allow a few" "Allow none" ...
-##  $ eimpcnt : dbl+lbl [1:19690]  3,  2,  2,  2,  2,  2, NA,  2, NA,  1,  1,  3,  3, ...
-##    ..@ label        : chr "Allow many/few immigrants from poorer countries in Europe"
-##    ..@ format.spss  : chr "F1.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:7] 1 2 3 4 7 8 9
-##    .. ..- attr(*, "names")= chr [1:7] "Allow many to come and live here" "Allow some" "Allow a few" "Allow none" ...
-##  $ imrcntr : dbl+lbl [1:19690]  3,  3,  2,  2,  2,  1, NA,  2, NA,  2,  1, NA,  3, ...
-##    ..@ label        : chr "Allow many/few immigrants from richer countries outside Europe"
-##    ..@ format.spss  : chr "F1.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:7] 1 2 3 4 7 8 9
-##    .. ..- attr(*, "names")= chr [1:7] "Allow many to come and live here" "Allow some" "Allow a few" "Allow none" ...
-##  $ impcntr : dbl+lbl [1:19690]  3,  2,  2,  3,  2,  1, NA,  2, NA,  2,  1, NA,  3, ...
-##    ..@ label        : chr "Allow many/few immigrants from poorer countries outside Europe"
-##    ..@ format.spss  : chr "F1.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:7] 1 2 3 4 7 8 9
-##    .. ..- attr(*, "names")= chr [1:7] "Allow many to come and live here" "Allow some" "Allow a few" "Allow none" ...
-##  $ qfimchr : dbl+lbl [1:19690]  4,  2,  0,  6,  2,  0, 99,  0,  1,  2,  0,  8,  6, ...
-##    ..@ label        : chr "Qualification for immigration: christian background"
-##    ..@ format.spss  : chr "F2.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:14] 0 1 2 3 4 5 6 7 8 9 ...
-##    .. ..- attr(*, "names")= chr [1:14] "Extremely unimportant" "1" "2" "3" ...
-##  $ qfimwht : dbl+lbl [1:19690]  1,  0,  0,  0,  0,  0, 99,  0,  0,  1,  0,  5,  0, ...
-##    ..@ label        : chr "Qualification for immigration: be white"
-##    ..@ format.spss  : chr "F2.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:14] 0 1 2 3 4 5 6 7 8 9 ...
-##    .. ..- attr(*, "names")= chr [1:14] "Extremely unimportant" "1" "2" "3" ...
-##  $ imwgdwn : dbl+lbl [1:19690]  3,  4,  2,  2,  3,  3, NA,  4, NA,  4,  5, NA,  3, ...
-##    ..@ label        : chr "Average wages/salaries generally brought down by immigrants"
-##    ..@ format.spss  : chr "F1.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:8] 1 2 3 4 5 7 8 9
-##    .. ..- attr(*, "names")= chr [1:8] "Agree strongly" "Agree" "Neither agree nor disagree" "Disagree" ...
-##  $ imhecop : dbl+lbl [1:19690]  2,  2,  1,  4,  3,  2, NA,  3, NA,  2,  5,  3,  3, ...
-##    ..@ label        : chr "Immigrants harm economic prospects of the poor more than the rich"
-##    ..@ format.spss  : chr "F1.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:8] 1 2 3 4 5 7 8 9
-##    .. ..- attr(*, "names")= chr [1:8] "Agree strongly" "Agree" "Neither agree nor disagree" "Disagree" ...
-##  $ imtcjob : dbl+lbl [1:19690]  7,  5,  6,  5,  7, 10, NA,  8, NA,  4,  5,  5,  5, ...
-##    ..@ label        : chr "Immigrants take jobs away in country or create new jobs"
-##    ..@ format.spss  : chr "F2.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:14] 0 1 2 3 4 5 6 7 8 9 ...
-##    .. ..- attr(*, "names")= chr [1:14] "Take jobs away" "1" "2" "3" ...
-##  $ imbleco : dbl+lbl [1:19690]  9,  4,  2, NA,  3, 10, NA,  9, NA,  6,  5, NA,  2, ...
-##    ..@ label        : chr "Taxes and services: immigrants take out more than they put in or less"
-##    ..@ format.spss  : chr "F2.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:14] 0 1 2 3 4 5 6 7 8 9 ...
-##    .. ..- attr(*, "names")= chr [1:14] "Generally take out more" "1" "2" "3" ...
-##  $ imbgeco : dbl+lbl [1:19690]  4,  3, 10,  7,  5, 10, NA,  8, NA,  5,  5,  5,  3, ...
-##    ..@ label        : chr "Immigration bad or good for country's economy"
-##    ..@ format.spss  : chr "F2.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:14] 0 1 2 3 4 5 6 7 8 9 ...
-##    .. ..- attr(*, "names")= chr [1:14] "Bad for the economy" "1" "2" "3" ...
-##  $ imueclt : dbl+lbl [1:19690]  9,  4, 10,  5,  4, 10, NA,  9, NA,  3,  7,  5,  2, ...
-##    ..@ label        : chr "Country's cultural life undermined or enriched by immigrants"
-##    ..@ format.spss  : chr "F2.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:14] 0 1 2 3 4 5 6 7 8 9 ...
-##    .. ..- attr(*, "names")= chr [1:14] "Cultural life undermined" "1" "2" "3" ...
-##  $ imwbcnt : dbl+lbl [1:19690]  7,  3,  5,  5,  5, 10, NA,  8, NA,  5,  5,  5,  2, ...
-##    ..@ label        : chr "Immigrants make country worse or better place to live"
-##    ..@ format.spss  : chr "F2.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:14] 0 1 2 3 4 5 6 7 8 9 ...
-##    .. ..- attr(*, "names")= chr [1:14] "Worse place to live" "1" "2" "3" ...
-##  $ imwbcrm : dbl+lbl [1:19690]  3,  3,  5,  2,  3,  5, NA,  5, NA,  3,  4,  2,  0, ...
-##    ..@ label        : chr "Immigrants make country's crime problems worse or better"
-##    ..@ format.spss  : chr "F2.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:14] 0 1 2 3 4 5 6 7 8 9 ...
-##    .. ..- attr(*, "names")= chr [1:14] "Crime problems made worse" "1" "2" "3" ...
-##  $ imrsprc : dbl+lbl [1:19690]  2,  2,  1,  4,  1,  2, NA,  1,  1,  3,  1, NA,  4, ...
-##    ..@ label        : chr "Richer countries should be responsible for accepting people from poorer countries"
-##    ..@ format.spss  : chr "F1.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:8] 1 2 3 4 5 7 8 9
-##    .. ..- attr(*, "names")= chr [1:8] "Agree strongly" "Agree" "Neither agree nor disagree" "Disagree" ...
-##  $ pplstrd : dbl+lbl [1:19690]  2,  4,  2,  2,  3,  4, NA,  4,  4,  2,  4,  1,  2, ...
-##    ..@ label        : chr "Better for a country if almost everyone share customs and traditions"
-##    ..@ format.spss  : chr "F1.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:8] 1 2 3 4 5 7 8 9
-##    .. ..- attr(*, "names")= chr [1:8] "Agree strongly" "Agree" "Neither agree nor disagree" "Disagree" ...
-##  $ vrtrlg  : dbl+lbl [1:19690]  3,  5,  3,  2,  4,  1, NA,  4,  2,  3,  2,  2,  4, ...
-##    ..@ label      : chr "Better for a country if a variety of different religions"
-##    ..@ format.spss: chr "F1.0"
-##    ..@ labels     : Named num [1:8] 1 2 3 4 5 7 8 9
-##    .. ..- attr(*, "names")= chr [1:8] "Agree strongly" "Agree" "Neither agree nor disagree" "Disagree" ...
-##  $ shrrfg  : dbl+lbl [1:19690]  3,  2,  1,  1,  3,  3, NA,  3,  4,  3,  3, NA,  2, ...
-##    ..@ label      : chr "Country has more than its fair share of people applying refugee status"
-##    ..@ format.spss: chr "F1.0"
-##    ..@ labels     : Named num [1:8] 1 2 3 4 5 7 8 9
-##    .. ..- attr(*, "names")= chr [1:8] "Agree strongly" "Agree" "Neither agree nor disagree" "Disagree" ...
-##  $ rfgawrk : dbl+lbl [1:19690]  2,  2,  1,  2,  2,  2, NA,  2,  1,  2,  2, NA,  3, ...
-##    ..@ label        : chr "People applying refugee status allowed to work while cases considered"
-##    ..@ format.spss  : chr "F1.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:8] 1 2 3 4 5 7 8 9
-##    .. ..- attr(*, "names")= chr [1:8] "Agree strongly" "Agree" "Neither agree nor disagree" "Disagree" ...
-##  $ gvrfgap : dbl+lbl [1:19690]  4,  3,  2,  4,  2,  2, NA,  3,  2,  4,  3,  4,  4, ...
-##    ..@ label        : chr "Government should be generous judging applications for refugee status"
-##    ..@ format.spss  : chr "F1.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:8] 1 2 3 4 5 7 8 9
-##    .. ..- attr(*, "names")= chr [1:8] "Agree strongly" "Agree" "Neither agree nor disagree" "Disagree" ...
-##  $ rfgfrpc : dbl+lbl [1:19690]  4,  3,  2,  4,  4,  4, NA,  4,  3,  4,  4,  3,  3, ...
-##    ..@ label        : chr "Most refugee applicants not in real fear of persecution own countries"
-##    ..@ format.spss  : chr "F1.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:8] 1 2 3 4 5 7 8 9
-##    .. ..- attr(*, "names")= chr [1:8] "Agree strongly" "Agree" "Neither agree nor disagree" "Disagree" ...
-##  $ rfggvfn : dbl+lbl [1:19690]  2,  3,  2,  4,  3,  2, NA,  2,  2,  2,  2, NA,  2, ...
-##    ..@ label        : chr "Financial support to refugee applicants while cases considered"
-##    ..@ format.spss  : chr "F1.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:8] 1 2 3 4 5 7 8 9
-##    .. ..- attr(*, "names")= chr [1:8] "Agree strongly" "Agree" "Neither agree nor disagree" "Disagree" ...
-##  $ rfgbfml : dbl+lbl [1:19690]  2,  3,  1,  2,  2,  1, NA,  4,  2,  3,  2, NA,  4, ...
-##    ..@ label        : chr "Granted refugees should be entitled to bring close family members"
-##    ..@ format.spss  : chr "F1.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:8] 1 2 3 4 5 7 8 9
-##    .. ..- attr(*, "names")= chr [1:8] "Agree strongly" "Agree" "Neither agree nor disagree" "Disagree" ...
-##  $ gndr    : dbl+lbl [1:19690]  1,  2,  1,  2,  2,  1, NA,  2,  2,  1,  2,  2,  2, ...
-##    ..@ label        : chr "Gender"
-##    ..@ format.spss  : chr "F1.0"
-##    ..@ display_width: int 6
-##    ..@ labels       : Named num [1:3] 1 2 9
-##    .. ..- attr(*, "names")= chr [1:3] "Male" "Female" "No answer"
-##  $ yrbrn   : dbl+lbl [1:19690] 1949, 1978, 1953, 1940, 1964, 1959,   NA, 1973, 1962...
-##    ..@ label        : chr "Year of birth"
-##    ..@ format.spss  : chr "F4.0"
-##    ..@ display_width: int 7
-##    ..@ labels       : Named num [1:3] 7777 8888 9999
-##    .. ..- attr(*, "names")= chr [1:3] "Refusal" "Don't know" "No answer"
-##  $ edulvl  : dbl+lbl [1:19690] NA,  3, NA, NA,  3, NA, NA,  6, NA,  5, NA, NA,  3, ...
-##    ..@ label      : chr "Highest level of education"
-##    ..@ format.spss: chr "F1.0"
-##    ..@ labels     : Named num [1:10] 0 1 2 3 4 5 6 7 8 9
-##    .. ..- attr(*, "names")= chr [1:10] "Not completed primary education" "Primary or first stage of basic" "Lower secondary or second stage of basic" "Upper secondary" ...
-##  $ eduyrs  : dbl+lbl [1:19690] 11, 16, 14,  9, 12, 18, NA, 17, 15, 17, 11, 10, 19, ...
-##    ..@ label      : chr "Years of full-time education completed"
-##    ..@ format.spss: chr "F2.0"
-##    ..@ labels     : Named num [1:3] 77 88 99
-##    .. ..- attr(*, "names")= chr [1:3] "Refusal" "Don't know" "No answer"
-##  $ polintr : dbl+lbl [1:19690] 3, 3, 1, 2, 3, 2, 1, 4, 3, 3, 1, 3, 3, 3, 1, 2, 3, 3...
-##    ..@ label        : chr "How interested in politics"
-##    ..@ format.spss  : chr "F1.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:7] 1 2 3 4 7 8 9
-##    .. ..- attr(*, "names")= chr [1:7] "Very interested" "Quite interested" "Hardly interested" "Not at all interested" ...
-##  $ lrscale : dbl+lbl [1:19690]  6,  7,  6,  5,  8,  5, NA,  8,  5,  7, NA, NA,  4, ...
-##    ..@ label        : chr "Placement on left right scale"
-##    ..@ format.spss  : chr "F2.0"
-##    ..@ display_width: int 9
-##    ..@ labels       : Named num [1:14] 0 1 2 3 4 5 6 7 8 9 ...
-##    .. ..- attr(*, "names")= chr [1:14] "Left" "1" "2" "3" ...
    -
    ## Compute descriptive statistics:
-descriptives(ess)
    -
    - -
    -
    - -Click for explanation - -

    One thing you might notice when inspecting the ess data is that all the -scale variables are stored as labelled vectors. When loading SPSS data, -haven will use these labelled vectors to preserve the metadata associated -with SPSS scale variables (i.e., variable labels and value labels).

    -

    While it’s good to have this metadata available, we want to analyze these items -as numeric variables, so the value labels are only going to make our lives -harder. Thankfully, the labelled package contains many routines for -manipulating labelled vectors.

    +

    In keeping with common practice, we will treat ordinal Likert-type rating scales +with five or more levels as continuous. Since some R routines will treat labelled +vectors as discrete variables, we can make things easier for ourselves by +converting all the labelled vectors in our data to numeric vectors.

    We can use the labelled::remove_val_labels() function to strip the value labels and convert all of the labelled vectors to numeric vectors.

    -
    -

    4.3.4

    -

    Correct any formatting issues that you discovered above.

    +

    Convert the remaining labelled vectors to numeric vectors.

    -Click for explanation +Click to see code -

    In keeping with common practice, we will treat ordinal Likert-type rating scales -with five or more levels as continuous. Since some R routines will treat labelled -vectors as discrete variables, we can make things easier for ourselves by -converting all the labelled vectors in our data to numeric vectors.

    -

    As noted above, we can easily accomplish this goal with the remove_val_labels() -function from the labelled package.

    -
    ## If necessary, install the labelled package:
-# install.packages("labelled", repos = "https://cloud.r-project.org")
-
-## Load the labelled package:
-library(labelled)
-
-## Strip the value labels:
-ess <- remove_val_labels(ess)
-
-## Check the effects:
-str(ess)
    -
    ## tibble [19,690 × 50] (S3: tbl_df/tbl/data.frame)
+
    ## If necessary, install the labelled package:
+# install.packages("labelled", repos = "https://cloud.r-project.org")
+
+## Load the labelled package:
+library(labelled)
+
+## Strip the value labels:
+ess <- remove_val_labels(ess)
+
+## Check the effects:
+str(ess)
    
+
    ## tibble [19,690 × 52] (S3: tbl_df/tbl/data.frame)
 ##  $ name    : chr [1:19690] "ESS1e06_1" "ESS1e06_1" "ESS1e06_1" "ESS1e06_1" ...
 ##   ..- attr(*, "label")= chr "Title of dataset"
 ##   ..- attr(*, "format.spss")= chr "A9"
@@ -1441,7 +1069,10 @@ 
    4.3.4 
    
 ##   ..- attr(*, "label")= chr "Production date"
 ##   ..- attr(*, "format.spss")= chr "A10"
 ##   ..- attr(*, "display_width")= int 12
-##  $ cntry   : Factor w/ 10 levels "Austria","Belgium",..: 1 9 1 1 9 1 2 9 1 9 ...
+##  $ cntry   : num [1:19690] 1 18 1 1 18 1 2 18 1 18 ...
+##   ..- attr(*, "label")= chr "Country"
+##   ..- attr(*, "format.spss")= chr "F2.0"
+##   ..- attr(*, "display_width")= int 7
 ##  $ idno    : num [1:19690] 1 1 2 3 3 4 4 4 6 6 ...
 ##   ..- attr(*, "label")= chr "Respondent's identification number"
 ##   ..- attr(*, "format.spss")= chr "F9.0"
@@ -1598,31 +1229,33 @@ 
    4.3.4 
    
 ##   ..- attr(*, "label")= chr "Year of birth"
 ##   ..- attr(*, "format.spss")= chr "F4.0"
 ##   ..- attr(*, "display_width")= int 7
-##  $ edulvl  : num [1:19690] NA 3 NA NA 3 NA NA 6 NA 5 ...
-##   ..- attr(*, "label")= chr "Highest level of education"
-##   ..- attr(*, "format.spss")= chr "F1.0"
+##  $ edulvl  : Ord.factor w/ 7 levels "Not completed primary education"<..: NA 4 NA NA 4 NA NA 7 NA 6 ...
 ##  $ eduyrs  : num [1:19690] 11 16 14 9 12 18 NA 17 15 17 ...
 ##   ..- attr(*, "label")= chr "Years of full-time education completed"
 ##   ..- attr(*, "format.spss")= chr "F2.0"
-##  $ polintr : num [1:19690] 3 3 1 2 3 2 1 4 3 3 ...
-##   ..- attr(*, "label")= chr "How interested in politics"
-##   ..- attr(*, "format.spss")= chr "F1.0"
-##   ..- attr(*, "display_width")= int 9
+##  $ polintr : Ord.factor w/ 4 levels "Very interested"<..: 3 3 1 2 3 2 1 4 3 3 ...
 ##  $ lrscale : num [1:19690] 6 7 6 5 8 5 NA 8 5 7 ...
 ##   ..- attr(*, "label")= chr "Placement on left right scale"
 ##   ..- attr(*, "format.spss")= chr "F2.0"
-##   ..- attr(*, "display_width")= int 9
    
-
    descriptives(ess)
    
+##   ..- attr(*, "display_width")= int 9
+##  $ country : Factor w/ 10 levels "Austria","Belgium",..: 1 9 1 1 9 1 2 9 1 9 ...
+##  $ sex     : Factor w/ 2 levels "Male","Female": 1 2 1 2 2 1 NA 2 2 1 ...
    +
    descriptives(ess)
    -

    Note that the variables are now simple numeric vectors, but the variable labels -have been retained as column attributes (which is probably useful). If we want -to completely nuke the labelling information, we can use the +

    + +Click for explanation + +

    Note that the numeric variables are now simple numeric vectors, but the +variable labels have been retained as column attributes (which is probably +useful). If we want to completely nuke the labelling information, we can use the labelled::remove_labels() function to do so.

    +

    In addition to screening with summary statistics, we can also visualize the variables’ distributions. You have already created a few such visualizations @@ -1645,14 +1278,14 @@

    4.3.5

    Click to show code -
    ## Load the tidyr package:
-library(tidyr)
-
-## Convert the target variables into a long-formatted data frame:
-ess_plot <- pivot_longer(ess,
-                         cols = trstlgl:rfgbfml, # Which columns to convert
-                         names_to = "variable",  # Name for the new grouping variable
-                         values_to = "value")    # Name for the column of stacked values
    +
    ## Load the tidyr package:
+library(tidyr)
+
+## Convert the target variables into a long-formatted data frame:
+ess_plot <- pivot_longer(ess,
+                         cols = trstlgl:rfgbfml, # Which columns to convert
+                         names_to = "variable",  # Name for the new grouping variable
+                         values_to = "value")    # Name for the column of stacked values

    The next step in the process will be to plot the variables using ggplot(). In @@ -1676,11 +1309,11 @@

    4.3.6

    Click to show code -
    library(ggplot2)
-
-ggplot(ess_plot, aes(x = value)) + 
-  geom_histogram() +                        # Create a histogram
-  facet_wrap(~ variable, scales = "free_x") # Facet on 'variable'
    +
    library(ggplot2)
+
+ggplot(ess_plot, aes(x = value)) + 
+  geom_histogram() +                        # Create a histogram
+  facet_wrap(~ variable, scales = "free_x") # Facet on 'variable'

    @@ -1706,7 +1339,7 @@

    4.3.7

    Click to show code -
    select(ess, trstlgl:rfgbfml) %>% descriptives()
    +
    select(ess, trstlgl:rfgbfml) %>% descriptives()
    +
    +
    + +Click to show explanation +

    Here, we need to tell R that these values should be considered missing, or NA. Otherwise they will contribute the numeric value to the analysis, as though someone had provided an answer of 77 on a 10-point scale.

    -
    ess <- ess %>%
-  mutate(across(c(qfimchr, qfimwht), na_if, 77)) %>% 
-  mutate(across(c(qfimchr, qfimwht), na_if, 88)) %>%
-  mutate(across(c(qfimchr, qfimwht), na_if, 99))
    +

    We’ve done quite a bit of data processing, and we’ll continue to use these data +for several future practicals, so it would be a good idea to save the processed +dataset for later use. When saving data that you plan to analyze in R, you will +usually want to use the R Data Set (RDS) format.

    +

    Datasets saved in RDS format retain all of their attributes and formatting (e.g., +factor are still factors, missing values are coded as NA, etc.). So, you +don’t have to redo any data processing before future analyses.

    +
    +
    +
    +

    4.3.9

    +

    Use the saveRDS() function to save the processed dataset.

    +
    + +Click to show code + +
    ## Save the processed data:
+saveRDS(ess, "ess_round1.rds")
    +
    +

    Now, we’re ready to run the analyses and see if we can replicate the Kestilä (2006) results.

    -
    -

    4.3.8

    +
    +

    4.3.10

    Run two principal component analyses (PCA): one for trust in politics, one for attitudes towards immigration.

      @@ -1762,15 +1438,15 @@

      4.3.8

      Trust in politics

      Kestilä extracted three components with VARIMAX rotation.

      -
      ## Load the psych package:
-library(psych)
-
-## Run the PCA:
-pca_trust <- select(ess, trstlgl:trstplt) %>% 
-  principal(nfactors = 3, rotate = "varimax")
-
-## Print the results:
-print(pca_trust, cut = 0.3, digits = 3)
      +
      ## Load the psych package:
+library(psych)
+
+## Run the PCA:
+pca_trust <- select(ess, trstlgl:trstplt) %>% 
+  principal(nfactors = 3, rotate = "varimax")
+
+## Print the results:
+print(pca_trust, cut = 0.3, digits = 3)
      ## Principal Components Analysis
 ## Call: principal(r = ., nfactors = 3, rotate = "varimax")
 ## Standardized loadings (pattern matrix) based upon correlation matrix
@@ -1808,10 +1484,10 @@ 
      Trust in politics
 
      Attitudes toward immigration
      
 
      Kestilä extracted five components with VARIMAX rotation.
      
-
      pca_att <- select(ess, imsmetn:rfgbfml) %>%
-  principal(nfactors = 5, rotate = "varimax")
-
-print(pca_att, cut = 0.3, digits = 3)
      
+
      pca_att <- select(ess, imsmetn:rfgbfml) %>%
+  principal(nfactors = 5, rotate = "varimax")
+
+print(pca_att, cut = 0.3, digits = 3)
      
 
      ## Principal Components Analysis
 ## Call: principal(r = ., nfactors = 5, rotate = "varimax")
 ## Standardized loadings (pattern matrix) based upon correlation matrix
@@ -1870,8 +1546,8 @@ 
      Attitudes toward immigration
-
      4.3.9 
      
+
      
+
      4.3.11 
      
 
      Add the component scores produced by the analyses you ran above to
 the ess data frame.
      
 
        
@@ -1894,12 +1570,12 @@ 
        4.3.9 
        
 
 Click to show code
 
-
        ## Save the component scores in stand-alone matrices: 
-trust_scores <- pca_trust$scores
-att_scores <- pca_att$scores
-
-## Inspect the result:
-head(trust_scores)
        
+
        ## Save the component scores in stand-alone matrices: 
+trust_scores <- pca_trust$scores
+att_scores <- pca_att$scores
+
+## Inspect the result:
+head(trust_scores)
        
 
        ##              RC3         RC2       RC1
 ## [1,]          NA          NA        NA
 ## [2,]  0.09755193 -0.01552183  0.994954
@@ -1907,7 +1583,7 @@ 
        4.3.9 
        
 ## [4,]          NA          NA        NA
 ## [5,] -0.21112678  0.84370377  1.200007
 ## [6,]  1.86596955  0.31083233 -1.062603
        
-
        summary(trust_scores)
        
+
        summary(trust_scores)
        
 
        ##       RC3              RC2              RC1        
 ##  Min.   :-4.035   Min.   :-3.706   Min.   :-3.139  
 ##  1st Qu.:-0.527   1st Qu.:-0.652   1st Qu.:-0.649  
@@ -1916,7 +1592,7 @@ 
        4.3.9 
        
 ##  3rd Qu.: 0.727   3rd Qu.: 0.742   3rd Qu.: 0.742  
 ##  Max.   : 3.302   Max.   : 3.452   Max.   : 3.539  
 ##  NA's   :4912     NA's   :4912     NA's   :4912
        
-
        head(att_scores)
        
+
        head(att_scores)
        
 
        ##             RC2        RC1        RC5         RC3        RC4
 ## [1,]  1.9873715  1.3233586 -0.8382499 -0.02172765 -0.0908143
 ## [2,]  0.1692841 -1.2178436 -0.5016936 -0.21749066  0.6758844
@@ -1924,7 +1600,7 @@ 
        4.3.9 
        
 ## [4,]         NA         NA         NA          NA         NA
 ## [5,] -0.1137484 -0.7891232 -1.4732563 -0.05843873  0.4110692
 ## [6,] -0.9195530  2.8231404 -0.3480398 -0.75699796 -1.3230602
        
-
        summary(att_scores)
        
+
        summary(att_scores)
        
 
        ##       RC2              RC1              RC5              RC3        
 ##  Min.   :-3.660   Min.   :-3.929   Min.   :-3.824   Min.   :-2.764  
 ##  1st Qu.:-0.616   1st Qu.:-0.585   1st Qu.:-0.656   1st Qu.:-0.748  
@@ -1956,20 +1632,20 @@ 
        4.3.9 
        
 
 Click to show code
 
-
        ## Check names (note the order):
-colnames(trust_scores)
        
+
        ## Check names (note the order):
+colnames(trust_scores)
        
 
        ## [1] "RC3" "RC2" "RC1"
        
-
        colnames(att_scores)
        
+
        colnames(att_scores)
        
 
        ## [1] "RC2" "RC1" "RC5" "RC3" "RC4"
        
-
        ## Give informative names:
-colnames(trust_scores) <- c("Trust_Institutions", 
-                            "Satisfaction", "
-                            Trust_Politicians")
-colnames(att_scores) <- c("Quantity", 
-                          "Effects", 
-                          "Refugees", 
-                          "Diversity", 
-                          "Economic")
        
+
        ## Give informative names:
+colnames(trust_scores) <- c("Trust_Institutions", 
+                            "Satisfaction",
+                            "Trust_Politicians")
+colnames(att_scores) <- c("Quantity", 
+                          "Effects", 
+                          "Refugees", 
+                          "Diversity", 
+                          "Economic")
        
 
 
        
 
        3. Add the component scores to the dataset
        
@@ -1977,13 +1653,13 @@ 
        4.3.9 
        
 
 Click to show code
 
-
        # Add the component scores to the 'ess' data:
-ess <- data.frame(ess, trust_scores, att_scores)
        
+
        # Add the component scores to the 'ess' data:
+ess <- data.frame(ess, trust_scores, att_scores)
        
 
 
        
 
      
-
      
-
      4.3.10 
      
+
      
+
      4.3.12 
      
 
      Were you able to replicate the results of Kestilä (2006)?
      
 
      
 
diff --git a/docs/at-home-exercises-4.html b/docs/at-home-exercises-4.html
index ac9e3df5..ac384d21 100644
--- a/docs/at-home-exercises-4.html
+++ b/docs/at-home-exercises-4.html
@@ -333,28 +333,31 @@
 
    • 4.3.5 
      • 
 
    • 4.3.6 
      • 
 
    • 4.3.7 
      • 
-
    • 4.3.8 
      • 
-
    • 4.3.9 
      • 
-
    • 4.3.10 
      • 
+
    • 4.3.8 
      • 
+
    • 4.3.9 
      • 
+
    • 4.3.10 
      • 
+
    • 4.3.11 
      • 
+
    • 4.3.12 
      •

  • 4.4 In-Class Exercises
  • 5 CFA @@ -371,30 +374,30 @@
  • 5.3 At-Home Exercises
  • 5.4 In-Class Exercises
  • 6 Full SEM @@ -411,23 +414,22 @@
  • 6.3 At-Home Exercises
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models @@ -466,24 +468,22 @@

    5.3 At-Home Exercises +

    5.3.1

    Load the ESS data.

      -
    • The relevant data are contained in the ess_round1_trust.rds file. +
    • The relevant data are contained in the ess_round1.rds file.
      • This file is in R Data Set (RDS) format.
      • The dataset is already stored as a data frame with the processing and cleaning that you should have done for previous practicals completed.
        • -
      • Check the documentation for the readRDS() function to see how you can -load data stored in an RDS file.
    Click to show code -
    ess <- readRDS("ess_round1_trust.rds")
    +
    ess <- readRDS("ess_round1.rds")

    Although you may have settled on any number of EFA solutions during the @@ -498,7 +498,7 @@

    5.3.1

    5.3.2

    Define the lavaan model syntax for the CFA implied by the three-factor EFA -solution you found in the Week 2 In-Class Exercises.

    +solution you found in the Week 4 In-Class Exercises.

    • Covary the three latent factors.
    • Do not specify any mean structure.
      • @@ -508,11 +508,11 @@

      5.3.2

      Click to show code -
      mod_3f <- '
-politicians  =~ pltcare + pltinvt + trstplt
-satisfaction =~ stfeco  + stfgov  + stfdem + stfedu  + stfhlth
-institutions =~ trstlgl + trstplc + trstun + trstprl
-'
      +
      mod_3f <- '
+institutions =~ trstlgl + trstplc + trstun + trstep + trstprl
+satisfaction =~ stfhlth + stfedu  + stfeco + stfgov + stfdem
+politicians  =~ pltinvt + pltcare + trstplt
+'
      Click for explanation @@ -523,7 +523,7 @@

      5.3.2

    -
    +

    5.3.3

    Estimate the CFA model you defined above, and summarize the results.

      @@ -545,60 +545,60 @@

      5.3.3

      Click the code -
      ## Load the lavaan package:
-library(lavaan)
-
-## Estimate the CFA model:
-fit_3f <- cfa(mod_3f, data = ess)
-
-## Summarize the fitted model:
-summary(fit_3f, fit.measures = TRUE, standardized = TRUE)
      -
      ## lavaan 0.6.16 ended normally after 45 iterations
+
      ## Load the lavaan package:
+library(lavaan)
+
+## Estimate the CFA model:
+fit_3f <- cfa(mod_3f, data = ess)
+
+## Summarize the fitted model:
+summary(fit_3f, fit.measures = TRUE, standardized = TRUE)
      
+
      ## lavaan 0.6.16 ended normally after 46 iterations
 ## 
 ##   Estimator                                         ML
 ##   Optimization method                           NLMINB
-##   Number of model parameters                        27
+##   Number of model parameters                        29
 ## 
 ##                                                   Used       Total
-##   Number of observations                         15448       18187
+##   Number of observations                         14778       19690
 ## 
 ## Model Test User Model:
-##                                                       
-##   Test statistic                              9188.922
-##   Degrees of freedom                                51
-##   P-value (Chi-square)                           0.000
+##                                                        
+##   Test statistic                              10652.207
+##   Degrees of freedom                                 62
+##   P-value (Chi-square)                            0.000
 ## 
 ## Model Test Baseline Model:
 ## 
-##   Test statistic                             75675.049
-##   Degrees of freedom                                66
+##   Test statistic                             81699.096
+##   Degrees of freedom                                78
 ##   P-value                                        0.000
 ## 
 ## User Model versus Baseline Model:
 ## 
-##   Comparative Fit Index (CFI)                    0.879
-##   Tucker-Lewis Index (TLI)                       0.844
+##   Comparative Fit Index (CFI)                    0.870
+##   Tucker-Lewis Index (TLI)                       0.837
 ## 
 ## Loglikelihood and Information Criteria:
 ## 
-##   Loglikelihood user model (H0)            -357923.209
-##   Loglikelihood unrestricted model (H1)    -353328.748
+##   Loglikelihood user model (H0)            -371404.658
+##   Loglikelihood unrestricted model (H1)    -366078.555
 ##                                                       
-##   Akaike (AIC)                              715900.419
-##   Bayesian (BIC)                            716106.840
-##   Sample-size adjusted Bayesian (SABIC)     716021.036
+##   Akaike (AIC)                              742867.317
+##   Bayesian (BIC)                            743087.743
+##   Sample-size adjusted Bayesian (SABIC)     742995.583
 ## 
 ## Root Mean Square Error of Approximation:
 ## 
 ##   RMSEA                                          0.108
 ##   90 Percent confidence interval - lower         0.106
-##   90 Percent confidence interval - upper         0.110
+##   90 Percent confidence interval - upper         0.109
 ##   P-value H_0: RMSEA <= 0.050                    0.000
 ##   P-value H_0: RMSEA >= 0.080                    1.000
 ## 
 ## Standardized Root Mean Square Residual:
 ## 
-##   SRMR                                           0.058
+##   SRMR                                           0.059
 ## 
 ## Parameter Estimates:
 ## 
@@ -608,47 +608,49 @@ 
      5.3.3 
      
 ## 
 ## Latent Variables:
 ##                    Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
-##   politicians =~                                                        
-##     pltcare           1.000                               0.671    0.639
-##     pltinvt           0.981    0.015   65.449    0.000    0.658    0.624
-##     trstplt           2.848    0.035   80.855    0.000    1.911    0.876
-##   satisfaction =~                                                       
-##     stfeco            1.000                               1.659    0.712
-##     stfgov            1.039    0.013   81.895    0.000    1.724    0.752
-##     stfdem            0.979    0.012   80.257    0.000    1.624    0.733
-##     stfedu            0.780    0.012   64.344    0.000    1.294    0.576
-##     stfhlth           0.706    0.012   58.239    0.000    1.171    0.519
 ##   institutions =~                                                       
-##     trstlgl           1.000                               1.640    0.688
-##     trstplc           0.777    0.012   64.538    0.000    1.275    0.582
-##     trstun            0.861    0.013   66.949    0.000    1.411    0.605
-##     trstprl           1.123    0.013   85.721    0.000    1.842    0.812
+##     trstlgl           1.000                               1.613    0.677
+##     trstplc           0.770    0.012   61.866    0.000    1.241    0.567
+##     trstun            0.929    0.013   69.227    0.000    1.498    0.642
+##     trstep            0.908    0.013   70.929    0.000    1.464    0.660
+##     trstprl           1.139    0.014   84.084    0.000    1.837    0.809
+##   satisfaction =~                                                       
+##     stfhlth           1.000                               1.173    0.521
+##     stfedu            1.106    0.022   50.840    0.000    1.297    0.577
+##     stfeco            1.415    0.025   57.214    0.000    1.659    0.713
+##     stfgov            1.480    0.025   58.764    0.000    1.736    0.756
+##     stfdem            1.384    0.024   57.904    0.000    1.623    0.731
+##   politicians =~                                                        
+##     pltinvt           1.000                               0.646    0.613
+##     pltcare           1.021    0.016   62.862    0.000    0.660    0.628
+##     trstplt           3.012    0.039   76.838    0.000    1.946    0.891
 ## 
 ## Covariances:
 ##                    Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
-##   politicians ~~                                                        
-##     satisfaction      0.793    0.016   48.672    0.000    0.712    0.712
-##     institutions      0.950    0.018   51.888    0.000    0.863    0.863
+##   institutions ~~                                                       
+##     satisfaction      1.391    0.032   43.206    0.000    0.736    0.736
+##     politicians       0.909    0.018   49.934    0.000    0.872    0.872
 ##   satisfaction ~~                                                       
-##     institutions      2.046    0.040   51.736    0.000    0.752    0.752
+##     politicians       0.539    0.013   41.053    0.000    0.711    0.711
 ## 
 ## Variances:
 ##                    Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
-##    .pltcare           0.651    0.008   77.811    0.000    0.651    0.591
-##    .pltinvt           0.680    0.009   78.658    0.000    0.680    0.611
-##    .trstplt           1.103    0.029   38.414    0.000    1.103    0.232
-##    .stfeco            2.671    0.038   70.723    0.000    2.671    0.493
-##    .stfgov            2.287    0.035   66.172    0.000    2.287    0.435
-##    .stfdem            2.266    0.033   68.444    0.000    2.266    0.462
-##    .stfedu            3.378    0.042   79.725    0.000    3.378    0.668
-##    .stfhlth           3.721    0.045   81.846    0.000    3.721    0.731
-##    .trstlgl           2.997    0.040   74.548    0.000    2.997    0.527
-##    .trstplc           3.178    0.040   80.410    0.000    3.178    0.662
-##    .trstun            3.443    0.043   79.405    0.000    3.443    0.633
-##    .trstprl           1.746    0.030   57.842    0.000    1.746    0.340
-##     politicians       0.450    0.011   41.572    0.000    1.000    1.000
-##     satisfaction      2.751    0.058   47.286    0.000    1.000    1.000
-##     institutions      2.690    0.059   45.613    0.000    1.000    1.000
      
+##    .trstlgl           3.068    0.041   75.262    0.000    3.068    0.541
+##    .trstplc           3.248    0.041   80.037    0.000    3.248    0.678
+##    .trstun            3.197    0.041   77.141    0.000    3.197    0.588
+##    .trstep            2.776    0.036   76.243    0.000    2.776    0.564
+##    .trstprl           1.776    0.029   61.361    0.000    1.776    0.345
+##    .stfhlth           3.695    0.046   79.989    0.000    3.695    0.729
+##    .stfedu            3.368    0.043   77.916    0.000    3.368    0.667
+##    .stfeco            2.656    0.038   69.070    0.000    2.656    0.491
+##    .stfgov            2.264    0.035   64.201    0.000    2.264    0.429
+##    .stfdem            2.289    0.034   67.172    0.000    2.289    0.465
+##    .pltinvt           0.694    0.009   78.255    0.000    0.694    0.624
+##    .pltcare           0.668    0.009   77.562    0.000    0.668    0.605
+##    .trstplt           0.978    0.028   34.461    0.000    0.978    0.205
+##     institutions      2.601    0.059   44.198    0.000    1.000    1.000
+##     satisfaction      1.375    0.044   31.407    0.000    1.000    1.000
+##     politicians       0.417    0.011   38.843    0.000    1.000    1.000
      Click for explanation @@ -680,7 +682,7 @@

      5.3.3

      model wherein all Trust items are explained by a single latent variable.

    -
    +

    5.3.4

    Define the lavaan model syntax for a one-factor model of the Trust items.

      @@ -690,26 +692,26 @@

      5.3.4

      Click to show code -
      mod_1f <- '
-political_trust =~ 
-  pltcare + 
-  pltinvt + 
-  trstprl + 
-  trstplt + 
-  stfeco + 
-  stfgov + 
-  stfdem + 
-  stfedu + 
-  stfhlth + 
-  trstlgl + 
-  trstplc + 
-  trstun + 
-  trstep
-'
      +
      mod_1f <- '
+political_trust =~ 
+  trstlgl +
+  trstplc +
+  trstun +
+  trstep +
+  trstprl +
+  stfhlth +
+  stfedu  +
+  stfeco +
+  stfgov +
+  stfdem +
+  pltinvt +
+  pltcare +
+  trstplt
+'
    -
    +

    5.3.5

    Estimate the one-factor model, and summarize the results.

      @@ -721,19 +723,19 @@

      5.3.5

      Click to show code -
      ## Estimate the one factor model:
-fit_1f <- cfa(mod_1f, data = ess)
-
-## Summarize the results:
-summary(fit_1f, fit.measures = TRUE)
      -
      ## lavaan 0.6.16 ended normally after 38 iterations
+
      ## Estimate the one factor model:
+fit_1f <- cfa(mod_1f, data = ess)
+
+## Summarize the results:
+summary(fit_1f, fit.measures = TRUE)
      
+
      ## lavaan 0.6.16 ended normally after 33 iterations
 ## 
 ##   Estimator                                         ML
 ##   Optimization method                           NLMINB
 ##   Number of model parameters                        26
 ## 
 ##                                                   Used       Total
-##   Number of observations                         14778       18187
+##   Number of observations                         14778       19690
 ## 
 ## Model Test User Model:
 ##                                                        
@@ -782,71 +784,71 @@ 
      5.3.5 
      
 ## Latent Variables:
 ##                      Estimate  Std.Err  z-value  P(>|z|)
 ##   political_trust =~                                    
-##     pltcare             1.000                           
-##     pltinvt             0.966    0.018   55.075    0.000
-##     trstprl             2.986    0.043   70.171    0.000
-##     trstplt             2.988    0.042   71.755    0.000
-##     stfeco              2.262    0.039   57.544    0.000
-##     stfgov              2.489    0.040   62.079    0.000
-##     stfdem              2.522    0.039   64.095    0.000
-##     stfedu              1.756    0.036   48.642    0.000
-##     stfhlth             1.554    0.035   43.930    0.000
-##     trstlgl             2.526    0.041   61.195    0.000
-##     trstplc             1.956    0.036   54.052    0.000
-##     trstun              2.350    0.040   59.017    0.000
-##     trstep              2.296    0.038   60.160    0.000
+##     trstlgl             1.000                           
+##     trstplc             0.774    0.013   57.949    0.000
+##     trstun              0.930    0.014   64.200    0.000
+##     trstep              0.909    0.014   65.679    0.000
+##     trstprl             1.182    0.015   79.401    0.000
+##     stfhlth             0.615    0.013   45.947    0.000
+##     stfedu              0.695    0.014   51.424    0.000
+##     stfeco              0.895    0.014   62.316    0.000
+##     stfgov              0.985    0.014   68.200    0.000
+##     stfdem              0.998    0.014   70.899    0.000
+##     pltinvt             0.382    0.006   59.215    0.000
+##     pltcare             0.396    0.006   61.195    0.000
+##     trstplt             1.183    0.014   81.716    0.000
 ## 
 ## Variances:
 ##                    Estimate  Std.Err  z-value  P(>|z|)
-##    .pltcare           0.743    0.009   81.579    0.000
-##    .pltinvt           0.775    0.009   82.043    0.000
-##    .trstprl           1.938    0.027   70.877    0.000
-##    .trstplt           1.548    0.023   67.052    0.000
-##    .stfeco            3.565    0.044   81.289    0.000
-##    .stfgov            3.044    0.038   79.326    0.000
-##    .stfdem            2.631    0.034   78.072    0.000
-##    .stfedu            3.941    0.047   83.419    0.000
-##    .stfhlth           4.201    0.050   84.093    0.000
 ##    .trstlgl           3.370    0.042   79.787    0.000
 ##    .trstplc           3.410    0.041   82.311    0.000
 ##    .trstun            3.451    0.043   80.749    0.000
 ##    .trstep            3.019    0.038   80.272    0.000
-##     political_trst    0.360    0.010   36.350    0.000
      
-
      ## Compare fit statistics:
-fitMeasures(fit_3f)
      
+##    .trstprl           1.938    0.027   70.878    0.000
+##    .stfhlth           4.201    0.050   84.093    0.000
+##    .stfedu            3.941    0.047   83.419    0.000
+##    .stfeco            3.565    0.044   81.289    0.000
+##    .stfgov            3.044    0.038   79.326    0.000
+##    .stfdem            2.631    0.034   78.072    0.000
+##    .pltinvt           0.775    0.009   82.043    0.000
+##    .pltcare           0.743    0.009   81.579    0.000
+##    .trstplt           1.548    0.023   67.052    0.000
+##     political_trst    2.299    0.055   41.569    0.000
      +
      ## Compare fit statistics:
+fitMeasures(fit_3f)
      ##                  npar                  fmin                 chisq 
-##                27.000                 0.297              9188.922 
+##                29.000                 0.360             10652.207 
 ##                    df                pvalue        baseline.chisq 
-##                51.000                 0.000             75675.049 
+##                62.000                 0.000             81699.096 
 ##           baseline.df       baseline.pvalue                   cfi 
-##                66.000                 0.000                 0.879 
+##                78.000                 0.000                 0.870 
 ##                   tli                  nnfi                   rfi 
-##                 0.844                 0.844                 0.843 
+##                 0.837                 0.837                 0.836 
 ##                   nfi                  pnfi                   ifi 
-##                 0.879                 0.679                 0.879 
+##                 0.870                 0.691                 0.870 
 ##                   rni                  logl     unrestricted.logl 
-##                 0.879           -357923.209           -353328.748 
+##                 0.870           -371404.658           -366078.555 
 ##                   aic                   bic                ntotal 
-##            715900.419            716106.840             15448.000 
+##            742867.317            743087.743             14778.000 
 ##                  bic2                 rmsea        rmsea.ci.lower 
-##            716021.036                 0.108                 0.106 
+##            742995.583                 0.108                 0.106 
 ##        rmsea.ci.upper        rmsea.ci.level          rmsea.pvalue 
-##                 0.110                 0.900                 0.000 
+##                 0.109                 0.900                 0.000 
 ##        rmsea.close.h0 rmsea.notclose.pvalue     rmsea.notclose.h0 
 ##                 0.050                 1.000                 0.080 
 ##                   rmr            rmr_nomean                  srmr 
-##                 0.245                 0.245                 0.058 
+##                 0.255                 0.255                 0.059 
 ##          srmr_bentler   srmr_bentler_nomean                  crmr 
-##                 0.058                 0.058                 0.064 
+##                 0.059                 0.059                 0.064 
 ##           crmr_nomean            srmr_mplus     srmr_mplus_nomean 
-##                 0.064                 0.058                 0.058 
+##                 0.064                 0.059                 0.059 
 ##                 cn_05                 cn_01                   gfi 
-##               116.444               131.098                 0.905 
+##               113.901               126.971                 0.897 
 ##                  agfi                  pgfi                   mfi 
-##                 0.854                 0.591                 0.744 
+##                 0.849                 0.611                 0.699 
 ##                  ecvi 
-##                 0.598
      -
      fitMeasures(fit_1f)
      +## 0.725 +
      fitMeasures(fit_1f)
      ##                  npar                  fmin                 chisq 
 ##                26.000                 0.598             17667.304 
 ##                    df                pvalue        baseline.chisq 
@@ -892,7 +894,7 @@ 
      5.3.5 
      
 order CFA is only appropriate in certain circumstances.
    -
    +

    5.3.6

    Given the CFA results above, would a second order CFA be appropriate for the Trust data? Why or why not?

    @@ -911,7 +913,7 @@

    5.3.6

    -
    +

    5.3.7

    Define the lavaan model syntax for a second-order CFA model of the Trust items.

      @@ -921,13 +923,13 @@

      5.3.7

      Click to show code -
      mod_2nd <- '
-politicians  =~ pltcare + pltinvt + trstplt
-satisfaction =~ stfeco  + stfgov  + stfdem + stfedu  + stfhlth
-institutions =~ trstlgl + trstplc + trstun + trstprl
-
-trust =~ politicians + satisfaction + institutions
-'
      +
      mod_2nd <- '
+institutions =~ trstlgl + trstplc + trstun + trstep + trstprl
+satisfaction =~ stfhlth + stfedu  + stfeco + stfgov + stfdem
+politicians  =~ pltinvt + pltcare + trstplt
+
+trust =~ politicians + satisfaction + institutions
+'
      Click for explanation @@ -940,7 +942,7 @@

      5.3.7

    -
    +

    5.3.8

    Estimate the second order CFA model, and summarize the results.

      @@ -954,54 +956,54 @@

      5.3.8

      Click to show code -
      fit_2nd <- cfa(mod_2nd, data = ess)
-summary(fit_2nd, fit.measures = TRUE, standardized = TRUE)
      -
      ## lavaan 0.6.16 ended normally after 45 iterations
+
      fit_2nd <- cfa(mod_2nd, data = ess)
+summary(fit_2nd, fit.measures = TRUE, standardized = TRUE)
      
+
      ## lavaan 0.6.16 ended normally after 44 iterations
 ## 
 ##   Estimator                                         ML
 ##   Optimization method                           NLMINB
-##   Number of model parameters                        27
+##   Number of model parameters                        29
 ## 
 ##                                                   Used       Total
-##   Number of observations                         15448       18187
+##   Number of observations                         14778       19690
 ## 
 ## Model Test User Model:
-##                                                       
-##   Test statistic                              9188.922
-##   Degrees of freedom                                51
-##   P-value (Chi-square)                           0.000
+##                                                        
+##   Test statistic                              10652.207
+##   Degrees of freedom                                 62
+##   P-value (Chi-square)                            0.000
 ## 
 ## Model Test Baseline Model:
 ## 
-##   Test statistic                             75675.049
-##   Degrees of freedom                                66
+##   Test statistic                             81699.096
+##   Degrees of freedom                                78
 ##   P-value                                        0.000
 ## 
 ## User Model versus Baseline Model:
 ## 
-##   Comparative Fit Index (CFI)                    0.879
-##   Tucker-Lewis Index (TLI)                       0.844
+##   Comparative Fit Index (CFI)                    0.870
+##   Tucker-Lewis Index (TLI)                       0.837
 ## 
 ## Loglikelihood and Information Criteria:
 ## 
-##   Loglikelihood user model (H0)            -357923.209
-##   Loglikelihood unrestricted model (H1)    -353328.748
+##   Loglikelihood user model (H0)            -371404.658
+##   Loglikelihood unrestricted model (H1)    -366078.555
 ##                                                       
-##   Akaike (AIC)                              715900.419
-##   Bayesian (BIC)                            716106.840
-##   Sample-size adjusted Bayesian (SABIC)     716021.036
+##   Akaike (AIC)                              742867.317
+##   Bayesian (BIC)                            743087.743
+##   Sample-size adjusted Bayesian (SABIC)     742995.583
 ## 
 ## Root Mean Square Error of Approximation:
 ## 
 ##   RMSEA                                          0.108
 ##   90 Percent confidence interval - lower         0.106
-##   90 Percent confidence interval - upper         0.110
+##   90 Percent confidence interval - upper         0.109
 ##   P-value H_0: RMSEA <= 0.050                    0.000
 ##   P-value H_0: RMSEA >= 0.080                    1.000
 ## 
 ## Standardized Root Mean Square Residual:
 ## 
-##   SRMR                                           0.058
+##   SRMR                                           0.059
 ## 
 ## Parameter Estimates:
 ## 
@@ -1011,111 +1013,113 @@ 
      5.3.8 
      
 ## 
 ## Latent Variables:
 ##                    Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
-##   politicians =~                                                        
-##     pltcare           1.000                               0.671    0.639
-##     pltinvt           0.981    0.015   65.449    0.000    0.658    0.624
-##     trstplt           2.848    0.035   80.855    0.000    1.911    0.876
-##   satisfaction =~                                                       
-##     stfeco            1.000                               1.659    0.712
-##     stfgov            1.039    0.013   81.895    0.000    1.724    0.752
-##     stfdem            0.979    0.012   80.257    0.000    1.624    0.733
-##     stfedu            0.780    0.012   64.344    0.000    1.294    0.576
-##     stfhlth           0.706    0.012   58.239    0.000    1.171    0.519
 ##   institutions =~                                                       
-##     trstlgl           1.000                               1.640    0.688
-##     trstplc           0.777    0.012   64.539    0.000    1.275    0.582
-##     trstun            0.861    0.013   66.949    0.000    1.411    0.605
-##     trstprl           1.123    0.013   85.721    0.000    1.842    0.812
+##     trstlgl           1.000                               1.613    0.677
+##     trstplc           0.770    0.012   61.866    0.000    1.241    0.567
+##     trstun            0.929    0.013   69.227    0.000    1.498    0.642
+##     trstep            0.908    0.013   70.929    0.000    1.464    0.660
+##     trstprl           1.139    0.014   84.084    0.000    1.837    0.809
+##   satisfaction =~                                                       
+##     stfhlth           1.000                               1.173    0.521
+##     stfedu            1.106    0.022   50.840    0.000    1.297    0.577
+##     stfeco            1.415    0.025   57.214    0.000    1.659    0.713
+##     stfgov            1.480    0.025   58.764    0.000    1.736    0.756
+##     stfdem            1.384    0.024   57.904    0.000    1.623    0.731
+##   politicians =~                                                        
+##     pltinvt           1.000                               0.646    0.613
+##     pltcare           1.021    0.016   62.862    0.000    0.660    0.628
+##     trstplt           3.012    0.039   76.838    0.000    1.946    0.891
 ##   trust =~                                                              
-##     politicians       1.000                               0.904    0.904
-##     satisfaction      2.153    0.037   57.975    0.000    0.788    0.788
-##     institutions      2.580    0.044   59.246    0.000    0.955    0.955
+##     politicians       1.000                               0.918    0.918
+##     satisfaction      1.531    0.033   46.494    0.000    0.774    0.774
+##     institutions      2.583    0.045   56.796    0.000    0.950    0.950
 ## 
 ## Variances:
 ##                    Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
-##    .pltcare           0.651    0.008   77.811    0.000    0.651    0.591
-##    .pltinvt           0.680    0.009   78.658    0.000    0.680    0.611
-##    .trstplt           1.103    0.029   38.414    0.000    1.103    0.232
-##    .stfeco            2.671    0.038   70.723    0.000    2.671    0.493
-##    .stfgov            2.287    0.035   66.172    0.000    2.287    0.435
-##    .stfdem            2.266    0.033   68.444    0.000    2.266    0.462
-##    .stfedu            3.378    0.042   79.725    0.000    3.378    0.668
-##    .stfhlth           3.721    0.045   81.846    0.000    3.721    0.731
-##    .trstlgl           2.997    0.040   74.549    0.000    2.997    0.527
-##    .trstplc           3.178    0.040   80.410    0.000    3.178    0.662
-##    .trstun            3.443    0.043   79.405    0.000    3.443    0.633
-##    .trstprl           1.746    0.030   57.842    0.000    1.746    0.340
-##    .politicians       0.082    0.004   19.635    0.000    0.182    0.182
-##    .satisfaction      1.044    0.029   35.902    0.000    0.379    0.379
-##    .institutions      0.238    0.024   10.064    0.000    0.089    0.089
-##     trust             0.368    0.010   36.776    0.000    1.000    1.000
      
-
      ## Compare fit between the first and second order models:
-fitMeasures(fit_3f)
      
+##    .trstlgl           3.068    0.041   75.262    0.000    3.068    0.541
+##    .trstplc           3.248    0.041   80.037    0.000    3.248    0.678
+##    .trstun            3.197    0.041   77.141    0.000    3.197    0.588
+##    .trstep            2.776    0.036   76.243    0.000    2.776    0.564
+##    .trstprl           1.776    0.029   61.361    0.000    1.776    0.345
+##    .stfhlth           3.695    0.046   79.989    0.000    3.695    0.729
+##    .stfedu            3.368    0.043   77.916    0.000    3.368    0.667
+##    .stfeco            2.656    0.038   69.070    0.000    2.656    0.491
+##    .stfgov            2.264    0.035   64.201    0.000    2.264    0.429
+##    .stfdem            2.289    0.034   67.172    0.000    2.289    0.465
+##    .pltinvt           0.694    0.009   78.255    0.000    0.694    0.624
+##    .pltcare           0.668    0.009   77.562    0.000    0.668    0.605
+##    .trstplt           0.978    0.028   34.461    0.000    0.978    0.205
+##    .institutions      0.255    0.022   11.691    0.000    0.098    0.098
+##    .satisfaction      0.551    0.020   27.846    0.000    0.400    0.400
+##    .politicians       0.065    0.004   17.091    0.000    0.157    0.157
+##     trust             0.352    0.010   35.005    0.000    1.000    1.000
      +
      ## Compare fit between the first and second order models:
+fitMeasures(fit_3f)
      ##                  npar                  fmin                 chisq 
-##                27.000                 0.297              9188.922 
+##                29.000                 0.360             10652.207 
 ##                    df                pvalue        baseline.chisq 
-##                51.000                 0.000             75675.049 
+##                62.000                 0.000             81699.096 
 ##           baseline.df       baseline.pvalue                   cfi 
-##                66.000                 0.000                 0.879 
+##                78.000                 0.000                 0.870 
 ##                   tli                  nnfi                   rfi 
-##                 0.844                 0.844                 0.843 
+##                 0.837                 0.837                 0.836 
 ##                   nfi                  pnfi                   ifi 
-##                 0.879                 0.679                 0.879 
+##                 0.870                 0.691                 0.870 
 ##                   rni                  logl     unrestricted.logl 
-##                 0.879           -357923.209           -353328.748 
+##                 0.870           -371404.658           -366078.555 
 ##                   aic                   bic                ntotal 
-##            715900.419            716106.840             15448.000 
+##            742867.317            743087.743             14778.000 
 ##                  bic2                 rmsea        rmsea.ci.lower 
-##            716021.036                 0.108                 0.106 
+##            742995.583                 0.108                 0.106 
 ##        rmsea.ci.upper        rmsea.ci.level          rmsea.pvalue 
-##                 0.110                 0.900                 0.000 
+##                 0.109                 0.900                 0.000 
 ##        rmsea.close.h0 rmsea.notclose.pvalue     rmsea.notclose.h0 
 ##                 0.050                 1.000                 0.080 
 ##                   rmr            rmr_nomean                  srmr 
-##                 0.245                 0.245                 0.058 
+##                 0.255                 0.255                 0.059 
 ##          srmr_bentler   srmr_bentler_nomean                  crmr 
-##                 0.058                 0.058                 0.064 
+##                 0.059                 0.059                 0.064 
 ##           crmr_nomean            srmr_mplus     srmr_mplus_nomean 
-##                 0.064                 0.058                 0.058 
+##                 0.064                 0.059                 0.059 
 ##                 cn_05                 cn_01                   gfi 
-##               116.444               131.098                 0.905 
+##               113.901               126.971                 0.897 
 ##                  agfi                  pgfi                   mfi 
-##                 0.854                 0.591                 0.744 
+##                 0.849                 0.611                 0.699 
 ##                  ecvi 
-##                 0.598
      -
      fitMeasures(fit_2nd)
      +## 0.725       +
      fitMeasures(fit_2nd)
      ##                  npar                  fmin                 chisq 
-##                27.000                 0.297              9188.922 
+##                29.000                 0.360             10652.207 
 ##                    df                pvalue        baseline.chisq 
-##                51.000                 0.000             75675.049 
+##                62.000                 0.000             81699.096 
 ##           baseline.df       baseline.pvalue                   cfi 
-##                66.000                 0.000                 0.879 
+##                78.000                 0.000                 0.870 
 ##                   tli                  nnfi                   rfi 
-##                 0.844                 0.844                 0.843 
+##                 0.837                 0.837                 0.836 
 ##                   nfi                  pnfi                   ifi 
-##                 0.879                 0.679                 0.879 
+##                 0.870                 0.691                 0.870 
 ##                   rni                  logl     unrestricted.logl 
-##                 0.879           -357923.209           -353328.748 
+##                 0.870           -371404.658           -366078.555 
 ##                   aic                   bic                ntotal 
-##            715900.419            716106.840             15448.000 
+##            742867.317            743087.743             14778.000 
 ##                  bic2                 rmsea        rmsea.ci.lower 
-##            716021.036                 0.108                 0.106 
+##            742995.583                 0.108                 0.106 
 ##        rmsea.ci.upper        rmsea.ci.level          rmsea.pvalue 
-##                 0.110                 0.900                 0.000 
+##                 0.109                 0.900                 0.000 
 ##        rmsea.close.h0 rmsea.notclose.pvalue     rmsea.notclose.h0 
 ##                 0.050                 1.000                 0.080 
 ##                   rmr            rmr_nomean                  srmr 
-##                 0.245                 0.245                 0.058 
+##                 0.255                 0.255                 0.059 
 ##          srmr_bentler   srmr_bentler_nomean                  crmr 
-##                 0.058                 0.058                 0.064 
+##                 0.059                 0.059                 0.064 
 ##           crmr_nomean            srmr_mplus     srmr_mplus_nomean 
-##                 0.064                 0.058                 0.058 
+##                 0.064                 0.059                 0.059 
 ##                 cn_05                 cn_01                   gfi 
-##               116.444               131.098                 0.905 
+##               113.901               126.971                 0.897 
 ##                  agfi                  pgfi                   mfi 
-##                 0.854                 0.591                 0.744 
+##                 0.849                 0.611                 0.699 
 ##                  ecvi 
-##                 0.598
      +## 0.725
      Click for explanation @@ -1125,7 +1129,7 @@

      5.3.8

      You should quickly notice something strange about the model fit statistics compared above. If you don’t see it, consider the following:

      -
      fitMeasures(fit_3f) - fitMeasures(fit_2nd)
      +
      fitMeasures(fit_3f) - fitMeasures(fit_2nd)
      ##                  npar                  fmin                 chisq 
 ##                     0                     0                     0 
 ##                    df                pvalue        baseline.chisq 
diff --git a/docs/at-home-exercises-5.html b/docs/at-home-exercises-5.html
index 472b2aac..990b9702 100644
--- a/docs/at-home-exercises-5.html
+++ b/docs/at-home-exercises-5.html
@@ -333,28 +333,31 @@
 
    • 4.3.5 
      • 
 
    • 4.3.6 
      • 
 
    • 4.3.7 
      • 
-
    • 4.3.8 
      • 
-
    • 4.3.9 
      • 
-
    • 4.3.10 
      • 
+
    • 4.3.8 
      • 
+
    • 4.3.9 
      • 
+
    • 4.3.10 
      • 
+
    • 4.3.11 
      • 
+
    • 4.3.12 
      •

  • 4.4 In-Class Exercises
  • 5 CFA @@ -371,30 +374,30 @@
  • 5.3 At-Home Exercises
  • 5.4 In-Class Exercises
  • 6 Full SEM @@ -411,23 +414,22 @@
  • 6.3 At-Home Exercises
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models @@ -461,43 +463,62 @@

    6.3 At-Home Exercises

    -

    This week, wrap up our re-analysis of the Kestilä (2006) results. During -this practical, you will conduct a CFA of the Trust in Politics items and -compare the results to those obtained from your previous EFA- and PCA-based -replications of Kestilä (2006).

    +

    This week, we’ll take another look at the Kestilä (2006) results. During this +practical, you will conduct an SEM to replicate the regression analysis of the +Finnish data that you conducted in the Week 4 In-Class Exercises.

    -
    +

    6.3.1

    -

    Load the ESS data.

    +

    Load the Finnish subsample of ESS data.

      -
    • The relevant data are contained in the ess_round1_trust.rds file. +
    • The relevant data are contained in the ess_finland.rds file.
        -
      • This file is in R Data Set (RDS) format.
        • -
      • The dataset is already stored as a data frame with the processing and -cleaning that you should have done for previous practicals completed.
        • -
      • Check the documentation for the readRDS() function to see how you can -load data stored in an RDS file.
        • +
      • These are the processed Finnish subsample data from the Week 4 exercises.
    +

    Note: Unless otherwise noted, all the following analyses use these data.

    -Click for explanation +Click to show code -
    ess <- readRDS("ess_round1_trust.rds")
    +
    ess_fin <- readRDS("ess_finland.rds")
    -

    Unfortunately, we need to do a little data processing before we can fit the -regression model. At the moment, lavaan will not automatically convert a factor -variable into dummy codes. So, we need to create explicit dummy codes for the -two factors we’ll use as predictors in our regression analysis: sex and -political orientation.

    +

    We need to do a little data processing before we can fit the regression model. +At the moment, lavaan will not automatically convert a factor variable into +dummy codes. So, we need to create explicit dummy codes for the two factors +we’ll use as predictors in our regression analysis: sex and political +orientation.

    -
    +

    6.3.2

    Convert the sex and political interest factors into dummy codes.

    +Click to show code + +
    library(dplyr)
+
+## Create a dummy codes by broadcasting a logical test on the factor levels:
+ess_fin <- mutate(ess_fin,
+                  female = ifelse(sex == "Female", 1, 0),
+                  hi_pol_interest = ifelse(polintr_bin == "High Interest", 1, 0)
+                  )
+
+## Check the results:
+with(ess_fin, table(dummy = female, factor = sex))
    +
    ##      factor
+## dummy Male Female
+##     0  960      0
+##     1    0   1040
    +
    with(ess_fin, table(dummy = hi_pol_interest, factor = polintr_bin))
    +
    ##      factor
+## dummy Low Interest High Interest
+##     0         1070             0
+##     1            0           929
    +
    + Click for explanation

    In R, we have several ways of converting a factor into an appropriate set of @@ -515,137 +536,128 @@

    6.3.2

    When our factor only has two levels, though, the ifelse() function is the simplest way.

    -
    library(dplyr)
-
-## Create a dummy codes by broadcasting a logical test on the factor levels:
-ess <- mutate(ess,
-              female = ifelse(sex == "Female", 1, 0),
-              hi_pol_interest = ifelse(polintr_bin == "High Interest", 1, 0)
-             )
-
-## Check the results:
-with(ess, table(dummy = female, factor = sex))
    -
    ##      factor
-## dummy Female Male
-##     0      0 8841
-##     1   9309    0
    -
    with(ess, table(dummy = hi_pol_interest, factor = polintr_bin))
    -
    ##      factor
-## dummy Low Interest High Interest
-##     0         8099             0
-##     1            0         10042
    -
    -
    -
    -

    6.3.3

    -

    Finally, subset the data to only Finnish participants.

    -
    - -Click for explanation - -
    ess_fin <- filter(ess, cntry == "Finland")
    +

    We are now ready to estimate our latent regression model. Specifically, we want -to implement the following regression as an SEM in lavaan.

    -

    \[ +to combine the three OLS regression models that you ran in 4.4.16 +into a single SEM that we will estimate in lavaan.

    +

    The following path diagram shows the intended theoretical model.

    + +

    +

    Although the variances are not included in this path diagram, all variables in +the model (including the observed predictor variables) are random.

    -
    -

    6.3.4

    -

    Define the lavaan model syntax for the regression shown above.

    +
    +

    6.3.3

    +

    Define the lavaan model syntax for the SEM shown above.

      -
    • Use the definition of the institutions factor from 5.3.2 to -define the DV.
      • +
    • Use the definition of the institutions, satsifaction, and politicians +factors from 5.3.2 to define the DVs.
      • +
    • Covary the three latent factors.
      • +
    • Covary the five predictors.
    -

    Hint: You can simply copy the line of syntax that defines the latent factor -and add a line to define the latent regression model.

    +
    + +Click to show code + +
    mod_sem <- '
+## Define the latent DVs:
+institutions =~ trstlgl + trstplc + trstun + trstep + trstprl
+satisfaction =~ stfhlth + stfedu  + stfeco + stfgov + stfdem
+politicians  =~ pltinvt + pltcare + trstplt
+
+## Specify the structural relations:
+institutions + satisfaction + politicians ~ 
+  female + age + eduyrs + hi_pol_interest + lrscale
+'
    Click for explanation -
    mod_sem <- '
-## Define the latent DV:
-institutions =~ trstlgl + trstplc + trstun
-
-## Specify the structural relations:
-institutions ~ female + age + eduyrs + hi_pol_interest + lrscale
-'
    +
      +

    • We simply need to add a line defining the latent regression paths to our old +CFA syntax.

      • +

    • We don’t need to specify the covariances in the syntax. We can use options in +the sem() function to request those estimates.

      +

    • +
    -
    -

    6.3.5

    +
    +

    6.3.4

    Estimate the SEM, and summarize the results.

    • Fit the model to the processed Finnish subsample from above.
    • Estimate the model using lavaan::sem().
      • -
    • Use the default settings in the sem() function.
    • Request the standardized parameter estimates with the summary.
      • -
    • Request the \(R^2\) estimates with the summary by supplying the rsquare = TRUE -argument to summary().
      • +
    • Request the \(R^2\) estimates with the summary.
    -Click for explanation +Click to show code -
    library(lavaan)
-
-## Fit the SEM:
-fit_sem <- sem(mod_sem, data = ess_fin)
-
-## Summarize the results:
-summary(fit_sem, fit.measures = TRUE, standardized = TRUE, rsquare = TRUE)
    -
    ## lavaan 0.6.16 ended normally after 43 iterations
+
    library(lavaan)
+
+## Fit the SEM:
+fit_sem <- sem(mod_sem, data = ess_fin, fixed.x = FALSE)
+
+## Summarize the results:
+summary(fit_sem, fit.measures = TRUE, standardized = TRUE, rsquare = TRUE)
    
+
    ## lavaan 0.6.16 ended normally after 82 iterations
 ## 
 ##   Estimator                                         ML
 ##   Optimization method                           NLMINB
-##   Number of model parameters                        11
+##   Number of model parameters                        59
 ## 
 ##                                                   Used       Total
-##   Number of observations                          1835        2000
+##   Number of observations                          1740        2000
 ## 
 ## Model Test User Model:
 ##                                                       
-##   Test statistic                               101.932
-##   Degrees of freedom                                10
+##   Test statistic                              1287.421
+##   Degrees of freedom                               112
 ##   P-value (Chi-square)                           0.000
 ## 
 ## Model Test Baseline Model:
 ## 
-##   Test statistic                              1221.668
-##   Degrees of freedom                                18
+##   Test statistic                             10534.649
+##   Degrees of freedom                               143
 ##   P-value                                        0.000
 ## 
 ## User Model versus Baseline Model:
 ## 
-##   Comparative Fit Index (CFI)                    0.924
-##   Tucker-Lewis Index (TLI)                       0.863
+##   Comparative Fit Index (CFI)                    0.887
+##   Tucker-Lewis Index (TLI)                       0.856
 ## 
 ## Loglikelihood and Information Criteria:
 ## 
-##   Loglikelihood user model (H0)             -10940.642
-##   Loglikelihood unrestricted model (H1)     -10889.676
+##   Loglikelihood user model (H0)             -57914.779
+##   Loglikelihood unrestricted model (H1)     -57271.068
 ##                                                       
-##   Akaike (AIC)                               21903.284
-##   Bayesian (BIC)                             21963.947
-##   Sample-size adjusted Bayesian (SABIC)      21929.001
+##   Akaike (AIC)                              115947.557
+##   Bayesian (BIC)                            116269.794
+##   Sample-size adjusted Bayesian (SABIC)     116082.357
 ## 
 ## Root Mean Square Error of Approximation:
 ## 
-##   RMSEA                                          0.071
-##   90 Percent confidence interval - lower         0.059
-##   90 Percent confidence interval - upper         0.084
-##   P-value H_0: RMSEA <= 0.050                    0.003
-##   P-value H_0: RMSEA >= 0.080                    0.121
+##   RMSEA                                          0.078
+##   90 Percent confidence interval - lower         0.074
+##   90 Percent confidence interval - upper         0.082
+##   P-value H_0: RMSEA <= 0.050                    0.000
+##   P-value H_0: RMSEA >= 0.080                    0.160
 ## 
 ## Standardized Root Mean Square Residual:
 ## 
-##   SRMR                                           0.030
+##   SRMR                                           0.045
 ## 
 ## Parameter Estimates:
 ## 
@@ -656,50 +668,126 @@ 
    6.3.5 
    
 ## Latent Variables:
 ##                    Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
 ##   institutions =~                                                       
-##     trstlgl           1.000                               1.783    0.835
-##     trstplc           0.615    0.035   17.401    0.000    1.097    0.640
-##     trstun            0.603    0.037   16.355    0.000    1.076    0.528
+##     trstlgl           1.000                               1.418    0.669
+##     trstplc           0.609    0.031   19.403    0.000    0.863    0.508
+##     trstun            0.887    0.038   23.484    0.000    1.257    0.626
+##     trstep            1.134    0.041   27.652    0.000    1.607    0.755
+##     trstprl           1.192    0.040   29.444    0.000    1.689    0.815
+##   satisfaction =~                                                       
+##     stfhlth           1.000                               0.979    0.497
+##     stfedu            0.602    0.043   13.872    0.000    0.589    0.416
+##     stfeco            1.266    0.067   18.848    0.000    1.240    0.681
+##     stfgov            1.639    0.079   20.638    0.000    1.605    0.846
+##     stfdem            1.521    0.075   20.180    0.000    1.489    0.793
+##   politicians =~                                                        
+##     pltinvt           1.000                               0.567    0.566
+##     pltcare           0.953    0.048   19.653    0.000    0.540    0.590
+##     trstplt           3.281    0.133   24.675    0.000    1.860    0.915
 ## 
 ## Regressions:
 ##                    Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
 ##   institutions ~                                                        
-##     female           -0.025    0.094   -0.264    0.792   -0.014   -0.007
-##     age              -0.002    0.003   -0.534    0.594   -0.001   -0.015
-##     eduyrs            0.072    0.013    5.405    0.000    0.041    0.157
-##     hi_pol_interst    0.178    0.096    1.850    0.064    0.100    0.050
-##     lrscale           0.110    0.023    4.727    0.000    0.062    0.125
+##     female            0.019    0.073    0.259    0.796    0.013    0.007
+##     age              -0.008    0.002   -3.740    0.000   -0.006   -0.105
+##     eduyrs            0.034    0.010    3.233    0.001    0.024    0.091
+##     hi_pol_interst    0.358    0.076    4.730    0.000    0.253    0.126
+##     lrscale           0.104    0.018    5.634    0.000    0.073    0.147
+##   satisfaction ~                                                        
+##     female           -0.147    0.050   -2.910    0.004   -0.150   -0.075
+##     age              -0.007    0.002   -4.598    0.000   -0.007   -0.129
+##     eduyrs            0.005    0.007    0.775    0.439    0.006    0.022
+##     hi_pol_interst    0.164    0.052    3.162    0.002    0.167    0.084
+##     lrscale           0.099    0.013    7.501    0.000    0.101    0.202
+##   politicians ~                                                         
+##     female            0.010    0.029    0.349    0.727    0.018    0.009
+##     age              -0.004    0.001   -4.490    0.000   -0.007   -0.124
+##     eduyrs            0.007    0.004    1.697    0.090    0.012    0.047
+##     hi_pol_interst    0.258    0.031    8.364    0.000    0.455    0.227
+##     lrscale           0.039    0.007    5.370    0.000    0.068    0.138
+## 
+## Covariances:
+##                      Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
+##  .institutions ~~                                                         
+##    .satisfaction        1.030    0.069   14.933    0.000    0.796    0.796
+##    .politicians         0.675    0.041   16.628    0.000    0.908    0.908
+##  .satisfaction ~~                                                         
+##    .politicians         0.365    0.027   13.544    0.000    0.713    0.713
+##   female ~~                                                               
+##     age                 0.071    0.212    0.335    0.738    0.071    0.008
+##     eduyrs              0.179    0.046    3.869    0.000    0.179    0.093
+##     hi_pol_interst     -0.017    0.006   -2.767    0.006   -0.017   -0.066
+##     lrscale            -0.032    0.024   -1.316    0.188   -0.032   -0.032
+##   age ~~                                                                  
+##     eduyrs            -22.750    1.722  -13.212    0.000  -22.750   -0.334
+##     hi_pol_interst      1.377    0.215    6.413    0.000    1.377    0.156
+##     lrscale             1.774    0.853    2.079    0.038    1.774    0.050
+##   eduyrs ~~                                                               
+##     hi_pol_interst      0.270    0.047    5.787    0.000    0.270    0.140
+##     lrscale             0.735    0.186    3.946    0.000    0.735    0.095
+##   hi_pol_interest ~~                                                      
+##     lrscale             0.016    0.024    0.672    0.501    0.016    0.016
 ## 
 ## Variances:
 ##                    Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
-##    .trstlgl           1.379    0.168    8.199    0.000    1.379    0.302
-##    .trstplc           1.739    0.084   20.599    0.000    1.739    0.591
-##    .trstun            3.000    0.116   25.792    0.000    3.000    0.722
-##    .institutions      3.019    0.209   14.451    0.000    0.949    0.949
+##    .trstlgl           2.477    0.093   26.743    0.000    2.477    0.552
+##    .trstplc           2.140    0.076   28.334    0.000    2.140    0.742
+##    .trstun            2.453    0.090   27.322    0.000    2.453    0.608
+##    .trstep            1.950    0.078   24.906    0.000    1.950    0.430
+##    .trstprl           1.443    0.064   22.437    0.000    1.443    0.336
+##    .stfhlth           2.922    0.104   28.103    0.000    2.922    0.753
+##    .stfedu            1.663    0.058   28.613    0.000    1.663    0.827
+##    .stfeco            1.775    0.069   25.755    0.000    1.775    0.536
+##    .stfgov            1.020    0.056   18.371    0.000    1.020    0.284
+##    .stfdem            1.307    0.060   21.953    0.000    1.307    0.371
+##    .pltinvt           0.682    0.024   27.818    0.000    0.682    0.680
+##    .pltcare           0.547    0.020   27.582    0.000    0.547    0.652
+##    .trstplt           0.672    0.069    9.676    0.000    0.672    0.163
+##    .institutions      1.881    0.125   15.077    0.000    0.936    0.936
+##    .satisfaction      0.892    0.086   10.386    0.000    0.930    0.930
+##    .politicians       0.294    0.024   12.224    0.000    0.914    0.914
+##     female            0.250    0.008   29.496    0.000    0.250    1.000
+##     age             313.238   10.620   29.496    0.000  313.238    1.000
+##     eduyrs           14.818    0.502   29.496    0.000   14.818    1.000
+##     hi_pol_interst    0.250    0.008   29.496    0.000    0.250    1.000
+##     lrscale           4.034    0.137   29.496    0.000    4.034    1.000
 ## 
 ## R-Square:
 ##                    Estimate
-##     trstlgl           0.698
-##     trstplc           0.409
-##     trstun            0.278
-##     institutions      0.051
    
+##     trstlgl           0.448
+##     trstplc           0.258
+##     trstun            0.392
+##     trstep            0.570
+##     trstprl           0.664
+##     stfhlth           0.247
+##     stfedu            0.173
+##     stfeco            0.464
+##     stfgov            0.716
+##     stfdem            0.629
+##     pltinvt           0.320
+##     pltcare           0.348
+##     trstplt           0.837
+##     institutions      0.064
+##     satisfaction      0.070
+##     politicians       0.086
    +
    + +Click for explanation + +

    The fixed.x = FALSE argument tells lavaan to model the predictors as random +variables. By default, lavaan will covary any random predictor variables. So, +we don’t need to make any other changes to the usual procedure.

    +
    -
    -

    6.3.6

    +
    +

    6.3.5

    Finally, we will rerun the latent regression model from above as a path model -with the factor score for Trust in Institutions as the DV.

    +with the factor scores from 4.4.10 acting as the DVs.

      -
    • Add the Trust in Institutions factor score that you estimated in -4.4.10 to the ESS data. -
        -
      • If you did not save the factors scores last week, you’ll need to rerun the -relevant EFA.
        • -
      • Don’t forget to subset the data to the Finish participants before fitting -the path model.
        • -
      • -
    • Rerun the above SEM with the EFA-derived Trust in Institutions factor score -taking the place of the analagous latent variable as DV. +
    • Rerun the above SEM as a path model wherein the EFA-derived Trust in +Institutions, Satisfaction with Political Systems, and Trust in Politicians +factor scores act as the DVs.
      • Request the standardized parameter estimates with the summary.
      • Request the \(R^2\) estimates with the summary.
        • @@ -707,66 +795,23 @@

        6.3.6

      -Click for explanation +Click to show code -

      First, we’ll quickly reproduce the Trust in Institutions factor score that we -estimated last week.

      -
        -
      • Note that psych::fa() returns the factor scores in a different order than it -did for the Week 2 analyses.
        • -
      -
      ## Load the psych library:
-library(psych)
-
-## Rerun the three-factor EFA from last week:
-fit_efa <- fa(ess[7:19], 
-              nfactors = 3,          
-              rotate   = "promax",   
-              scores   = "Bartlett")
-
-## View the factor loadings:
-print(fit_efa$loadings, cut = 0.3)
      -
      ## 
-## Loadings:
-##         MR3    MR1    MR2   
-## pltcare         0.815       
-## pltinvt         0.806       
-## trstprl  0.405  0.381       
-## trstlgl  0.843              
-## trstplc  0.776              
-## trstplt         0.571       
-## trstep   0.503              
-## trstun   0.543              
-## stfeco                 0.703
-## stfgov                 0.594
-## stfdem                 0.462
-## stfedu                 0.701
-## stfhlth                0.664
-## 
-##                  MR3   MR1   MR2
-## SS loadings    2.178 2.101 2.023
-## Proportion Var 0.168 0.162 0.156
-## Cumulative Var 0.168 0.329 0.485
      -
      ## Reproduce the factor score from last week:
-ess$trust_inst_efa <- fit_efa$scores[ , 1]
-
-## Subset the data again:
-ess_fin <- ess[ess$cntry == "Finland", ]
      -

      Now, we’ll rerun our regression as a path analysis with the EFA-derived factor -score as DV.

      -
      ## Define the model syntax for the path analysis:
-mod_pa <- 'trust_inst_efa ~ female + age + eduyrs + hi_pol_interest + lrscale'
-
-## Estimate the path model:
-fit_pa <- sem(mod_pa, data = ess_fin)
-
-## Summarize the results:
-summary(fit_pa, fit.measures = TRUE, standardized = TRUE)
      -
      ## lavaan 0.6.16 ended normally after 1 iteration
+
      ## Define the model syntax for the path analysis:
+mod_pa <- '
+trust_inst + satisfy + trust_pol ~ 
+  female + age + eduyrs + hi_pol_interest + lrscale'
+
+## Estimate the path model:
+fit_pa <- sem(mod_pa, data = ess_fin, fixed.x = FALSE)
+
+## Summarize the results:
+summary(fit_pa, standardized = TRUE, rsquare = TRUE)
      
+
      ## lavaan 0.6.16 ended normally after 44 iterations
 ## 
 ##   Estimator                                         ML
 ##   Optimization method                           NLMINB
-##   Number of model parameters                         6
+##   Number of model parameters                        36
 ## 
 ##                                                   Used       Total
 ##   Number of observations                          1740        2000
@@ -776,38 +821,6 @@ 
      6.3.6 
      
 ##   Test statistic                                 0.000
 ##   Degrees of freedom                                 0
 ## 
-## Model Test Baseline Model:
-## 
-##   Test statistic                                71.266
-##   Degrees of freedom                                 5
-##   P-value                                        0.000
-## 
-## User Model versus Baseline Model:
-## 
-##   Comparative Fit Index (CFI)                    1.000
-##   Tucker-Lewis Index (TLI)                       1.000
-## 
-## Loglikelihood and Information Criteria:
-## 
-##   Loglikelihood user model (H0)              -2352.608
-##   Loglikelihood unrestricted model (H1)      -2352.608
-##                                                       
-##   Akaike (AIC)                                4717.216
-##   Bayesian (BIC)                              4749.985
-##   Sample-size adjusted Bayesian (SABIC)       4730.924
-## 
-## Root Mean Square Error of Approximation:
-## 
-##   RMSEA                                          0.000
-##   90 Percent confidence interval - lower         0.000
-##   90 Percent confidence interval - upper         0.000
-##   P-value H_0: RMSEA <= 0.050                       NA
-##   P-value H_0: RMSEA >= 0.080                       NA
-## 
-## Standardized Root Mean Square Residual:
-## 
-##   SRMR                                           0.000
-## 
 ## Parameter Estimates:
 ## 
 ##   Standard errors                             Standard
@@ -816,25 +829,75 @@ 
      6.3.6 
      
 ## 
 ## Regressions:
 ##                    Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
-##   trust_inst_efa ~                                                      
-##     female            0.002    0.045    0.036    0.971    0.002    0.001
-##     age              -0.003    0.001   -1.977    0.048   -0.003   -0.051
-##     eduyrs            0.024    0.006    3.820    0.000    0.024    0.099
-##     hi_pol_interst    0.154    0.047    3.313    0.001    0.154    0.081
-##     lrscale           0.058    0.011    5.159    0.000    0.058    0.122
+##   trust_inst ~                                                          
+##     female            0.004    0.045    0.091    0.928    0.004    0.002
+##     age              -0.003    0.001   -2.229    0.026   -0.003   -0.057
+##     eduyrs            0.023    0.006    3.642    0.000    0.023    0.094
+##     hi_pol_interst    0.167    0.046    3.599    0.000    0.167    0.088
+##     lrscale           0.059    0.011    5.258    0.000    0.059    0.125
+##   satisfy ~                                                             
+##     female           -0.125    0.040   -3.115    0.002   -0.125   -0.073
+##     age              -0.005    0.001   -4.102    0.000   -0.005   -0.105
+##     eduyrs           -0.003    0.006   -0.534    0.594   -0.003   -0.014
+##     hi_pol_interst    0.073    0.041    1.782    0.075    0.073    0.043
+##     lrscale           0.085    0.010    8.510    0.000    0.085    0.200
+##   trust_pol ~                                                           
+##     female            0.016    0.046    0.338    0.735    0.016    0.008
+##     age              -0.009    0.001   -6.480    0.000   -0.009   -0.161
+##     eduyrs            0.018    0.007    2.839    0.005    0.018    0.071
+##     hi_pol_interst    0.464    0.047    9.801    0.000    0.464    0.232
+##     lrscale           0.055    0.011    4.801    0.000    0.055    0.110
+## 
+## Covariances:
+##                      Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
+##  .trust_inst ~~                                                           
+##    .satisfy             0.437    0.021   20.609    0.000    0.437    0.568
+##    .trust_pol           0.498    0.024   20.480    0.000    0.498    0.564
+##  .satisfy ~~                                                              
+##    .trust_pol           0.367    0.021   17.664    0.000    0.367    0.467
+##   female ~~                                                               
+##     age                 0.071    0.212    0.335    0.738    0.071    0.008
+##     eduyrs              0.179    0.046    3.869    0.000    0.179    0.093
+##     hi_pol_interst     -0.017    0.006   -2.767    0.006   -0.017   -0.066
+##     lrscale            -0.032    0.024   -1.316    0.188   -0.032   -0.032
+##   age ~~                                                                  
+##     eduyrs            -22.750    1.722  -13.212    0.000  -22.750   -0.334
+##     hi_pol_interst      1.377    0.215    6.413    0.000    1.377    0.156
+##     lrscale             1.774    0.853    2.079    0.038    1.774    0.050
+##   eduyrs ~~                                                               
+##     hi_pol_interst      0.270    0.047    5.787    0.000    0.270    0.140
+##     lrscale             0.735    0.186    3.946    0.000    0.735    0.095
+##   hi_pol_interest ~~                                                      
+##     lrscale             0.016    0.024    0.672    0.501    0.016    0.016
 ## 
 ## Variances:
 ##                    Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
-##    .trust_inst_efa    0.875    0.030   29.496    0.000    0.875    0.960
      
-
      ## Extract R^2:
-inspect(fit_pa, "r2")
      
-
      ## trust_inst_efa 
-##           0.04
      
+##    .trust_inst        0.866    0.029   29.496    0.000    0.866    0.958
+##    .satisfy           0.684    0.023   29.496    0.000    0.684    0.945
+##    .trust_pol         0.902    0.031   29.496    0.000    0.902    0.902
+##     female            0.250    0.008   29.496    0.000    0.250    1.000
+##     age             313.238   10.620   29.496    0.000  313.238    1.000
+##     eduyrs           14.818    0.502   29.496    0.000   14.818    1.000
+##     hi_pol_interst    0.250    0.008   29.496    0.000    0.250    1.000
+##     lrscale           4.034    0.137   29.496    0.000    4.034    1.000
+## 
+## R-Square:
+##                    Estimate
+##     trust_inst        0.042
+##     satisfy           0.055
+##     trust_pol         0.098
      +
      + +Click to show explanation + +

      We don’t so anything particularly special here. We simply rerun our latent +regression as a path analysis with the EFA-derived factor scores as the DVs.

      +
    -
    -

    6.3.7

    +
    +

    6.3.6

    Compare the results from the path analysis to the SEM-based results.

    • Does it matter whether we use a latent variable or a factor score to define @@ -844,54 +907,97 @@

      6.3.7

      (i.e., the values in the column labeled Std.all).

      -Click for explanation +Click to show code -

      First, we’ll source a script that defines a bunch of convenience functions. One -of these functions, partSummary(), will allow us to print only the interesting -pieces of the model summary.

      -
      ## Source a script of convenience function definitions:
-source("supportFunctions.R")
      -

      Now, we’ll compare the results. Specifically, we’re interested in differences in -the regression coefficients and the \(R^2\).

      -
      ## View the regression estimates from the SEM:
-partSummary(fit_sem, 8, standardized = TRUE)
      +

      Note: The “supportFunction.R” script that we source below isn’t a necessary +part of the solution. This script defines a bunch of convenience functions. One +of these functions, partSummary(), allows us to print selected pieces of the +model summary.

      +
      ## Source a script of convenience function definitions:
+source("supportFunctions.R")
      +
      ## View the regression estimates from the SEM:
+partSummary(fit_sem, 8, standardized = TRUE)
      ## Regressions:
 ##                    Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
 ##   institutions ~                                                        
-##     female           -0.025    0.094   -0.264    0.792   -0.014   -0.007
-##     age              -0.002    0.003   -0.534    0.594   -0.001   -0.015
-##     eduyrs            0.072    0.013    5.405    0.000    0.041    0.157
-##     hi_pol_interst    0.178    0.096    1.850    0.064    0.100    0.050
-##     lrscale           0.110    0.023    4.727    0.000    0.062    0.125
      -
      ## View the regression estimates from the path analysis:
-partSummary(fit_pa, 7, standardized = TRUE)
      +## female 0.019 0.073 0.259 0.796 0.013 0.007 +## age -0.008 0.002 -3.740 0.000 -0.006 -0.105 +## eduyrs 0.034 0.010 3.233 0.001 0.024 0.091 +## hi_pol_interst 0.358 0.076 4.730 0.000 0.253 0.126 +## lrscale 0.104 0.018 5.634 0.000 0.073 0.147 +## satisfaction ~ +## female -0.147 0.050 -2.910 0.004 -0.150 -0.075 +## age -0.007 0.002 -4.598 0.000 -0.007 -0.129 +## eduyrs 0.005 0.007 0.775 0.439 0.006 0.022 +## hi_pol_interst 0.164 0.052 3.162 0.002 0.167 0.084 +## lrscale 0.099 0.013 7.501 0.000 0.101 0.202 +## politicians ~ +## female 0.010 0.029 0.349 0.727 0.018 0.009 +## age -0.004 0.001 -4.490 0.000 -0.007 -0.124 +## eduyrs 0.007 0.004 1.697 0.090 0.012 0.047 +## hi_pol_interst 0.258 0.031 8.364 0.000 0.455 0.227 +## lrscale 0.039 0.007 5.370 0.000 0.068 0.138       +
      ## View the regression estimates from the path analysis:
+partSummary(fit_pa, 7, standardized = TRUE)
      ## Regressions:
 ##                    Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
-##   trust_inst_efa ~                                                      
-##     female            0.002    0.045    0.036    0.971    0.002    0.001
-##     age              -0.003    0.001   -1.977    0.048   -0.003   -0.051
-##     eduyrs            0.024    0.006    3.820    0.000    0.024    0.099
-##     hi_pol_interst    0.154    0.047    3.313    0.001    0.154    0.081
-##     lrscale           0.058    0.011    5.159    0.000    0.058    0.122
      -
      ## View the R-squared estimates from the SEM:
-partSummary(fit_sem, 10, rsquare = TRUE)
      +## trust_inst ~ +## female 0.004 0.045 0.091 0.928 0.004 0.002 +## age -0.003 0.001 -2.229 0.026 -0.003 -0.057 +## eduyrs 0.023 0.006 3.642 0.000 0.023 0.094 +## hi_pol_interst 0.167 0.046 3.599 0.000 0.167 0.088 +## lrscale 0.059 0.011 5.258 0.000 0.059 0.125 +## satisfy ~ +## female -0.125 0.040 -3.115 0.002 -0.125 -0.073 +## age -0.005 0.001 -4.102 0.000 -0.005 -0.105 +## eduyrs -0.003 0.006 -0.534 0.594 -0.003 -0.014 +## hi_pol_interst 0.073 0.041 1.782 0.075 0.073 0.043 +## lrscale 0.085 0.010 8.510 0.000 0.085 0.200 +## trust_pol ~ +## female 0.016 0.046 0.338 0.735 0.016 0.008 +## age -0.009 0.001 -6.480 0.000 -0.009 -0.161 +## eduyrs 0.018 0.007 2.839 0.005 0.018 0.071 +## hi_pol_interst 0.464 0.047 9.801 0.000 0.464 0.232 +## lrscale 0.055 0.011 4.801 0.000 0.055 0.110 +
      ## View the R-squared estimates from the SEM:
+partSummary(fit_sem, 11, rsquare = TRUE)
      +
      ## R-Square:
+##                    Estimate
+##     trstlgl           0.448
+##     trstplc           0.258
+##     trstun            0.392
+##     trstep            0.570
+##     trstprl           0.664
+##     stfhlth           0.247
+##     stfedu            0.173
+##     stfeco            0.464
+##     stfgov            0.716
+##     stfdem            0.629
+##     pltinvt           0.320
+##     pltcare           0.348
+##     trstplt           0.837
+##     institutions      0.064
+##     satisfaction      0.070
+##     politicians       0.086
      +
      ## View the R-squared estimates from the SEM:
+partSummary(fit_pa, 10, rsquare = TRUE)
      ## R-Square:
 ##                    Estimate
-##     trstlgl           0.698
-##     trstplc           0.409
-##     trstun            0.278
-##     institutions      0.051
      -
      ## View the R-squared estimate from the path analysis:
-#partSummary(fit_pa, 9, rsquare = TRUE)
-inspect(fit_pa, "r2")
      -
      ## trust_inst_efa 
-##           0.04
      +## trust_inst 0.042 +## satisfy 0.055 +## trust_pol 0.098 +
      + +Click for explanation +

      It certainly looks like the way we define the DV has a meaningful impact. The patterns of significance differ between the two sets of regression slopes, and -the \(R^2\) is 26.7% larger in the SEM.

      +the \(R^2\) values are larger for the Institutions and Satisfaction factors in +the SEM, and the \(R^2\) for the Politicians factor is higher in the path analysis.

      +
      -

      End of In-Class Exercises

      +

      End of At-Home Exercises

    diff --git a/docs/at-home-exercises-6.html b/docs/at-home-exercises-6.html index f4e30a03..4e3bfb87 100644 --- a/docs/at-home-exercises-6.html +++ b/docs/at-home-exercises-6.html @@ -333,28 +333,31 @@
  • 4.3.5
  • 4.3.6
  • 4.3.7
    • -
  • 4.3.8
    • -
  • 4.3.9
    • -
  • 4.3.10
    • +
  • 4.3.8
    • +
  • 4.3.9
    • +
  • 4.3.10
    • +
  • 4.3.11
    • +
  • 4.3.12
  • 4.4 In-Class Exercises
  • 5 CFA @@ -371,30 +374,30 @@
  • 5.3 At-Home Exercises
  • 5.4 In-Class Exercises
  • 6 Full SEM @@ -411,23 +414,22 @@
  • 6.3 At-Home Exercises
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/at-home-exercises.html b/docs/at-home-exercises.html index fa128d06..e053083f 100644 --- a/docs/at-home-exercises.html +++ b/docs/at-home-exercises.html @@ -333,28 +333,31 @@
  • 4.3.5
  • 4.3.6
  • 4.3.7
    • -
  • 4.3.8
    • -
  • 4.3.9
    • -
  • 4.3.10
    • +
  • 4.3.8
    • +
  • 4.3.9
    • +
  • 4.3.10
    • +
  • 4.3.11
    • +
  • 4.3.12
  • 4.4 In-Class Exercises
  • 5 CFA @@ -371,30 +374,30 @@
  • 5.3 At-Home Exercises
  • 5.4 In-Class Exercises
  • 6 Full SEM @@ -411,23 +414,22 @@
  • 6.3 At-Home Exercises
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/attendance.html b/docs/attendance.html index f91f356c..e8dc6d74 100644 --- a/docs/attendance.html +++ b/docs/attendance.html @@ -333,28 +333,31 @@
  • 4.3.5
  • 4.3.6
  • 4.3.7
    • -
  • 4.3.8
    • -
  • 4.3.9
    • -
  • 4.3.10
    • +
  • 4.3.8
    • +
  • 4.3.9
    • +
  • 4.3.10
    • +
  • 4.3.11
    • +
  • 4.3.12
  • 4.4 In-Class Exercises
  • 5 CFA @@ -371,30 +374,30 @@
  • 5.3 At-Home Exercises
  • 5.4 In-Class Exercises
  • 6 Full SEM @@ -411,23 +414,22 @@
  • 6.3 At-Home Exercises
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/cfa.html b/docs/cfa.html index e350ca20..2d5b4aed 100644 --- a/docs/cfa.html +++ b/docs/cfa.html @@ -333,28 +333,31 @@
  • 4.3.5
  • 4.3.6
  • 4.3.7
    • -
  • 4.3.8
    • -
  • 4.3.9
    • -
  • 4.3.10
    • +
  • 4.3.8
    • +
  • 4.3.9
    • +
  • 4.3.10
    • +
  • 4.3.11
    • +
  • 4.3.12
  • 4.4 In-Class Exercises
  • 5 CFA @@ -371,30 +374,30 @@
  • 5.3 At-Home Exercises
  • 5.4 In-Class Exercises
  • 6 Full SEM @@ -411,23 +414,22 @@
  • 6.3 At-Home Exercises
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/course-data.html b/docs/course-data.html index 12d8e9d3..68e0f9ed 100644 --- a/docs/course-data.html +++ b/docs/course-data.html @@ -333,28 +333,31 @@
  • 4.3.5
  • 4.3.6
  • 4.3.7
    • -
  • 4.3.8
    • -
  • 4.3.9
    • -
  • 4.3.10
    • +
  • 4.3.8
    • +
  • 4.3.9
    • +
  • 4.3.10
    • +
  • 4.3.11
    • +
  • 4.3.12
  • 4.4 In-Class Exercises
  • 5 CFA @@ -371,30 +374,30 @@
  • 5.3 At-Home Exercises
  • 5.4 In-Class Exercises
  • 6 Full SEM @@ -411,23 +414,22 @@
  • 6.3 At-Home Exercises
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/course-overview.html b/docs/course-overview.html index c57359bd..120cfeed 100644 --- a/docs/course-overview.html +++ b/docs/course-overview.html @@ -333,28 +333,31 @@
  • 4.3.5
  • 4.3.6
  • 4.3.7
    • -
  • 4.3.8
    • -
  • 4.3.9
    • -
  • 4.3.10
    • +
  • 4.3.8
    • +
  • 4.3.9
    • +
  • 4.3.10
    • +
  • 4.3.11
    • +
  • 4.3.12
  • 4.4 In-Class Exercises
  • 5 CFA @@ -371,30 +374,30 @@
  • 5.3 At-Home Exercises
  • 5.4 In-Class Exercises
  • 6 Full SEM @@ -411,23 +414,22 @@
  • 6.3 At-Home Exercises
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/data/w4_efa_attitudes.rds b/docs/data/w4_efa_attitudes.rds index ae10f08d4540c4d6433ca93afd0f68eb600b55ab..40480505780d3d2f91c8e3569405e91db583cab5 100644 GIT binary patch literal 597328 zcmZUaWn5d&zV3@li@RHKD3szBoC3wIcuR4LYmo%EBBfZ--~@NK0)+rA?(Xg$ASAc{ zz0Wy!pL_FSy~<~1&HSEuX07Kti{UHwvw!{2jyrv{NSEebNd_@T;c{^?a}B)+dHH&s znuw5;h_pY9(a2rk88PYW;2?Ba8HGArVnv0I1Y&zxwh>tcFNvf_$I=h3=ub=i%U1!X zPi5xi9P@PE)#gvfwo;B~GbbrCqO|cS%^|_~&>Hsx&E;XWGWU@0YD>begnqe-K6t4C z$h96>0&70c(T24>o>cJmgcdY@Ot}al@dYO9P-`;gMnUfws&lO5t3=5IXN4u}i12pA zLVV;X*mjM<^u0{FQi*An+^*|k@0ocG#F)6}61pp+G}b%JN~;2tvE6H}n$&$ml<^Fm+_pK8hae0&epnV=3@}eKhufiE1cX@-glX@(6->9L7LO`IJ zW5JT;F1H)l-|j*OWlER`^y?0j|E9k;&vay>@d0r?3g`Sp+h*IT%v9lE53o(6+gtzM zIY~@>v4oxLrzLPOij7a&%o+9B*{H4=aje~6t+5}MW{WkHjn~+C4u76adW}5fdxj`$ zYuNd@rIzEliMpf?x7ad(eQL5fETPpEQb#1_>4EE)8Ak&V2%rhVb!`u?hO>a^H z!Hl>8QO{}f<)?)-QCG6JS{x4@*Ctj!3pkWXIw?IWl+RiwI(z=y?U@Solp3ENo)D{V zZ;3Rq0m5lk=}w~kO!m;3oQfmgOfxX5G37+KftX?J9x)8bDUsjWwVfpNFb7F7nk*3o#g{lAku)K_x%~NEzccGLwjJ11Q1+REx{Q^thquNQKMk_9U zu2ohhm6qGu6qX|S?Jh|O&muV|rqDza2MQl7xt$9u7YomqJyWi@)iwqWmpmXMz@PanRdOeyoqcU}r{kSW zbsZ0%OQV5Ou%U(%A$D2rI)idg0{;jSbPSch!Oz<^n+)AoS2DU)M?;m=j;XacGD9g& zx<$DGyA9qZkEIQ-EDx$L*2*d2JEXvr>H@;g=^*!!*BG&Lt4j5GXoJI2j5N`vrFH*4V zkJh_B|D7asFM+*3P;RJX*yeivcIZldnMLx4M6J~H*B+M2F%c>N#n!jG{{BlNHQ~ZX z^Q=8m0(>#LnU6z$#nGk|eKx#rS|{ER+&7I`GMz7fJu!bR8XTw6G-6Z9N>3$zI7u9e zcK4OeX^(+Wn{g#FK-pjN^64q|YT^O6=EL%UMb2gz{9V6Zp@g9JxIYe-7?v|Ga-DN4 zEi(kfi~-QlL+YMH+N4OuCn?4f`mwEN1}b3uFrgFP18T_3=0xiIM2qK15Zp5h)U{4z z<|cb={Vpye7J(dz}1Hx>cdScz__MWwcMEdC+2>@3w~~ zcPeO-MC>Q|T&P+A4PoLB9TN|RoaVd!Glt&a73wE4NddL7jTuoFBUeevR`30?&o6tI zDow)vd{Nd$M?_5JL^JxHN_q@h7*cjjdpj0Z94N-ilE8wC34Pi8o1|5}Nl^)m0&y=s zhNNvlGjEEc1rf+T@G>caNo#;_g8<G{jOtFJj@n(Lmbf zPp-RA&nk5O?nP!ge;GDCv~{|U8ie+QZldxb+LzyUpS@<98^<}W;Cf#dg}$u~ zF?Xu1MjcY$ixqvvV^$px)YN7Y@zZw&lpZ#s)S1l%{>VvtRt-Fs$eWjc_T=)b^_t27 z&I*sVfMSDS7MORDfV9mA2E;~j+!_EjSGchUKp;aNKLObB#cj^)JO$RmzAbpDPCmyw z-`%;}bWhWE&B7&m=`Nna>nQ?I!rN3BzAT>(2B{?on8&Gs3KY~q)UaMb$laH?=?`Yiy=qfQupMpw$ zS+J|ZxpL3D`NIL+9B_HM?hI1+bKP_K!UwP=2;&=z;L8eFJ)@t0uxb8u;Pu7)ecu6t zly+3hnIaGkKZ>K~E=jmTrmfwz+%zr~wztno0~f=0O-~o6YtyIv(rK9-|8wJK{rVu? zQjne%yPK7g+?NBf^?E33W*Uq`M2!0&n53n1_k&ndSQ+2}8ePo5adD>!cPPH>W{U$L z&l5nRl7r#$U3~BgR{@0>AdA)wfj$5p-CZxL3eUd;1owh7Krq{HZ|VMuYU=z7ie%Dz z3?6J<^gFD07C@W};_c&)ky}b?R#6XQym~iAjOmSv{u+b63NRXxlOZbHXuSUI>MDe$ z*v5DJj!ndIq6=m9x42Se9Aj?UmKk1foyPu`-8R$8Al7W*PW{Ny7?~Nc-qWoz@>hh` z4H_##XDf{@9C<;CC_LAsw}$U(Uufp8+w#pI^&azHd(I^MDiy|->NGdG>U#NI6Ec@z zK|SVE)*!qa$QU$%aXMHrqA_^=))uIdK_V&@{o66VGKk$e9DlCjVsKh*7dHHY)*VM} zSI<}$-Ng&ZH)NoY0Ynr4vN;~`G^YTdYqT6Bb*)Hy+55YM*6vfNNFL`#nNtSgXpx1Gt{bgX z9>tL4Vl0})+HKm#=Nz@I^Wa}v%XnJRN4D$D{ZA7}*=5l674a%!U(-4U*oT||H2TH> ze{P zy@{VfoZn7Ql^p61=iUn&xy8)kRPZb)f~nSJ%vHTS0{QVh0Y{Mt>^>%u9NA8E6k~+k z43hNe&yzApy62ik5vg_42ZsHIDt)y8!PcUC!Na781vRQ}1zf~yI2?*hb0vKg%KPew z)>oeIweOg1YF{oQ5hk=}vEFKo(nTLmO{0o^zU$Um@-Yzw_E`PiwvRK~QxBEe1^hF# z)+12vP6v3$j&y%W>gGA<2RvBZCjiNRK23oATS_0|%0PAx8=CNb-7Sg~Bqw}GeFF>w zY|YX3AsOKO+VC@#`fh?douB-949j$y10Lpk5hXJ?k51d0J%EGQR9G$5xhF#zAe%XX zSsv@p(`*W3bq!;?yoDI#o`n_3Jbp=0hA26-jX}T**5q(VpZb(02w=#+jzAkv+^xh3 zA5-7bMZ7Clsu+r)(k#W1gUS$n~l=G?Va+700Xe#P=Ri6=y?} zmHT0LUoaUo4~1mT>$x#vd}&!hd&Rrm$#Ct+?;xDNoqfqbAPd0#b`MEqYlwXY^$Z}4 z1atw~oTH8JMPdjd;C+DQhKolm`$wis;~vPA^zk@z-?$~UwjX9FE`pk^E+(6;$UoqY=Xs+5 ztxF#j2ZY)XLoNJB6R~-(jPxJVx_P4y(z(uLS40y2)db+;4(~z}ZX8{3-aX2Ic+sy7 z!&0W|CewrKo=?eL0CFq=M1JQ5Cjng7=LaU>r5>{7>3fDnIF2ak29EeQR}S1-$kYI$;>))nE9RPd8!{|E6qgtr-*nbRAO3j0bYyW()6#@IA(WMwJRBE!eCF%G#}KMQ*8UTSZD zDUwE76)DAsRhj+WRv08 zr0JuJ`cu1KG04&3Ae|DKaD_Y*2Yo!SU9l)Ys~(FhfIm)Yd}$DSBJT>Dt%5)4S$|r2 z*@y>h*6``Jp)6=}o9_lZuI0Jwv^)N) zjtSEM>$nSmykmFOo!DvLnFiGg$joI2JmkA?uIwK+mNAs6|cQLq0HI^zq}ny==zj0>W4as_&wbcps*VcODQB z+jCt&SQ?+-ivc|n&}~@qPrGd=fS9AkkLhz-3EeI{C84P~tN>~3t=;>KjS*V@2V+aK zd3*C8Qi&qNbZMcs4Of&pVpE~Eg2eR~kiPGpnW3hNKC(?SZ?Ktd&D zy&vf2T@e>8dHKW4RUrNuRW9|XG3@(pNjZ)v-Q$uxx1Yu&hxg~C=mJ_x@dY5fApN}Q zHBb&mAM*~AfL5EW@RY#xOjB2&VOkxE^5fGTlhK9D-bz|B1@bG?08He^_|eJ(mg_zb za#a*I_ROx#M{Q~G&*b45)(QeD8=Gn~+g#zr={?nyVUm%)+=d+cRM{}XV=q%=ru2|n z)L2{9cmrxJTA{LG=60C3b5iRCeT+-u;d1i>@YRpQC0nNWXD0=+HuuQB<%%Z`odNK0 zOzAV~%(uRs*9ot_G9<@TLp&Hh5P+(4;|rhfy&BFb6pkG^Jxerz;7!I?DbN%36^B8^}l%WI%+3G;#VWkC7Kf; zXw`Lur_i@Y$gz8?VFW<~LgkcqtGzWU`G{Nhu_UWIrmj2N56-YR!Mn+i()j&S{-$4* z(s+EqiGKhc?d8Gk=B46(f|Oq0`|lWm&WZkoWmdT*pW)}!!D@d6%6{lmwNmt4-rc0C z3|tn1z;PN)oLe1d;BAtw<->VM7;9Nua*(;&8Qio$k`oTyMk3NzaqN($Yc~dHSR;y# zJ{b>>1%5;GF17~+oR3h>rm4L6eVxpzc1g$$c6Q^zEWcfxTg+hwXF-#hzsn76XxDnZtEax>+>*)bAT&My7s(!GUWoUepGaQ&e8fNaSXJm5J&h<$!Z_-tt zSww(ZhpwY%8E4jlKAxglEY4KFB@)`T9eiI#G4{h+eVj$IYT4LP0&O0eo(Z^2d{c?Y56Q zqSi;Ur2m`$8V#bhaE_y+;e~TW=L^kEHL63Gt_YqLDwdt^^^k4pzW%6sN2al>p#=QIc2UVun*z zzaXu*9}#FRDzM9YJdxDI3qH>%{L;FMqFUv^sZNR?O2Gv0#6jiVm$$bZHDgnI`!c@Y zy_D}l&kTB(lBt;YdKl3qmrC3+1QowyH4UDt;$kqY>c-U;ZdQv^eqhSKj#%+GlG@H$ zB=R~M$4K_NSV-fL*GOhWYjes(``Q1?LaXryjugOuQb|d{OZBz^QCS@7LYl3-@jyzxn1QJzUwh@{?HN26tETvmlQQN z#pu`ik@R%I!ehcT(fh`-5-hDs`BZ$a8bGacO?;hXL9fJFU48KRGjFhKmpd=G*=Jmr z0@T`8CAlvN(m!ipz!#^|Yb;_HLm~P)?XPp&^P62#^{7{vsk;OVO@p>Z2Hc+#kiAf% z<(oUvG}`tw*dBMPuwQAr?lC!cHAti6yTsl8bhu0PRx-N!#-1>w+Qws%scGn~ZQ5L| zf;|?tgdo?4`#Ov+^TYS|ZQH_r+*R6DO$Mlzjckgi$U}sMYtJme^nA|yp}PsG$^3u} zn*8)QCZG52WmY?5$kP|>M_t-SUD8M0SYj6Bp=J8}2ZE*QUtbZ?&wW1*pRFXXSthS- zf)Z#2ajw?hL=o$?`>wuuB&ZBp^<&eXZP^%)-EwSLvK zm(}uts?btBTeCrc6QS2@!rVr;->F`mi(0K!DSMq+gnHt~qW9{7^oJVs2{othVUdmq z+g$jXTD0BI7+#UU(iZdzytC+yas(=Q8y|l=IuG>7c2wdON;&f4fpLuWLqBBYLyiS% zYEI2kXesN5((<%aA0!6PiJ2dFYLJ)gZoJQ(#ZP0Lgz&T=)+d`+#=bc37b2E?s^6o| zT5*4j{7GCxt~*RpoEo$vU`%7*fnGr{bUJ{q_Or}>HMPzM=luciEAMw8l;11j%s`y? zD16b}#;HM#Z;-&@aMlkgLuQrQW~j3WOEH0uIPYyI9%lrQXcfr0pqr8o_XwTTc{$EK zkAmbLb)8}81E;m2Sc5yOYC1uY7|xn2sj4d1YFt)xuw-8;G=R>6apk4h zcDyEL09>%Kc;GE@uOnO}u*HL6^|a;VLs=}2nJB|2t4l*E@cl=!R9VgI4C&VHQ+^u1 zM_CJIH8YYx(k>Ubkx06JEF&^teQ(~jD#7fFaZ0p?g{|`T6}c!n>c9pmlU7L8B80?d00Y2U?})= z;X>0v690qAF8=8uH>I&7aEkFe-2-@qYraz#({D;T%v9V2OKRwIp3Nh^yhvJa+?z1V z{TEt+9u6C5o;iZVpzq1gtPJzWC6q82Gv8(bbExqOC+$;RN7)z;`V%cohp;RcM6B=F zOzb3Twtg;)6WDz$J))E_z*kQzbKI9~$qnABzz|Epa8M5oeGfJZ&e!By*pfDl=y=1B z z8ihT(wmWGQ_NN+%=K|_pc6pv3@>3E?TSS?UWF0NMKE=IT zTS@%P*{tZs8c$u$b)fxo(c>J*^Q^P`LVVqifCXIvRF6;Bg=cOl<rG^DXGM$IzG9z$Q4!fnAkQ7NR+t_eBqcj+YdvdE#8m8J4Zsgi zS?_xDa5$@|?n^^Irz4V#4c1tGy5W&Q4}5QntiRS4l`9NNa>T!q&v!V54XU6G24?OY zH=dKG*n`Ew)3ib?cS>;oGPVMEG&`E4<%U)hg$4DSHhOCUaJ&C#CkaMlKwPQ!WN4VF zk5{~+AGVpUHgy+-71-iMNR4ac4+Zj$EjkBHE zyR`%dmO#zy?sN>Ns#lfGz!l8z==C0IH@c)vnz0{d5V$}dHgsZeT5h2exj*rz6sCP! z`4grDr~Q=xhXI2(6mKV;8}E%%aYKn&L($q{dC;v2@0Z7qus{J>r1re={vukG#Y%H|I$Av?MunJALjT{ z(J@zvY^4h4(h-6hop~%9xbHYnjYm}#DU*@+UR}8frKsk}(1CI&A2KPaM)cc&G{x*o zN7Wv7lslb2wfkoV_pLPjA6FF$)nY%UVk&qyu4L*zO^^BPcQz*1K)8rD^MC2N)KL{c zxDuZ&JvSR%F72B7T!*Siu)26*qxt+gz(zuaVVYn?oTnN1ot~GoFl>8p3;J_!ViO@> zY&rNQ##h0&iZ1Yhzey@mm!Wfn#aYU031mSW*?~$Z zZoru_Q@v_yhU1m=OX_+vt8xbRfZf{IZq2G-TE8rU^ga9-Y$kPSN}b^9RMxV#W?}?8 zI|8Y^Zq@hXqce_d%PMjT`0cR36%9rt z*(2m|vbu+1iKy(4L;(LmiEtTr1B3U*8#%B0wJxs4E*U(q_m4_I4W|BaDN+C2ap>^2 zZKlx_RC~rL4nJpuh-3A9e>yUfL5mI=ari9Fiep=}w^xJ@YsTOQ##F<%Br^{S?~{q~wGM-tgXhCdg+iC|wCQY0BH} zIlsO8lmoe$T=Zl_~Fes^!SPujDL{LOYtuduQK?5E^h z$ue^$VgtY*2hcrhcC!o!cJGT~K*?hPSw+MHu&3f(YGOEj?<+s?%Gj3Saa$V{9#+_-xHXb1mj@$Wdt1Y9!J%`&<&qP`4 zE`)x$z!hVp-j!@$Fp56?0^vaV3kjt)#ibAXTu2L0lC9{H5|P+o1r z1FaD;XQtMa8Ho4~tCH6p2n|_%w(*2$1K;iMWCv>P+NZVe^z%Uo;M1ebY4Hw$ zbgdRRICAP5=9lRPDv)eEo7CeY=(%PO!9Uq~>6Mt7QcgUEH zk5a$a(JO#wgZxfwY>snqrdHA0+=HtDF+FWz$=gLK!kPQ;ET6~&w4BQK$(fo$A*)@* zam`yCT%9IfFh4mIQm<_%F?No4uEsSBzuni*G1%3=A?<#h*sH;Z&fg8Cwi1JkPrBaTZ+L0l_33^65*$AOn7&Z!2ao~&fvF{<)kM5q zpHJ$1WRej-A_VO&29UZJl=$R%~G0VLZJPVo?nxeF4)1)w|)m z&EGRVfdq4T)m&&3+J9euLV~rfPG4E;-Sw|AIqu;>=BX*o?pX23;;`al8@Cu2AdiCXS6&=3TOQLKN zfw3j3dxAvo17IxyyW!J>)hE9!r{Oq(#!fI?bVW_SoKae zjw@&1*bN6dijIBQGQgq9q@UPfK-%H0 z%6=|S-uE&7Hr< zK@0j<`K0i1_&>@=E`O6OAb}>L^Je<(HO8gn_VB{;E(%oahA0Ah)xU)rdy&0zj<$tt z@9~|ecc^G>_-wn^-ihlG`diqHaN|MrpR0T_IFYaMYOyN?y2}Dx{#sr2D`;@|q?Kf1SoWFKcKcg?4_y z6Ul(0i@>%faRH^{J~Y?+4Fo3fkv!LLWla#Md% zr_#)Sj2jm;3H!fmajxPPJpRKxYZ@arla6sDO=%!|#Zg^z zH&Y4ud)i@w2Z2?lX(m-f!Ghmd5bEV+x&TStTMWt0z;u`wzBsKP<_VN?jM@_ z-$QJGDQ^6`wEtrR8BfVK0(yiL=P;{AW>WOaCC>8XTrPt4wb(XeEpjW*(&b7lYRg0Q z`tXE;7}CtlwOL?9nIRD#)N3)%X8HD25|qW^A8`4C{mtU6%y@Y-nCTcjw~LgPy7l%J&5@bJuj zY8NV5%tOFc?t zXQ78E`7b|@_-{Wqic~BL$*j(>Xbo|APQDwypPbwFx%sQ|^hQxj?dTrtz!9>$Q`VR{ zt1^CR;edEM(ha$Pu8-9HH%FsU|I2mJg#KdnzhJc#uCl{03@h3M$1s#3OK+!UlgcZk zPP1yg6%{MEMhCX(w0c>}zWd*kCxeFP1Eg%C{vx!i|J|tiUpPmj|GyjEVmSQoMmcYF z=lZp85^{n{yMAl2`|Uhjsp8u$?^*Jmq$y;Ba(BufB5g55si@~*2R^(7or zzhizZjchPv+&Q%KjnXAY+u%l*O95qU`z9cR!0%Gre?tk1Hj+pSi8caL0BAyP%nqseKg|4&CkP8{BF68>KW-$($n8^jf1g z03Pg!$%Gqr(RoScm;rk9GV#*kdfk-d32wPI$0$DtGmf*5NaGC&bAc zg~(8JFQDBe!Z?{nwW{A-Rsm4uWC5+pcXxYm4fZFJKUuBqj3H@T%^F@)^j5DDVSw`-Ux4!bWFU*%{b#OBop$< zFHQ=0Nrd7vZEU*Xb~Z%Mm^0k+vz8TG9J5;?wS7$9sc`*cC@f+8BNyS(t~FlV_jD)r zn9IB*7S0cv(B~YROV?dyNBDX>$^p{CuY=2C&(?2th& z7wC^^Cd3`!Im?^t;GJ4#TZ{tP9SL9dsMym|TsNcu)72DaH3>^gy^5z^B+D4mdq?dy={`#hOIy+9`>(ItP^|&)3B3$g+C)KaeXp$LJo^zA0yLDc44p7@&5i48hz9%FvHC)&N#8z|DsK5X;PJpr0!DK2>?>An?t2OViyG39FLbm&BGe*9eBP zf~dz|bf{V?BCCIFneYzpd@NwcpA#;P3}?VEj)TZKyF(qx`3gc_%aiv@L~8Yq>KBU) z)60kc%z>O-G)9ydSqMdm#9B;79QqjBQTl6Mg zEx(BjmaLSzI!c+7O3-VV@AiOiRfDv+iqPbAS6TFM52x!)==D!VVSPM-GkXzZblk7P1a+K$Pk_MjOc zg8cB~yT{>xhicz}ktf~@96_~4tU@Up1q(h~HVcbZ%tQ?`bLqp2sM)o_8z0R$U_#!l zal#^6AI(M;awmi?v8Msy@4R5aQD#iQ@_eN&?6(}`Q@8uid>JBU>&o(ul2*L#O}Pr3 zuQU1zDw|a~S`+yEe zm1;tgDK<~uhZ>z9CE^)en9X?bt>P`ccz@$s&OMU+rLZ<`j!AMRVjk%A?Q zlCd=>{`JkB7&~p%Rg%^XbZt-y*d4h+?_1cX%C&NqiF&_+D(Mr}Y}e0Bxdodjo_DE* zTfg?T(&OIMuaaSAfK_dIr^5iXjF$IVo16tVQ81pkT&<>OW*b*bG;aooYPa_lJI}rN z2Q%kvGW)~)noeP_7=c8*g>++L%*C=lrQDZVu!SHqyFD3S_q_8yf{))oL%NAvA5*NR z-t?#5=k4p=v$fcV44x^tcQd@ob!J}|KC50n<&>>2Zx4_Z6ZhtB*cTYYq#q9pK5jcW zP!C5Q6DaxEoa%|4;PyEkEN-8L@XsiHD+GW@)y=;k^nh5EAi?xsy#cFh#F%$OnnOo%-4Tl?Qe#S_rQz3^W-?Am5{ccD@6<)Ou+iAW;xEUOxiK{Z zdA`h*A7zF(%=nUze{d=x(t&wnMlz{3wr*i<>L&#Pdd{B}SF0nTQ<^8lUd2iIHwE@$ zEM80IlrwIHM=M?r7;gYkue|G5f1fs61cS&fG+vx^Xg1NRS1xASKbGt_(-ODqpeyVU zG|x;(>lJl@mUzE#+qE^e`8<+ip0`4NV)|E#G_+U#7swE2RcJf@S;o>9*HW%zbEh z-}j5V2zGy6v~^Jb=@BzKW{GwXf!xIXQ_Vdg)|g_-e|YMDKGRMt__{0mz@$Os1!%S7 z^xYk?%V1fGCS6PPAnl76l9Azg%@KT>>$6|Z>C9=&RA0#}Onp7uih&VLG#xk%?{P6o znY?~56~y=YOyXAf$RpS8y!I)lS83UV@&nb$@+TlooL}G@uc6YAiI6w`(XMofKW1grlh*vbDouR#&Pls zC)b-0uttq0^p1<#=-Fql_oSmdHh9 z_Com{%sP%CGp&<87jH$%#5`0svIZon;|^POs^?_dfhP4Tu3$a<7YQ|`IrFrU z^ZTZu7I@boLsQ1Tt3*tV=k5^wDh)8m4=Kx|=lgJ0sbkYX+?!@WZa18rzH!m|aSD3wsr?AU5N1b+|S@kI-`}9vHAtjY&+;fk4*&~rL^U%5bkMu9}#{kr9fxkAh z)jBVVQ4FRLJ@d)^LijfwzWwz!kF1Pu$5ei)VEoJ_y|amr;e@@q>Xm}d6V>Y$h%mk! z-Ur`K-R^+}7n2+Ap5@4T?v_dW^PKo;Z>x1wf9EE&e=6v*i`GYaOM7RBTrc%{gv-2= z7{1^D6GgyajWEM89*L?Uy>&IYHFqGbYnx*ZXJ{xB%chpB87^rJIC=!oL6mrVDj0M1 zJ2F@%ZKd9k0Dg8XQ#fi-k*}ZF0A$aR+!FjO8Tmj15_{tMR7^M5!};S{p6k%!tmJVp z^U=7A`{SvVVoTv6>VPARz;D;3m@qxONG9h%1ZA($to_UTu%IQ(lbE>2w3{u%w}iTD zJDBpl_MH?tlW^>U%NTmcs%t(eh7ty%H0Q}CpH~B2$>4X;Q{`JAj@xH;VROC&|7Njv z8F~E)-g4-?=x;sV1;!-9X%E#;ai(;{aiaZUQ49l98}-tYQa7UO z#?8#{CaIQnQlA$r0Llh3num|1g-^}m1ot#5oSV~`QH`tR%v`fFJ2Jv!4b|9 zc$GQuJc~J+SN@|n+ZMqwr!KkW0Xp0&p*Ve4(U@RVA7zNM5wFCshD@PtDYvQBb1h_W z-&P=3qShJg)*Nx>FkQ-3FabY#_Tm_1eONE@V$4APPC|JP?q~dztA}Oz`MC`+>nw86s^*b zlKK%LoqSLU7H>IzpbJT-&llEJyLfpxP~7XQz-6X;y_rL+ZYW?i)}iCZA?qcJpaWVI z+wUbBRuVgi=3=r`&jo-$cUXdRolBEKEUN9RN6tPEv-<7!P(J)G6Dl=81*ul%PM!=4 zJ80wRz~}m}_Q4&&=@K-e%l6P_-yd}&UypmM<}QD+!2Bhc->m4xjNB^xiq=#2n&Nqn zU9)1)gq84ivG6zaI=~*%E%}8$Po6s|^tJO~YgMtJBP^K#`}h6ZGXzaU3{P57tjf z6w;?-18$R}E_UR@8v&{k45-_iue#^Bf-${hUy2ao^hqrEMgU5A&TMl;8ppFE{HG7W zmH`XY%17)=7QQ1jg)~-16@4a@(&1vMZG!F2phbk;<9WN`!tn?Gz^l6j&cJD&`ndY% zAkwdA&(}Oq*_(LUU#TFcaVq)7Y9Eu$u(#=|<>`+8CMq&QC!!&{F@ljs=5|868A}fy zX9C?`4&X!8`Ex4c4>#@dJ3APz9(A$bGjw@j=@t(R zI(T0E%sP~0twz)@s>OP68}N1Bb%QW7(!*t69m- zUUJ99P4Wnvr2H=mhLY?c@-uVs4dODCz&FDS=Eg@K;AqMebAKO?k101!GE(dQI9nnr zY-Q2VVsj`BZ_`A2;S?0W^UYCh$oVJfx?w3gOwj-k_vg-N>2wwP1kKjaC>yqfbUArD z=-3)jz?5l7Bz_ekT~Z?oq|rp6K=e;=kV!$Hg8fe_)<2DH1MRmxKf_xb?_Ds7N{e(_ zgsT|_3KeEg#YAJG67b@ZUenba z;f`ZXyPV6XXM)CDO$d6wm6c1a(VM>jed0lzy7_^RxP5GXaPHFOFFflsOyG=cdMa7> zC`q`ch7_+y<(LFZYJkqV4rf7_pMgKcJ>=Eb@o-^uB0(q;8|Ia4_lJ>0(-mgy!-g zPM4E>{V2ts2G_)4m?QCRzq98m((NU2yGdVAxI=3$#Om3aZ(+KQ3dTu}Yu?U6hk~Hl z*U~xIC%fNo-o%?j#@@*xR>N3gHKpFrrj+&Oc$8N!1f`7#m^MsbmotnK{5*Enr9Zj2 z2$oH%-9NziI9)n={404zp&hYYm^SxeMTl`>N#t(%uF#3y_a2?osVS^Vje?+fnM3iD zC*4aEKU!+t`}>bb8{0)#Vf)XmSKh)@ThY&%4F}WidS2e?i_vG(Ue`AqKQ0{TH2m8XLM6U^K{Gkv9#40 z;hjwoXiMv?v=*6~roj~R_rtb(n@+ z_*q%m>38MGjbhs;*r1Y+p4fHtU7gNyBBWaBC~<;c@Pg%o0u=g8h}cZEC8tAno0H`= zF;*PI__l~n)9Mu(!s1u4$F15Y{-Pu|gHA#wiqc=MwCEMBl@`u?^h7X zT}6o9l_I1!@En~E>og-N!|n1o1Qnji8oUbRmL;8WhI!;p49~X*sN;( zgmfTC{L)x_b9XiGTRQU1KGqS|y!CNk%O_Y7XQ}Q45b9M;94k!Z`lo2J=$^I#0X_Nk zJ9AFTlo~6`dk5u;lCM>e%_2}EVUVT7qp;2?&9_WL79>I6)`!6+=4E>R;*L*BTZ}L` ziv8D4^=bdL#n4En@bvmfqKffYIUwMjgE6s1X$3(5F z_Z$3djL)mI=M_L=t9mSv*Zu$(RT&KVpbV($1yltL0Kh!cg}p{yLj4rH>V|!MHOkDn|tS6iMGFU z42$bzLD%BpJ8hbi3VXwT3_M_JWrL6_d59D;GxIb%$QrsQwD#cF_vdu&VlqY|(70#Q z;$n{S^^q~Og@LyU(BwVgkM*K5GYj5p{{qcl+0rn3U^ng|M#|kg#UeahR4rleZtdrf zZL3dhu_Ccc%Z5}|?bv;5&=>%Hg~H2;M%ij*Kl`UT~jLs>=8 zPWTXSdmlF!wxs9V9N(?n3$yv`oNw9~(4@BSx5Ggxu&|cxS=5hKT7RE*Ivg^p!^Ztm z9NduWV}!_Tn{^G$uAo|XHmC5xGnoBMq7OK3MZ#?Jxt5c!uLC*cW$4CQNE6ar}eLch;*#9 z==pmre9*9Va!wJA~T|g}?U7A%ShE zlc&imEkPTR=~4!Y%_Q&)iyfB}gD)(dH+OCLA*NfK6@|kM>qiN=fDT6{&O|JZw%l78 ziXR7}Q}6j5Yn(6~k9<@n7k`AoO5DKuSFDKW6+XsJN22YoS>tQ${8k3073COWE@AUl z1s*Ih?_;$IxP7Surzfy^tAP{ezS^%VULL<*KbGA(i_IHcGBxJIYc28d7ZdG$7nITT z_|@Xo(bzlz`Ro%Hofw*(c>01_#?Q}!pgI3Rn&<8hh{%(vtvrl}((kIP3vFoYh{$HU zYf`<}YDA1+t0r_|ypT2NXW<-}f%#XB>+7gUKWbMR{dR290TEm4zjs)*plJ)q0}`vF zfS9#-Qn6ABRp=Nq=2kDyyIaG9vyY#?3C)%73ww-TW4e*!h~WoWB22T0IRtI6fB|ud zqd$btLiS3Lb}v_57!z|;Xp|m8^(VbGvaIHzRc!vj;k(9Y;YqxTkLX1hGVHSI=~G4| zlh)n?y2&uqB2_e_IF98}MLz|cZ>poocwy)Bx=yHOdq>{>p%YMIws>9s7q$-K{48B@ z<2ipA5N3B*OLK(QOHu5My*eIM8h1uk2@&I?GJfTdG2L~TN$WwP9nx{I@e*9o38S8;5|x_XLB-`blH^kcl&4o) z6<0g) zzO`&g8?O9oTyAZ674W;G7`l^#P9JyAK;%Pe=dLKwFU8CGDf7vlh@|L{(sdvN!}T>e zjU7?6c~Y3RuGlm2;vV`nIakzwRSkxI88fW%Y+iOBZ3!ImJMy%a^2e4-eqN$b9&+l8 zebqr2Sfwe;MwgD-RchBB^L&C5CU|RY{j*SN#vSgk`a{q-WhR;Z#RXd1`(uZCX?hff zqQn)C%j`u&fiowz+BQMkyKBv*(o9R^cCDX+#3M9oWIAB|9KV#;S)TgIvj^h8g}iKA zhvBirKak_aHOvoLYmTLxDAW9pY8_h2@aiv6SN{2WeMSZJ<8uBzAx{n3{^_>CY&UJS zC7scgKdPw$L(i+#PW#hz6#PmQe#ooNLTmRYGNo6e0%+aQ!NE9l5sDWXH<>kP!I-?N z;O1lRVOZ{w$>-41OFH)WNXbbV6k=|d!tRIhWdA3Qm%nE5@LY7)Y|buho@Lgab5+Ty z5h&TZrE=AE3+Uwgnb%c(9tNBq>7P@?=25zK4}}Kux*(#v_ARa&9@NxQROCFp0!^*H zrjtX@`mef-d@k`+@7F7+JGt5o_h{MtBg*Yyc>Eyc60Kb(vD^IP#?8x)t3A#Kx_FwN zfr4P(yN~+hVSHn>x8cYO)Nxf^WA8MU-y~@a&y5~PK%}W1*KU*tz~E$RJo}F2bp+&w zM<1)Y-!0o-7aOV7#Ye;8Ql4<)=PCd?9(D9DOsIA(TVEmI zf;lgk{JMwnPFkS5v*K9-?uNOpG3z@#!`63kpN2UJxUY<1PqH5T^L*&{y=t+iEMl3R z1#fY4@mH?FcwX&`dnOf#=y2?gjF}!HiE+|pO{G9XztiNSt6a2x@}5Nt(gX6G=B}7wDiz;Pa?bf(_*E71944vIbRe5bpmv zC5VdlwgvC18Agjmd!!Dpo`Oo5Ei)f#`w?k-n9k zrB#Y1M$Ye){S*p8!R{u^Vg8W0u5-)rgk0M77PSf$h!e~5Z`9hj`K;E}cc}Qq$!m$8 zHfSuib%Q|lQr_`(qG?sxPDDJ;zvmTq-m>#X7az4Y&TygDea+8Yn(m|NfXDM$wfA8V z7h4|u&>Mued-v1v;m~5#ahAC`EZ-p=4Sg!BKLkVF%}w!xOsKVO`@&!?R(H{1@Pl0> z>Nch~yxzE<>SITpYtFDqG%bt2G3Ui_{;%tXw0*2jYdR$Rwm-OvE(#qT{l=;!= zQ&Nu`rZ;vg7O1luqK*qZDs^rTp*8unt)dA5MmTkNWnPC9aADD59?xnK>8T^*qIN70 zB@8y5s>XEVhyvr~Z@!#})F8Ei)psrG^!ctEF*k=u_XF3kBxb>ofqr|Nu?`Gt%D5kW zoQynO*B8=9)9PJ-c-+m&;Ceo)Xm~XKfR`Q`?wq==9e*4V*D$dCW`B$7XYuNv9^tU( zc`0}0iI@D#?sMdhR)Rf8C+gd(y2gb6KI;6S;jJ5q)w^}DjQu!k%|O7Zzg%T|MH2Fi zLl@R%US5XF#VDrc$xdrjYgtA<=J5gzKN6IEZ$is|A#K9@LCMe_)R8b3aX6PYze2#_ zJ!*C^(bj{JS3E;AQ(OXaxBI^H;psf+vL-$;T-$=_sv$lOhMHybD75|Mp621HXS!fm z^hy^gxdST1)lTT?e?on8!F!I`zel9)Kh9siNz1#T;Ci1Cmr@bTI8SXo81IewE2?g- zoCK|I3u>6IAL|!MM5C^AZ&igcJ{a?V=!Z?CL+jM|OPi4ZHsv2jo<-4DlUc^K+2r>}Rvht0zlshyK% zzI_!NzhdN^>IDbv`5tST)|m>#Gf4I}a7o+J`l)0XV;wM(LgPtj5#NknMQ|2?M9bxm_WYh0x5|3>$F$1b5{n2{f_QS&}Q zJFYi4#;E5TR=+u8zMon7ObsG(XmK%j?8npBja%@vAOh~TJB zTDUH9Xc3jxJg=CAcgP?=V84WjAnM!_!dHm5gh3aL;ahQPD38Q>ZLeh^B6)D`=w0x_ z{KdF=PT|pOs0+y(U^x60Chi2vZYZYZ)39LWJg(`@hDi5{K3fy9`{|+ox?W&QIGPia zGVfQ>!t6OEzVUi%FCq?`iC)jebZ1ePv)+LZ%jTI;&Egx!-{trf(0Au&>>k1jARPc(%(ZF6+PKslqx2 zVI1UXe%vjRxwvdUQT0uYs>Lmo;!tzx#;IJCJ^V#jauUN&eWZzp?!GjbVDY%0l+=XU zg))fb=3zwo;C}dx0Uhe;o6W8G(F#L*UdPMlE#)~+G|oG0ltb;+rTQ}_{V>d$=<+my z2a#U}I!IeDB4WrcUf&4jWp=cWzaTSiuY=()G70th#}M(9;N4M^bEv%FP>%V@X&5?K zTYXPx#j^7{k8lY3H!&d6?)_?ppQfM-ML>` z5I35OS|6oz-rx&>fgiVs)6R1LdVHG-9X(cgJPhxvIk4D)=?+qj#!X4D zwafMWoo20uhGqVP5g%p6ipFK@o|s*lb3?B+(wEzzwKcRljuD1x?=aeDCZP5WT}On^ zq`@S!>!qV0i21oKDJUZmTc>YNis^M7`S-e{uJr1(j1~*jdQiW#Yg~{(C-ksn8c&Bv zE2CGJnk}narJc99o>b1lM9WK2tJ&pJ(T*kE=N>`-^9%!&_g~cbkweo91RTSjRjk}g zdPJjTrv0cPnz$_K82-iukqeB&%yQgea7V!1Z@b`MpF^kmx;+)2WYE{^fgz$rG(JQf zo?FL!c?)6ah1;y)M*Af_voTagCLdbeG8ig6OAt|_`e?8tAJn-C5=wTb;pqo$Rhsvl zfY5Ct-U@paVAMc#^mB|SN;+5ReY*288fI={-_eqaBBSq0u?Vh!!HrTYgqu0hXmTL2 zD&cSABY>%^r}Q}@7gorAHD|&6ZBv}6c=IC+bFPo@I2`kK%$byJ{?-5$gF zPLj`2A*>FdQmuBY&9#rHrr@GpLAurQcy+*=s#$sM$G7)G6Y=aYoM5_M|2<(ZsLMWILTebYR62X$_GNN;Ae6-uhl z=t-WvfGR~|W(~gxqcq8D6E%SWXiTSh;F>p0mqY1OeFl*&F)%#XfKPACM+>JPQQ{sY_Iv(QP+^x}v?V2wfZLf;@KIkLM)1R%z2EmN@nV=e z-BS)kRIYQ}G4B8cVuTn^_GNJd&mg3C^4+8tR2+~jcrv)8QyoNIwwaB_%U2deKV@WDsz^eMH@Rk}$_3Ib8gTR!;_l8bUr2H7{Xg++nx7r!oO2 za#&G8N|6;h8*Cl{jn;>6?~SQI zq)91}+2la@Q95{@$;Sp|wm57n>EeeWI}KJ-h8fhi;r5ycF9S4haC(t0yav_AoH$=t zv=&CxUi_|6yo}jvt;kOtevLYuZ*09;M_boKeK)`MCDENmNo3QcMD0W<63c;89b=(Ln?Cy0kiSi|6lqwB_q$tcX^q8h30 z1GEWU%h@m-0>o+Gu#NmAMEVqB6?yzU8aXW><}7Oq`S{U^)FdZF`gtrk>ANE8^-1#; z4h=!ISJ}N*#?E5ne*fMk-+QZ}YG1FPr-c+ETDKj!&6SJDC85@;3IM~ZJND+sxuF== z`ACk%VU^4$a*6`y#{qOSx8JR7C>2F z1v|30A>g+DGSzZyS;`wV8vM4iL(^N^zVLKOT}VP!-ADYle+-AqpaxS z_tb3}ZxG1}52d|55&77ubH9wT5V_0hF~fxmcn0%SwNSsm)#0Z}d5zi{!1$KpFVhy- z`)w7)f)jZm-w>%njE>Kz7IpCCmCQ-Rpte~g!!m@`I}TMu2lt)UMPp|__3kb3!qa=_ zRQ;NDLlvLJeic^oF0*Su@jbaEvj>_B8G0Ux|3;(MY6OAbwDl?ij(Z_+jU-l&Kw|79 z-H&uZq*|#5>iW(w5GyHATJpC##Cr+@*BjTv=#ilp-@It^JJ9P8FH;sf4&#{;nj5&H zV6fxs&d8Dg%s+fPBbBvO5s5QkYqISI?D}a3+{vSlV3;{zjfV$qUKTo@zua}Qjix_g z?o{B_UB9t;nVuDOW!zt}`kL{hcaD)ux>3tX5w5Vf7?>MP_fShVMoA2%DPoD2QTq=0 z*v_s%s0%aHz0ZNYA4NRaRxG9$PQXER!?uM!L=KXDtswdnTGFrfzf#8fU)(;QRBUny zktQ z)gKZ;Z;wRi6RhN6s7oaF>yDOXeDFp7UBVZCtR7|dL}W}}ISI2@M2_=`1U6qttd}qL z_Qd8t7R#hlf`rauTf3C^MWw-+oADpmk+Je4VLyJWL!e&8sRwi?>YP5Cb)`kv~1}4SKT_Z+5VXCg<8lTJR=+wG78=6kH4-EE`&+K@5E?@ZO~BJ)w=M} z2a)EJn@)0Ef#HWofjbXd@2YZg{&+>(5|PzyS|!8lVPU9b41w=yI3+u*UM;UC?aEQVW%)TOr!R51I$lY;PTRsqDQiKj|1RBwO4ZR% zQ5(Lh(KAdF8y}f!CPITvFjf{G8oqaVeK+|on^TEp1WfPV4?Ev%!nSAR6}8PjU7o*q z`X%&snIU4V)c%VJOsJ2}vW9ctCOjQJA?N7(Ak+|nuKC%v!gyp**7NZteV8DVe(@Fd z-qA#eQ-*q7C5&^s*2MZ>#{9#_GIg)w5Da;rR2W%H2ZQn}oprJdp>Ovct&ewT^XGVm zYck{2>gEL8DZj>yZ>oss6SKd9omTIUNbhb9?&39o+8XcrYpH4k96NtfYfdie#6N5r zl{$p#4(IxG#Bao&S8#3c*U&KNSLxT;QL_bFJY?-0JL8x19}DZ__H{7CX6>-zekH1> zix3+BJ`3pd7GX~797I|;9=luP9aI<`kAKW11jBZ*+qZXH|I=SkBYc=Am{AZ0iZ>pN zNce`x@pHRt%i9szWT;f>%R88t-}=SXc^~wNT5@aiES6om`W<8Gj66WNid{-{n zyQ*i>T@Ur?t*pv?SA!;b+M^$oe?`PDt>4{0AHu+cotBLGl`zu&qnI0&FT-V3X2P&b z#uFIIY3)*5e6XbNddcMi(Wuiv>2AI74(zy12cGi(;)bD4krnjqhcVnFe!SUIza54q zEFD1z=*2bD2qZ2fe|c;9Zd zR@yoo6q)AOSM9{!Z+moDAUfU61kFinY%w(XjYzih;+&7^m)%b<&$qW8346m`Oo>Re^T@QFjbij;5m>Vb4(%>&+|W=q7ZZVjT%Q z$cMVVTK#TyrRBAd7ajAh(o$mBmyx&EED4p?tgMm-HbUv*wFLb^x zwGLAnBH1N8{(hDfv%8%6piJRK7z%H@v-c|L-|SePj~xwUPA~{uW^er=HM_>$l9*k| z9TNxW<}lorvOE+{>_8o9Uw2)0n?Pe*4&5Gj$Vy;H~Nflg;mgd`9^FiM>HY_Y~U9?pbA4qFL%FG_Jl&#<~H>k88GrN zCg{*h1w?xLs5|KT7f3t&RYWbg9SYpLg;wP8EVH|%?rd@K&TIdAzeHj7uewv)V45jK zGWw|mObBjcIF;pvay5A74*XmP#DmI*3icSG`ja=lG!?}lqEhMGlXliHaQ(XP+L!*Q zQ&FPbqj!XW6Dipu5wq<5JKA-ArY7cRDPZsUq%yszKf3uC>Tq5l&Fx}G#IOG4;&uy& z)J~UQb7Xmae3?634%6GTl%I-n-0Vw${-e)&{35XR<*s}2{(EX-(I@V1m1_2E7~tL^ zDeuFHGS>SZQoM^`M6o`l`d}C;sbuNyJw?lt(Sl5AsDe!n4AYq$-7bK=H#fw7`w9EO z--x*AyiRk9#(xCdcFT)j@8VG+|6@zXhkh{hJ^0uwg;LbMAk_Estq%-u?4KWExdg4- zH)tI3!(->Ivk(>4bAVy{oO9*5iZJHZv{t*88M&UjaeYX6nSMr#iPt!6*xL}1CsBL7$K{d~4n%I9P5wH@1apcG%@Ia7q4(>BtqgZx zqDjRLGEra*<73I`H$MzwP`P&FD;EY@oflM0MlcGhSwd%C;CS(sFQ}gM^23FrHRUtWXw0$tljU)Ud~@B^yX8Nid*{k$0(kl*{2V=08iuWhj!d{a`U#$b zGS{tf?)Psl>B-AYFL%=Vg^A;Q4+yWFQM-A#+8sYlXnNG|(X%)Jwc{i6A8;-(_w()0 zu9DlRbMrb0e)YSsXu++jv~v$479W-*2LCVW6>q%DxGR|kosn81<&0j)lxsTh{YPOu z!@Gys-Gg~cJhn;aR{lE_WmDC0*8BEutFFyM};!B(Te(TNNfh$BRZixD7c!e$y4vzd&Of z+1~CsKZ1z*yEh$Ej9fO}r$6I#;d!N2X?J3$R76~n(_wGu{5m0?uM2&OW@^!Q_jl-+>!j`SOqtUUe&*-;b zgRCF?N)>Fb=;zVk?DyZY(d?9xZ1okkrMxY`L4h$DwZS#ZqqQ4RRhg>TCkK9LiF%5P zJwKq**4+t_DUOK5HkbVCJ2z^}_A*_aSOvr0BfCc4?M0-(;L)ws5vbho!W{+9VHh_Y z*|`|6in>qU@d`dCjyl4E$6WgQP`jUAf2S@J47hOfk7fy@#qnm&z}hZodS~3*g!76%KrLB>nIAL++U5P5TPY?XKp4E363%&kg=zP%!hlF_+PaF?%2 zTpQB~&GvSS>}V|_?a1;zAGW++rsM6aB)_xEzdKN)awrIwzI6X-R{MF?u3*QnJ==Le za2`g5G`9QPXGG*~M!}sYXzO7xuw|&yQNRyRKP$EVKC?3rPg<~XJM^Pp`InRn-2|cK zO#ChhckI2T0focA<)7!{8BRXDB;O$ftv^j70}S7$m$|By z0VOhz?p!=ei+^~ALvIq**t?-D+nK)emm)&GvI63l30Ry^+E@ROMN8{+n(-|Hxh}?*+PdSnG9qE-#GU*&cZ2?QbaD8ROG_7^~YDI+eJ7 zhXFR9F?|7rSt?2)Qh1}#;Wu2%>>`KFpIe>ljat8iZt`MeMnw0y3DK5i>)T*i*S2e0=zsY_rf)DFw#umS2qZ{;Se+&p=JqbClKj{0fnnjZmqJEz@O{4x$ zTi(stOfWrUWTRHJ2!k;Xo`1Q!AI(dAU`XP?zGF70J!5xGryPE%+q~MYsSSf4LUi{r zf5Ov6O8lU^c^-L=dW*|+UxtC(4g*J=X!F%TCT%f(2pn3m3XoWq6L#a5QOrFee% zeIF2`lli!8ofr}2B68Q9HiL1m)R&Ssk3-|*y=E5qwD*Nj=X4>XxWlsgUrZ1FQa!sn zse0M=l;E5nDwix!efiy)t1@lS)9bP5&FKvt!M$FV2WanuqQ0}E)3-Erpq+V}x5Zo9 z_ia$A(2Q2HwF2a7mVDGKRL1h9Iww8GXS8`>Ji|tN-nrs9=rZ-;u=|45xphc*am(6{ zLw7)EV`Pc?vgg}>dsDDf;j(%PEPuJ0zKiHtvuyj|6#dB?!O3X8`tH|v>e%c`?>r1e_;r)~L&RNd!L9l!y-eN8b8<>jdJ)u!V-%*?32-&IJF>n1d+ zYgqi?9f#t*pS#ngPom^AUXPjlXn7*4eEML&l;95-{M2nByx_OQ)Bfvb_^@>=9FOwR zaLId!tU>-DXw0S|$%gcNGh&R_8#8P_pN2f-CgxH~JD2dKGo# z@ugm2mVjR|{ZPJ91+#~DTCXL|PSBXngY@LX-=Si=)OcgXv zZ@yZhYE1e*@yeAoXnt+( z)M5L@kKrZs^u;vIdy??ICpbM^ZXK%qoce4wFBnEFxGtXYQd-vk=)vxr#Yqh?F;9*z zC(+h>pf!`TOYqbV81eP=zjFE!0mq)R&LcZ}+4+6<1K+WfX(96WTH)z-Csg**v&m@T zJT&qZBpTdRgl@8>T&l_i+VMnzE993UXjwnQtH{?!iV-3Ui6|GIS-E_FDx>~qNBe(n zN8KL|X>!DdEywpH>!aSNW#5@XzlN?>#l5)%Upgklw};na{o5OJ#y#pqbutObp+?_O ziPrU5w^^*-y*)Us-8V*LDX&kqdA0#that>41mC-}3w7{*dc*y29rWN2_pxfy*5{$+ z?T!Y+pP%r!8#cjb?hT=qLw7jOq|oY4U}n#FaO&lFTD#jvVJFY2r96Laxo4NkdH8}A_Q7Eq9*<#GmhrXuBoMh!ugH9Q-alT*rFlo)Bcy7`jCM3ou^%H6J z^zbuj=jsH~MIiGZRvUTggC=t)r8`mZx0hi=e%)fjX2fD~0SAgx-y(oo4>}Ueh{>eOd|Fd&MKZTp8|zH_&&{1I;~k zSuk?U`uiqpMN}ag^5vviG>nxD2NvnJ!}x(5cku}i)K+^0&zR|kUm>7z>Y8FT47q(D znK!`n`M}9dvw9I1@pSbf2blGX@eC5ruCHYH1p}M%yJ`aVd5T3tSj7U$|Wf4bw0h+96I zu&dS+a7v%0TXd@N43GCbh}KyTiw5l7#pLHO*tlZnnRS1!V~@8DZE1q8y

    zroU0o z*4c-n&xUjKTb=h$pxfgk5Sr<`z?CsB-;$kqUTYJVrgKmjeMXxztHAJ-*DzB%1 z(&7`UTlMYu;Hh31ROM#hc!~?PKghGbBUeViiHmx^9lt`r{rK|ZcZnQy583Q{XwgEz zIc(cvev(#qkH>9I`fzfq1R5N_M9?o=gki6N-)A?jLlfJX3NMANLw2QV>#itc^LN7u zwjaN|#onWxJmGx;zhexty}vczb#6zK*Ofl~%DVu=8yPzA6(7*#>W3TicP^kF=R;4= zvKOOTpYrBPTn%dcC9+}gES`Wf4P#=-`T`X!yT-)}ZV_-c8b4d{qNpSC?s-^C9^(jATNp6h|m+Fb@08L@f$&Yf(#FLmyK%8c79<9MfF_?pL|Niig;HKvUk03w!tbAmR&wPSZ!9PzRH3X?@af0*+(t?K@ixVU)ga zqQo`elQhWaSJn?iQe(RIwK0|%f z(ZKanI?#*HS78rZ!l}EfrSyb0G?`m5DgR1Y;^CjAcczBXaB}sS;t>OwR5>tpu}T1b z^Z19C?o3DRZ$&Qq$t=gqJWo*2(I`wu7N>?9990TH?FQj5L7c{)(2{@8p!Xj3ovFtB z$NJwN)}g5%0m%v1+n}8B_5NwwL_{|D{KQ|b8@dC}9C-Mofq;t?uaIup3fXuWpS$U` z;N_E2wRn12UWbU=XOs-r3Zg_!u`C`<1t9laesgHWFbu_|oyfi>v5Y_J-wqYM|K5f& z`Oo~4KVgZ+T+-UM1SFtVZ@cX7uml)#_&&RLA_Rt+TG%%{JcK&-e&j2)<%2d?o~aQj z+B`F!p(XZp0GdFKj~r5dL@l?gb8Ou?mhgwDf49%Gx?6Zu%|WMldIOdZbt;QG5TbN2 z{@53|2`9Te-i~fk^i40u>a_+;VqQxb(eiZY-G44DJ;R=W8_qqznCb=t7aEw>$`~P1 z!4bt?b_Hlv>2K6re;QSNs}A;Txrazw#e>fA_&}jh#rTS0EdM1sJ?we2FCG!+7*Bn^ zUIW8x!UruDUc=zvy-Rz#vHGj&$5w9A3hNMga%$Dq{nrptV9>{8nt(b54oRQoxqygI zjUS#fe+2^{)iw82uY_^$E;lWPJxlmI{bR1v4f9{~?>CG6?@_mp|MaZcAS%|JbhhgI zjuvNMt@<&5%@Yo$tMxt5#OkSw&L1^S#NNw?p51Ra-&ks+&PKQT9*1Knx9!DkzELcW z4JAh3+fh;v4K}W4>Gd*!c*e?x3nG?udfh_lYTCRC6f6d}8?K@U=LUK`BIwcdD+T|X8b48| z^8!!GI42_Z*Pm$1!Nb7MrqW^|+B`I#9>*4RTr>jDF#a>vHyV4tb8O3Fn`C3``vs(1 zqYmBsA1&7r=e*YH6EYBS!v=YM=Chc-+31uO!OI1GRlnA(IiCu7zx~F==8vP-8o$M^ z|9*?Q6;$Ru$1@P2Tj^C!L=@)VFdxg1E!cZ>rPDRfu0CA;{_Oa#WB#X`H=+(He@>yT zCo%sDM(Sq>(DEth%F1S#bhk#8j`wsUej1^+_N&`&f2Y;|p&?ac_h;hkkW=IACVInS zH1b&3YV|J#C}kmDzF+ncHCvxdEWBNgNHG(Mj@L1Kw$HQQ;vb!Xc~dj7;()faYK*$KFx?cdGEuyvk>WS6w+iW1aQ$srP;k&hNh z)8{j8TcdsfLbYHn0Twu{EN?r~>dNs97TM(rs;uzH-yd&?>Et;pt71)ZM*UUFLWvbI`F8v=lE)n1hH zZmRUawwvQ4_n)gm0|_>@>tA&da2f~j-?w7liwrxe$@)IR5=M4sU1eRUMI@>0csosA zME=xyU{Cl87|7LFdE7y2DgWIN8PUD04suD?Jp7f*JPgD7IK4c7IT%ulihI6jgIbbI z*I%-bK#PWtujx)<>l!1%oSP$W(B5B!<~$$0kUKswmdEEe)iMLav2R{JAG!rYs3c}S zing9f!0p>68y9^YO?p_d2Ol$pNe!z3D`VRF7@pp)tIv$5_iy8oHJ4AY6(&+z+iOII zVc^ZmER{Xj`oY75-mJ?VYX7_+=zDoOl_R|z)uawFj?V}oqWDm!uvZ53-Bw~)ZSn>+ z=ajpoAMZqq=3DDLpZB2-QEw?heXLHbTG85>Y16hPp2%^lYJUtpiY%?avS{C%L@if$ z?qWFd9+52tJwsiWbZBF9NjI?;eSc7t*% zIC-C8%g=W(loow$?dcznA^QSHcMA_9HDBdIVg`xt;BI+M!*w z=UhU*5USiVo9lU|2qx%!&z)6T1*0$4nq*GWzC#NI#l)#KOrcQ8A@r;xLl*`us@wNJ ziANPH@{eB^?1IU+GSJDPi->7iyB&{VeEMsJYKXE)3d$0>@SsqI_Wm&x{m%0LvOxzX z!Udb+@0r6;hPk=weqI=$+Y;EIkDDLG8nZU9UFW{w;MdPw?;TBVc*pp*`&5R+g%J1FJ1jEH$p^}zGStRrm|4X z)};E|C?1ieR(-ChQ%6KYEnk(Emv{!B%o~11Supb2Q*gTuFX}wHMoly(6DFo&dCL4o z5t$G)ksE4@hz_?Oir&QV-6PUQFP?UmfD@yu_-vpCO}ZWo=KHbl^9?7kUM%}oMQeW# z9=On(qJqdD52VWGenb=6AtjmjjbP~Z_7s_ld@A3aN|S78a3vpz-W7|8y7#qR124eOl(D{uu+1HLV~gMX~2kMVF}{RK~-ryrp${_5y_a0T)pH;&GGwXt$-zVZx-PgH&Z>(Do`JznX@FrT_DYX9HFEtj0;UTZu zbrWNm0a|QckmBCgw#@Fq5;1ndoyW_zi_#k{G9@_ut6w|VCyx)1^$|&PR@++_o1g6V z@jRe$AReYI`0T5|j;}e}L*!Rz>bxk!c^kC~s z@(#flQ5cm4tF;1FzOVIDlPqB4Ix#J^KXhQ z`%XCYt**JxZpVT;87oZ#0z09d;4IZrMnL3Z(U#P%5IkKpm*cM8G3bZ(mzYJHazs|| zi`#Q^JD$EL|6n4g7z|Dvt~}Si6ZOuC+R09BfZ>z!tECNSc{fVefDeT96r{m1qYeNfK}2mgDq7BE!HF{HTp5PC>lqrP|3CLnnR?~l;M zbZ$M)Uf)}MKa6N{xEtNMiDFtG553dJbaiob`bQZcA)=;0`1#vkVffZ2ZI05nF!0j! zNfL7~40mj(wa{0AoYGmd4MIygEBaNJu8<09SJ1VpmcN2X*(o~pFKb~qDASC%HW#|p z;?F;iu|x~OANF5h{s4<*8k=-##;|cbUHIvdx(gz`oj6qUJ`%Op)QK`55r)Q_FWN&L zg<<6VjbmwS>j=1K0-`M5*gRVAsX5q_Wr{v?H+4z96-Q$PvClls>;J0vuN(R}iV_*l zrr)XG1EV_3Z?1+*;2G4U?w<+9>Z?ZLn5DIMu0)yE8Ioz1d$f z-}QRw#tAqZSE)^DpV9nPuh_LeNT|b3g={{9y${@x&R5P+)-#ft!#-9|7vnLdz<^xK1@H2dfj8{BUu6%@HNobu z$UDC9t3JPgh>7n5Jq3JGiG9~rS*AXi2+|bf2_Zo1&DkIJHQRu+YrOu1Spyn9(=}9h z0eerTW#;&YjKmy7rn?$mop~QMjd83AJUR-Ur)UZ?iEv6?Hw4&{>=H?xD&%u zYTo9q$o-HTAL-nB*Y}^}mVo!WXwn7~~S&sosOt=aRQy z$=JIMTX+1GmE@iLstTq9?Q;)i(Bd>Ae(_5#aXt$~G^-#mkA0sfwZ>DPv9t)0_}RVc zPx}yXJG>T;I4Q&U9Gt$d_7&3FvVRHHyh0HV0~bAa(&}=dxr#qTv*alh*NZ5ozk<0Cfg|A~*gSLFz-va`Oj(#nTNr#;Ukx>zTC}!g zWA%@tF(oUy*7F_jIRa5-IlbL}hSf04oqdfXS`9{JFUgtxR>7WU zvqEP^fdlkb;L_^~8K4m7@=BQPJAQ?*oVJmb84P`&=V>pzgh;PKYvgw=z=Zg!1(&EA z^gFd83%SoN=`Y;t6YCpLwcuh8`GEv9rP-Zd6HcoihUs^sPn&AxmT+4> z<`fqNJzNS!<%=|(yrjb{&w98kL#^9K`C}H>pp&cAI&wS}bsbQ;#<44(fV<7F6V;N9 zI&Sl>y)}xxw@?1S`Sf;L1X|pOKXpzv5|KfAQ0$`uwtvMZQmbVWYCn5o`?R%cD1Vh82cpP^mfo}cc z;1X0}l#l&&uhSq5*|X~rZ882D4cgZc#8d=rf@`gXZA%f^v!L%(oFycL@Q`*s!v0^F z@*uMiv-Zz0ypv}#KK30NzSY52DTjUEwza8)q#8!kW6&)#9xiyDHs6n5v95Be*m$CXbEAJJ9zb?AEFxUqrffylHXI-@e1V zZiO*Dp7wqZ>fq?zQ^a=@k=K@9Fdib)wznrA5)#{g4ORAp`?6-#z|f_>miE!FP*m{L zV2vqleF+9LR!wf_t3jk=hca4c0%5TE@jAM0UPKH$*AX)N;-9<&6YAB{Ym1J;kbYE# zd=>3`BhYfqAkerUn+GNPXg3JAxTA`l6DP@z<1lh4)xjmt^Pl};Jf16Aym;af)^OYy%QO2CHW;#E{Z})sV(2?kmVNgPn!Hul za2{W_ZjKsOx`~BMVtRPcFHY(R=WzngSo(3>&)+coL31=~$`{5|E;m=5v_>rnul+97 zTtH-Z^MIHOKOw=PPf9>Y8AiT(@ExdI1w&hCM$&(cbTfr%#f-r1s94fO|X_ zNLTO~S~he$_zm+yt+&n39@#ah)BW<nB&3@G%VNQnJ z`v<>9A5em^)l#`SOV{@qP!JWjx(P-4dC_+h^$^)s@KS|X>oR{2oox>LC7!)BE^~4Z zXf42`z!BjAb)hA_{LyfYyB`c?TO7%?!rr57>Hfk-eu90kc!>8x>F%ON7&Y(-KeQte zhR+*|`F{0=3XAlloOdfwd*qP*y0vNlhAWs-(9dm54Vb5e-BpFSjbRu7S4^3%47tc1S9{A+#|&!f?+ygOBg)uE}&@Wc5BEijUH z;Nn~P3Ya|3)p+7jC6vlt`BbN(fjahb?AO(qfEefJ>LpjRMP4?B}4y3ao; zi-5aUCI35h8nw-p2z;`NM+LWwQbZ1&hryvWpDNn{rs=(O=|(od$ik`HyXulraB_L( z2cJiXocY#u;%+7)?$A2pU&jT*QfX%`YS=<_k2zy1GgbiqbAmPF{i&gh-0 z@(sGb3CcPv(7t~K3pVO=6A{buK2%n(LfYSX6)lcEh*`~z-9J6|H+*oFx-pz`2so$dQ*AZ^xxOXj#8ww;6uyVx`SzqeXAqG~}=qk?amW?DxRt*MAs~+pByf zp^%25D+++cliod5XSNJDqW065wq&tNLyJ@1|3})J$5R=7|D(rLh?FUmQluo2N`|sd zN<}F`ktk6jp@>Qml0qs(&RwH9p25So_@R1Vb!Pkw)kxpJ zbxUU}s?%>!PpGScqG#)Fa~->gh#rUiPQ79Lv%hrvw+H>hJ2YbTn?YQ`^MVG~X0z@O zuLs&k9f)`E#N;bEoqBd_}5E122(TKVjB1sMmYlPiLl#Y}K z6nqp6hmCojx_%igWu6YGJc{wtl-}~L9FkhLAH9k^C5TLajDS0KPEeT6wk#KZ0#$e4 z!PLe$1 zUb;=dsor^DjNm1On6l6Y%YD?_+qjzZ_ z^12h96w!Bxk^X_CR8tLQntnaX!sqe0y-kY-Gz_0eIWF7B!E*(gwKsmfKD`I#D}yf2~GzTdhM-F+0XO$9g>wmxW`-#bcM`l6E)@M(t z(+&dTo<+7HmgmCOE60K}!%LWe)3m)A&T9j0S&pN3=}dbVm8Pdzk+-`4=Y81k(e}ic zuMdgP3iE2#dZ;&>&8cbs&e+}uA8bW-{*vD zV~i){ijxb;a3@`&%^hl|H

    lEBZ(;J%r5|sQ=r4{ad$V&x)tylC8gb_XA(uQ!zdz z(iPrh5j_lNwxlv)^u(S81s#c~`BZuqsBy_H^>=r|ka6reeQj+RI4obc77Cge9#+kl zQ9HjzGm&2pXmaZ%L~XISUhiETd(zljph*@C8?kLg<>Hmv4+Z^E+gL8?+;dZGo_@zF zMz2dI;M$6c(!!r1%DBoByIKKsdwhO(kN)nz$>&P>aC!gxR8lKcg5s0L5+=SL&l9>Re-vIeZRCO_nKcc9dh;PzbQP;cD4!+QCG+-uyviX&i zs_tII*5SAKqS?os<@1QPRc-VnjCd(K^9LV=$bp?%uOczN=f0$y%D=Z}p$?&�O8X zxOu)_+}UsKcEI3BmT37AjK~*{SdEr2`Pb2aJE2h_uL~l2#M|Qe31~t;wCUCMNIV-Z z?v41tY8X^onhL&IPrz-WU6;&W!Q%}5(lG^p6S01FRK;9QnMEdVBrMs495vjx4cddg z>uT5*!d%#nw78vb@of0NYqCnMfG*RUx7X)2B35%>Y#HlDWL?{s-cSs0LHaW#Gd;%w zzh4||h>>98co4b6AgzY23M#Y{dJgGh{px1sORo*-{#PDJJnn>b><7gnX!$C4cBAQi z#4t+{SbFk}%Mme66SrAE6QdR{fA>K8h7` zyeLue504OkvFF@7D)I_f(+upy{P?i}``!ETTYwtOLJxk%wEv*Ph1Ji2yn=ThsN`h( z>`SGnFRs2^`OiQ6QwHl<2T})7*EXkj+e$E870LDb#aTORDd){QAmaU0Uc!tFte^t`XeoEj2>edssDXxd0V(w82-|7Ukv`;tS`0(lF{n#;S z9g5*Z%K37@(cCt%z8!?!TRuys&wl)8)0!WvR+_jlKm$bSS_-VUx*iI zlO=JvhOR&lbiL(kV}`oKKO>oY zAO5~Q%_$vbtVom}WA2F3GrUi?g@!!eelRBu4WQ_a3HN`WUjT-h`{|u8d7=M=&#S(N z2T-lf%g%Er_9HS$!0j;qSwvDhIKiHe^%3swSs0+cx`asQ{b*-{miJG^$=}77x1(gf z3U-5sW4!fjjh1@Nk`CD(h2fu{W&3L)BlMor9QE7esJ^b)K7x%Ts0mcLVG^)62Aj)MgPy4>@(3d9mE8jo@hHMW^ ze7+Hm>Po-83<%IcHB?q=lA#Np#eAy5Q#}U8?7!apvDy%N6_1|FxmSy5cUm?at~-Xv zUGw*Z%a{FrWu$p`i4zS}?QE8)I}KA!)B0Vz1W`9f%oY2q)6kpBf3Np$4D?^K`Lj;_ z6EFh)cBTw!z);jH9&=u1{u0k_u=$}r>k9(Td7XM^P|Aw+B1`uLzw5%*x0~mHDH>R| zo5UT70(36O-{O1onr7e%y*9_QRCN7AG34?eU9Wk&8c`H3-b~xK6?zk-D_Afd1E4+- za4j}|`JeY=c2}LYz=k{*D4sXYOwH!Svy|@Lb^N+IfyMsfEBj~(C=Kna;|uFV;oP!+ zs@a+OcsyIvde)r6yNH}|(XJiW5A*=S?9OlY(3v&p_Ka&M&~uKdTzz*DIx|-fZ1a8y zU3fz~8-;T~+w;^`z2bYU<#3s43MrhGI{^eNY;YX zU@w+;prk2$(vVI>+2i6{ZH##l`9$-;pt3&d^e2aRwVi?XJs*C%gcmQ{N%gnQHfO?A z@da7jq%R_^%Wlxq#qtCc>p2O^t}8H9Q$Q+MEew5YMN{*gb_2CzL1liYA)f72rM%j? zAyng`>Bu$ji>5^-Qx1znpg}ILZ+)|ch~l*B2d6kcvY&f!i86)NKT(2@-EKI>k0_H9 z_KYcEL_9})tye)o-InuyiO(Wn%-sH~X@w)SZ;YRpQ!aDzlVk-O*^IA;Xre6kv?(l2o!;)!ZF=d z(ETuS{eiM;%kRP0I{g;Q_mopEH%Y3cAX3AuTw&E)v}AJ7XOln?^ZTKl9A>o-iea+# z%bgqRPNMow@xOv0+hO=OM`O}CtgbaGy#AQ8$~V;6Tqb(ZEfWUgItv-vn7`QH|L{TW z`(-_b>$$4LUxLVT0g?D#4BuKIZ8MmaCQra!m1BSS2Zx9|h>4o~%zlhe{6Heqmyd*J zZMs}}(^d*~8)f`0p|8|)8LocH=}=1;*`fjj48EdsP1ey_-vqlmgaCnqg~o&){FGndNPZu;zqX6*$NG#c2JFgq7b-_-wP9C8T|3d_bRFa1hM#ygH)DxP@I<7aPR@ARBI@D- z4+&0l)Da=_?yg}nYS)BOE=ePDS%e2?LDh(?r05vMl3X@n8Kqxt(!y+rdD9DKd& zO@K@`s8OF%M_qe=r=R|H7A>8UkQ8G8M=Q*qG5x3BpC!FtqyFm1fI+>)QJ_k}?o6G}6Sr-V!Dgbj5T&Ps6}k65EmQ3()l_lhtSI z4b&-8y|Fmh8fbP??H^>L@GP577bVOEAhK!Bj7kBsZj69C@aoP!=?}oz)av!2RtJ%G zZH8uJ3!wXN{G8cy6za|fgb(N5LA^u2M&JI%`aZQ=gbs5Fty}g_^d!UWUg24XIJeyW zcnXok%(#vlgb{F?_ocTI?qT}yp=RCK!5={0abSRMvWkHFe7xX-TtDjN_*u1>$&4pL zS-#<;TP0XsH`%v()0^pgXwczt-W|!UNBFuzUmR;cv!a zUQsBy#b5R0aVOOJNmkNC;SFj$9>^EA&WpgJDgW*4hzDwj(zWRvQULn8h*SrG=O{TR zm;3AmEH74Xx$oJ6>U9`~Q`)aq%P>DrSi{q777+{m7o99q1%iQA-W9KO9^*~#QB!D; z$`?h9OH;9t#vQ2Zhn@&KO9JX}693X&w`bX2@{+h&kL}AYL)uJBLNE-7eVd$+X+We* z!B&1lIn*W8u}OM_73MB?Tg{4_L9wToN$IRJP;0J)M#`Gv*}S)1;hozHj8P5CNYxcM zD$Fj?^-;T~w4y$C*-uzyeTWo6z&(yKkDJcH*4rh^@zlK{_>mB`yS$t!Z!o=TY2@;T z%cwti{8f3kFn0Y8$HQOF7y>EkVApz=%KwSaEFHaw^)Je^ey;ve&kB*O^j?xqVDbK< z@dzH8*Dz}GQ!F>^-2Pw3&)A+C_n|}swa(WEu-2wvzsqj@^NVO6%GA(WE9;BZ?=IO= zXX1^pxH09H`%wrqcmM~kIW-_M)1A@Y@O$D6xyplg57ud95`kn_a`#m$ue*Y_Bt z@5vQ;FZiFI6L3d;uAY9fGVat%Qkogb#OjQ?3!boCNsU1y-B(fvB6c7O+Xk<_DNMdS zsCJn<9qADbEbs??rB=~k2RRg+?@x$%?r3VEYS#JNi1nfWTH5x}3Y{4RugXL=LACSiQ6 zEt8ubp_^hrj{lJq_a^}T*6Y&$yJOkDVIzjtzrb*+WFd)a?%{dVh4u%E@4@WYrn~9B zg~%N-)ValTyH5dCq$6Cw59@DCSy;cTW$ztK-kpgr-rTqmk> zI{wUcO5O*$^){qg_uqkmufmcSim~{@k4LS7Cc~*{i1(hf$~8$$5A^mLezTfKt=8wS zii!wAPko^ZKK(Lu*@{`ae;NS##p87fi`s~6{h&k4dkxT!al0h(VK~3xbzY3ECwWll zj)k+UvkyVn%+K@APX%D=wl+f6dtuUYaK1XS1&x#MKj|x8;pYHJw9n{XvKC?oNuMY8^%nF8{!WX$^9+ilg^G-aT z@A$G`zvf(i{a)ynjH%9*zl=x$AucRKf@tZ3!6~;HJ4CkJrn~P4J0c6N7Z9l8hf#H& zl-=vx(ZKLJwVGA$QP)E0UX4%8`c$CX>Z%E;d9BcgWY_kKZfQTEFg335$P*J7HV@r= z+?apGb#32g{;uW0{KP(M&TAhRw_x%(MRpP8o=;&wuQ<- zDk7O%QHq9aV75?k+Y?Sq|8`YLxmxdb1e*3BUj^^7J^EJeo|Kw6RAt<8yfuT>3H4@Q zQN4IR7P_mtrXk7(b#KcPPYn1B)fT?OpVm*oi@8D4W;Frmu^rMbbg_r_2L~I3J|BaL z;63`;xBXza>2cHThv`tE?KEP-SqM2JuO6rcO~Ygz?0O<#f*Mcx99o~v3e-Pc`)?oV zg64-sVOM^u0_||Qi%uVg-ywhG-Rg2h9|rG@lK7Of5t*T3mRMK~)TlECS2*s%@TP+g z$QPJ(Bg^*aj|a-3MlhE-x%N2MN7P=@S43PB0%O9rB(GX0BT9sP4Erl%D9hsZ)`?*H zO{gn${zOA@JPfF+nNO;(%|Z#v5W8E9t$F8eLp z$CUMC*4^hW!ysv6j)3sB72|08b*Gr0gAY+!O3tmTV1AEh?XHaEv%d(rkN2j`$9w^5 zyHFnrF^0}6*Ri!1um896oPgVv{iJ#IbyRmK-_6JFIi`=!_DzhISwixiLniA^HlbH0 zM`DIw{?q5OS!-V-=@FvWH=1b`1JJ%7U~gQoMwAgV1@?ku7+o+vP_S(Trrfq!epggM zUFLadDNk0&IrIdcDtaDt0BE;lIP73M_WX|=dIevoFzh&Sf3$81S}qr=FFc7t6?6B$ z#tvCS@8xgO-S3Q`?3MF+u1+PG`dwCh>2D`gUl3w_!#)W;8xJgwdM?kyr^p>u>7TLZ zD#=h+7lLKfiayIR8o?9U0k#*&*|1BQ4IdFWbTI z7v_%a6T$ZV)+dK#wTXDvSwn4}QfrvSnS2!vTh<$smGnKwnfVj6#F#ejXuPs)f2B56 zDa>K#bJ9b>yer>OwfFriwmB=uL92S^a@DXI5rl-f+!@r(#^OZdEXZIC6TVQh6X|)x6^oT4v zOd+a$gt}d^kfdSYz<;+3h;w7Ue2GSA$=!H3*s2>wgTpJoQ<7j# zD^PGtq76{mk8boh9!dDfmiNjRtpef*joCjP95+-b~__*II4(|+W;;GmCYF5vzwUUq@L z^WkS-*1ko_2OJX9*N>vx^Gdmo8nl7-m3K@_X)lVt?ZfgW6^m1J@;4r8cUi&ng=bSb zv?`&lZP}lCpY`|I7oepd4k?N@1M0J$NDgnNJp<(N2SK{kaxk&8GNI191sHpHt(H$B z82##ds^L3^S0E0(`m_E0ivGD+Jb>#?^0eJ?L^L$xc_MQU1~hk@Dzodr2)>_?K6VO* zM?)31W}BeSX@320cFs_vEp7D5>If?PrK0i6QyXe|kE{Gd*AaR28=EF^l7Q0@zuT9) z52#noSpTqK@$_LH-nh??|H(6(Xof%cyA6X5yeyu^EvS0yg|~vcxKMVDqSC=Kte*#C zTVYM262{L%k{mD9#a%{iqCvV%XI}y%(T%nDH6Qe*84DKGI0Josvdp3cA9Mz9soYx2 z%pV|9gU<6?4@U?%uWZ+{>+!%4JVoCfkMSb*=YM@M5MqGlZ$-$j3BYg}!`~R^cR$#N zmOjxhG_;y?bB@%fHPj^6fEVS49kYwZm%?QkRczPMB=qM{}DLH!e>x^uJiN znTz0udH0Q%AIj!K?Rk+;oCi)oiGvNkI6nhH{Y1;{*sU(ZS_hu~w+?8l0 zQJWb=Rvy}47<>|uWD39ME4@Hu?dT6iywlJTe`MK%%>$&VVqTH<87*1#S#^8op`}zA zR;$ZT(9dUtT6ALqMyasl;RhB(^4{}4_tg+|J=rAixXBoiepLIks=q^hsxz8vHsrvt zw;VEc*aVDQtYObqKSJ@&nY6t3_kk#~-Xv)+R)?_E6sYq{=QA2~{o+oK8-o&Q9tq2Z z{fIWtR(3u81~8t6{L)DXLN{#^@n1grqc-let2YGSLJQ7gFLAdJ&`xWO-Sd8nS{q!6 z$)^g@z_odXg*4_*GlI%=K40xcBo6MgRn_WSR$oV?*8(hDx}VU}a3#*~yB#8);NTrgn_t#bs7ptGPnPaH zzy1e|`Hgcq2>*YWj8TR?Ac-_oNXN2rP99yM^#fwt-&x5N1-VaQ*{eB%!sqU<{#xmqO) zle?ar=;4@YM2plkH`fV+>Pz>IU8ep6}8vW2E9F6MA z{44<)Od0C9^e;)Fg-fFGy%}TBbM>27b%qVJjmH~kB#NRj&qI7~auT5bd-FtQ)n&xk z@*sZIjg|4Jz8A47`c+u`k8W{rf76u}_&y*mEq=?>Cc@oot$%0K!l6%D*nYrH4XSsC znOK!#@%#aS;8#v&*C4m_(e^0r??7{muCFNxhWu%@N7uGyLD8OqO}pRfWBXLfD|0ct z080JwuJgOFe6jwQPMu@lB+&fZ5`XiToKQ_)INl^mK}{8og8&j6Kn#bNtAoOX|#yODBNrn%u&tP))#D8n~AXEw6)a zdqJWGKcf8L*5)MNMa22~IYoyI7){Ej6O#Oa#>Km5=YtjbIp`ByQ?}#LA2dMycuOY| z!%xs2d-H@kccaCC(C+ZaAOAP^h1$5=VZ3FYpH zJFqkfO{D0iTY-OID*IG&CO-k?o%<{kw#otqTz~4Z7>fWYa(BsrJq&~iu5BznQv3&zBX5MgcwT_I-SUGRiwaSP@r>-Ko)22;fB&StrUH?T^w#=Z z$9PT{QR{{+rw+qNRi$v<%S|Z6Ha&ijyB=ypZ<*wGD8i6_N@&YErDZ#Bt}JNi9?DQy zs^4=Ihx%WfVzH3c2fFglfO(lTM6syXGhe{^sCB3C?>)O&5@=pZW*Y+a@oX=m<^N>z z07=4d&?n(9=Za8h3${}((t>R?<3NAPEL+ECVxLrE&HZyQaI3P z%Z&-%s0!xiB{oIFSsU>vc~v-P?zNXNWW*u4GnC1L36!(;_Fr=`9$aE!Tl-ZRCz#f_ z#nRm_iyED`x;gA)+CPYaPrR1{ODLr!y*TDkEuKx`8Ty-u2gmbfjt&k)=zT0{wQ0%< zQ5sx~_y5Be+j}0yoIYh5o5V#O2Ao@Gu)p0W8->Ck9e+K4P&Dt3)v&bQ1$D`BU^ntP;I()kaVpn zBKsfiP-|JXvx3YB_6;$J#B$QP-)}`egMhm@o>?u3&BugSg|+?Vj)MxrQ$YAH%cxt8CtIAJeO2ub$WP?YITB(w?4smVt=u z)_=0TeF2RhGmXynYQ(c9$GT`-U_*?J^v0?l%uga^I|t7hKL?t)K$qS*jOQuiMO|9) zfo&`1k+jX*ogl#Mj{w7h<5A;E%>Klvofe$qbJ7O&z0voWKKK=A20|x1k|PjV=faO2 zl9y4_jezj;L)#H?msf;8e{lFssXnGz66Wg(D=D5RnHC~a^T7Si6mxx7X@~} z(~(cnK$;NIyZ42Nx;}s0nl!i{%Can)mAi|8A$#RlJj*Q5M6SeKWa9_A%E3>*CD#zS z!TeR-ZzG@vd>+&kya?Ts!SZ(OnTS}=+3ZJOKA*(vIR0>~MIC+Cgu|=zSB%?$-@*2r zZXnDaUZuG4@GC?y6R$3~g!Qc&cUH)9^K}Q>ZZSTZ+EeIfOPz1zy@<#Gwk4TgFg}3s z#L<*@BA7owDi(O3nOq13EDO$C!?1pyv*nGtVUE`^d2?#ok5hh&rh@YGb{}TO&!DQN zoWET3I!YNKEbQjMaF;#JA1-qF5up8`?!+OU86fSIWaswuK_$6;zM1WJVcN-f{^g5h zJ12HmrRe}WOgdMH9a+o0;=Re%o;yefO;P8$2XUPKP0-A09saix>l6P&arbtMVeI;# zHr;#j2YV6mMq7H8=?Z>tpuKn@l6P;}E+f^S{=8Zab(2n?5Ye9f&+9NbjX)~bznCuj zNyJSdJeOc_Yjd5@SO*Y%wu(k-Gv^2T9d0_ueP!Z!2)MOHAukd-fi8H%@yZ__sPXB2 z7#V@#uSoTIlCp$yL^frOrKNVFguPSdMjsvlUFe~)Y9od-W4IVu&FV0D(qO3k)V-5) zSRSI2OZ7p(C#JkWuZb&-<&y}Eb{^D|lgdH;G7hhDmzMQ^NBX;WUoiYZFYn`ei)|NS zq$$UIPj4c&59XwyVZ9sBT^||fdtQO5YWKPr-W7Q)^zAB_h^eXzTDo}Ma&4tPP*3vc zh8%r^xO@~)JCpxq2vV6QC>NYDIcIX*EWP<+%d-56> z-+4#x@}V^ld-20k+^irPc<%0PWQnkl`Z6;asNv~OoIh8(W# zd4-7~sE;r2LEiuH>0aC$n)S;Wdi5UdP*O8TzV9OXd(}Z{?(YPAs!9<*7^n#Izs?5LY;Xbk33hRd+d8Q0hC*-Y!(abxePJ+r z{r8Zsn}8zK_?0i^>OXpG^<%oN4))$>&PyKsgYm1-?_ZL#WT-z> zTYgh8{)yQ?1|! z9g9tqMqQ%UYK`c3uyJQc{@C<|3O#|v)rTLH0<|uYCr-Z@G1Ny+gh?Ml!*&Z}pk z=Z>AW}wBr`jHMpo`pkLT(fcl(+bjNt^T zEiqn=N@kGDe~&9D1G`ilL5I`)k1 zj_ol+iwW9Bt<9P+rZ$}+n05yG2yPFA!ts!IshFmE{u>ItmpCE5UgX({q8SaeNMlffZLLDX#8v_ z#8hg%NhjWg$zHyU9roI&h|sWa{mcS%roB{dw68+d;dh9v$IqZH{B&W_f(+_P{j@Fi z1LjZDogRG+<-3d0?RdgnBP_?u3LWas?Mde(Q5|j7{1Ms&8Yhj^|1POSL>sn)epakd z=+3xzddM2OT(d&>PL!Zd#n9q+?^B^K`@rbS$^oc$3O;>o_hv*EzZ+@n`Ui?}Lx+$5 z!6Q<$okyJJZ?u$OEg*ZS8Id|(Dpbi;LhoqnsfqGDC{hR)5UOTdeosD;_6r{o$qT=E z)d4{*IvMN?KYmk#iXqprT= zADr|40R3L$)+(_kz?tTq*uote1oH!%lk z2g{=*r`v#}b1?&V$N_rHWO+=Z^`UCyw3BemJosB*1cMnSZWVRfb*x+L;6aq#FXesx zR%3eKGpaBzx!0;2{PyQ2F zK6yk*#jS8B7Yq#=8l899iaPm~ySZAh_$Z@eJAWiW9eP~;dIxW1_B(_juUpS199dyB z?BnzNv>PzA>sIn7Uk1FT@xS_QiOK8Y9h}2`yOXHVbd9bO*Iu;1K@B=)bqpmjp4E0q z^+VgMu84`@D=6QY82IfdeZea#`Bj!JoZ?uU4noMEp5=CVSQA_3{Es?gbxFi zMgC}kBC{?LicG!dR{Jy0Gdyc2`vLxJc0@kkqt%?b?fhUudf(oXyeDD8f9|3;6Y~o8A{e4aPh8X;+9Yo z?j1?LsT@YFzPSA?HbI>QhS4gna)|u8pWEe*838AE`O3RzF{tW|PKn_kM?@NmR&9>L z>RtM5GRc7(77)ec@tl+c*1x!XI#lF@BcQH_Ek{n>`ltVom7i{hn+5a!smm;VNzYc+ zA(P!T&5-6J)RQ;m8CLue=s@O*%~eA&+~u8|#DDs%{kd#`0-X`%()*EHU5^muh=lmP z?Z1II>X{rGz86*3S9H0%Qh{z0zjP*P5=Pc6{vEpAj{3@nrn*M3I*X3k(y#Cx;~Ayi z$XFDl>ZLwzIccX=> z&lio;`(W_ow^QQ$?Wl?(s3rB25B03NT>!p4P}a3uPo>oZ8hBcw(yxRedh1WmUANc& z=llY7pn%xCo|(5pT@Tv+&SWs{3IYx%Tk^E07f96gtM^2cp#Qm|!{^cV(x3_s8?ed=YRek%^W{m|MBQNP*ua`9Pyb@eapF|qdgcQ|ppN8SvK{h8olX&12{AmVwxsll&_QM$FqB?> zzw_fy2~18VJj56~^bna#XZ3o13Jhu7Q1@ZHMk&_aDryZeh}49QTc)uK8g zM{S-}8&ieh+;3-&cRmIhxKwws`YikB(Mbi5*-`38{Duv86HtA1-tnr<`(b#O{S6e;zS$h79&xN+Yq^7rIzILj z4AB;g88IPy{<1?9)DR_UyrD z@mT%O09*Fl-Z4=a*gGBQqQvw=A;M38@Sz?J^>F|BXh2f}hQOiUm$M$A{?I6uwCh*V zQcV7s&-^@^z9#)-y{G|NiZpnib{3$V*u3z;qZR0_HJa&;!FZhL?E9bjQ>bmx^W*8i z)7neYO0-zf7T?kC0&TlKb6iqY0BXuH|NUXRU}nveh4XiF5NXQkgkX~>w)xIuYz=M0Q~tA2Y{ zr2$bMx=gb^!tiZf_UF&4_niRx&6uyGp3Tte$iDHW7PcPMU4^mt)@(-;Mj?z%qa85r zBsA-7fz=b!0)9H@T{i~W0iCwf#dfIE|s8MK%z-uO6|7f6dNLK(;&O}|~Q^GMnnE#`X$zisD0sg+;FdIc_{CG_Zk z$hv@Kzv@o8)85-?($OqOmQxvZ^Q89WaOeSX_#)2TJQ;dKkE(mV&_N`fc4A$*2y{RG zo*Qjh1T+>Y=QQpz)H)xlO(glkg3Lwbm$Y^m3W-xZI~f4&3Q5YAZUInON*0aH5JI&M z`^?(Aw?XH{(g`j73ozc+AoEl8IHHVEQ;D`%UifSFj*2fx6;WKR2xTKz8Ja?Dt0xFk(Y4qMV6eb^T8Q43z3M~e6zSR9?0Ymq+Y;EK)JR*6w z(`Dc5S77lptMquqx@CQP%d%>1BqAD4#PYd6h0fRCx16D?K<^jLi=poWfyV9`8FwZh zsA(q-)G1;9S|z`zVtrmQBBeju*qLkr3|w%nxB3~N3plLq{JVd}{_Ltc@tGEkfIcF2 z(*MZ}>QpO`@U1)oeS%42=Qck;A6<^LnHFRH1bS^>S+lIi`Z^F5cYpKv%FF|xW{twU za_baKPKs7V{QX>v$SK?*IE5Ab<tIWN31)$auimA;q3k!E1(z`bI+)oGYq-E4g}R{|>1op;mU^r}ce?ZR%)E)fi(}%IOC~c?5 zt_S;AV3PFg+J;C2H0?6DTF&_*3@@DBd?3dNiUs+037A+xf5@55CBz-jTk`yLR}nX~ zm0J1Laa@G%)beX*xqhP(`7J@d_6Hzy^&dswbR+1~*sO-z_7|x1plJ)smq7fi@iJok zB~ZP%-ffLpflmSIvq+)FM{9vTcv(xFxCwReiPF1i7~X%eLpejTF#^7tID8g3n~WCc zE{fhz6i2Q0zmxj8q=3%TvUw^Z2WV~Hf3pmnpzFb7|HjjmsEiom{fl1(==nM7C&-vx z-lvq%@>|~mh6}pVBom%N>C{@D(#3hy5YgbIf3FK@>17=BY^G1+25rzX07xL9S4M6NP0AUygnH5RoJ{h>CVaqqey9r;ckWKzm+`aAGzdb&Oql zlj&lB7`KE@U%$-cg(cuPgd_Y%PCy6e_>TQ7@`zM7v2R*!2=Y%E?q7owg1)ECr&Ac( zP(XS@7Ww%Y&yMruh?lDchJnd%a#}OA{_^JbO2_cKlal+C%FqCeY(DffUgI~kR>^#v za$?3|mhF$7{8}a_;inDRB4gAMXl(S|>CP-bo4Fh^bvFXfhWejcO6-EpN_LkK^UOO{{NfoE5<$=myI*)KXzJIdcU1$jHl>gaKQfp&eb*%g01 z^ybKei;?%S{zhGI_sT~}{Q^?4B(2!O2@&IZKD%rYL3Jm;1q944+Z8>(^MAdqN43;Z z?uK#}U~E?^(bDOIwheWB?iGy?qvV$4VEh2OBICzTY}g2l=b~Sp`WQp+k`<>oO&V2q z&1<}kY(r(GVVD12=LCAgp^0jRmHCfBzdKH>R}KO#$T0W$y);Dj7<{cY&+L~D3=vW1 z!Z)jdTE2MHzqbnw@^F_{QwfOla#}lPRtA%|`&8|>M*D!S8Aj~>g7vec^MC95^q?B` zm1-Qe`-|~64$f86pN{)LPt6W1v)34o+wbNnTelY&Z);W;wJAHcW|*FIl#fU%Qa zi9N4wK$SwT$NAN5nBOCzetwfi7>wVh^~Km@c`(x1>$HzOmzU2M-wkQ;SpC>=(w2s_ zoib>t_oTCK0vU$I%}I`?#!z1Q$%T^X42=TD*Z1BYL?gDG z--#EIRo?i2cmP{Kq`!9pXluqRuitFM^lYANtOg|t$R9gia@o8^GrmG8-Zfvb`KUZ^ z`0mdl^hKAS+4AfujQ;8zYl}Gs^jrHwb+%)8gW4KHB5o}mQC7{rww5$S{clo=u6$+2 z0a5M4LvJrvEU%lm%0}7S7_aznT<;5xn}?x|y?dAVepj?KcInBZOw1nXcr2tF-6M>M zCFeJ-=XeP;obQ*|j5VlB>rW5w>QVG-_F|T_BZezm=smFcsoGUUslQRr`E%8>ofZz# zLbceu=QrJNXqZDC-F)_Y8JnPQi>1|>KW|p((ZxNb!N_j~ulI`U`fsEs@vQ$0d8F&# z?>S?9rHtL$TS6_}QOBe8sXA#G-m@dDKP+KrML$63(-nVdSj)^OLbIHWs_dOo0&a_E z^DFh^P#}J~bNU-5cT@&jX=y+wv|VufmA{69NTWlZ&Ihd^KSZQr?Db)wec#OIs@D%$ z+PSNvKOYBLM*S9z3U#2j(e%3&Kf@UDL(R^~7Pyu-2%`IRq2s5+RXX2kRB3ldo-D!% zjBQHK-90}c(&odL%u2QZqdIHft4A!T?!d3sZmBX*;-^eJ)Z^U&3OnVyLl!iugv z7hF)GtNwPcLR+8@^F;)dq@ks?{SK;gcobPLTugtMKb51T3?&a3d~x;#c5ul;N$e=zGcVIZ8Ignx{d z?LIDby&JWNftLoz(Z<4*Jz~mo-+1DEDG8s$%RW2&?nnR;FJyFG>zF%XG|L|-V zzc4Qk5+>m8*5r)xi=doU5x!2=Og?m=TpQ(@fWwGl#?U(ctq?j{f7fOzYyirX^?}RP z=U{Z^1iz`*R$!QVOG;^S!QhmRhnl)9#&WoGY65ftj9INg7s}26CPN+-vVVT`gtZ!SV zMkdQEb~QBPvUJaao!Q40Xy@-gN-@txeNG(_9s$Y$t2D3MSJp>mrU2)Isc}MkrdqTv>eR$+1A7PIIG= z+yTJYVfVx1>rq51Kd+E(kO%{Y7PhjqPpH0fTk9*0YUmqF;B~3P)|d8bR`}VGDPZIu zZnw9eMU;7Gk2GopqCdYFY?LqoWs+Kq2gUU;+WBGn_S9XVA86RAR&WdTbk$QX{o;r0 zRO_1bn^Vx4ZmsnD?L}0(Jzd?`Q3kqBPw&D{976S*cc<&gCPV)r$wX)REkr5ych-K` zh$xpsG$vFY!u;o}R383mMB4jv-C=!UR6|nrKFBrz)OY-2#>o*d7j-D+`mYO+6Zu8% z{QmbaqdlU_a{&j##YSdNkBc$qadiD@ZaW$iFAc5R#8p-=(U$#w*PmUB?J&)eptdTE z0)=I{I*Gy0{x9&e3ExznTZQg5O3=ec?M0TNJO1qJpP90jsN zoiLud6B38$H(7qMK{*@Q&Bk~F#!l@H7k{7!ulE%<+v;OHZA;bO?|i@GVKy{OHs77u zzYnI3G#b{g#_+2Q^3GkE&KFTvDY;yY{2hiTMWlC1EkWzClO1ZkgcW+AN2EH$Hx$Dw zbV!_#JNbqVw3f9W>`FxckA8xV=JT-hjTs-tREYyAes$nYb$$tLvZ2yL#oww`%AYWI0q6hSjMs;xy-ODy{*d*TYZdLvE-| zHlO|W%^W-%OF}duN(&9{+)LektpI4-REBnZ#rk=*o?nfx&|sjhKM|P;R3<(d`omI2 z+dN;QYIWMz3*Sv)ILE#9yQ~o+{|>fnnPJWYj7MjRCCf1T7U9{5y$YO}5eCbeur1TK zfidUSqw0#i_rT={+ra`Q7+AkU(1*JSrrV;09baSi4b3nw^70N%R9f$xb$0tDlz`jx zjDva$jlFxE?%i_@)fjQywH;^nyGHN&Bk_fz`G_nNK02Wwin_M_Rp0pWGqn9;d>;-Z zqukF`;WU?0M2>2;xxMx;485BCu53RJwaVm<_SQ2nG-bwe>&aL2(O?~SNaiyv z&N5KplkTiZfDu~w(cPg>fw3eMbWXeoUOD+_M7@ZCxkFy10mD)-cHxS{rL$I;T$z#B z|L7KCq~avdq|LXh2=D;uM7*+?kP=F^`55d`#^e`($;8BwqKEJQ zQx9P0WumKrVZ=Z7;P$YaVR}ryKgdy9%{^4M252-@n()>vL@In`kbiv=jmYVfk8>12 zua7Dsyud@7yyREoX23XhR7#c`*++553Jdd0#sj0mE)^ z^Bet-FytQJZhI2@ed-xA6G2f|=;Z$G)M?}n)SJqjRBjA++I?Q^K_-tAO0(k)IvTJ6 zri<-dRn``v=|zc>n9nLO5ORa_>x*m@6`uV0ZQWr4E_W?kmj{+-=yW!_JGbFEBCbw) z@+dL}k-yHp`x3o;KJ`RQrCg0bV^gFqmm?HJyfLB|vVRy+D*BJEKDD9`8??BI9JhFU z1`Vic%ns};LZr2`M2RhjQDZoMyH@HOpl=dnxfUmZszob*g%C?n_v${~>cC%IBL#`n*ffT^MI>-iFz2a}s-AZ`AljzzJr(eQLc5 zh@WXi{L@ART*}MRl)ina|FVb3>MIz|r@Z>mbMDF2C?R6~cA7L3{|>YsV>io$pNPO1 zmG{pZSRpr*`T7{tp$Vk}>qev*m5AcqyI$%96aRx62#vui+>LfbM3g$`|Dx(C%s#CMc+ixCD4*CkEn{Sv*Hs<>J`=pK9bsR}b)V=DIwXy&g41bpNwDo)eWc2n-){i&5Zw|AA`XQ<*pe2(UR@Y?rqMQK>dBa z<)9=E&-O=p%N5^xD87;u9~#QLa-KGnDLkNHyxY`@wH!nWx~z3_%U49#f#eqiEDzVm#r{y^?>$60Eox@3d>%&TGUwEej?cJT$x^*8j?$=kS)W|8d=g*$uhWBr;c>8@Ql z^mx>_?wN^3OV7&p5z*S|E5lg!_X^_`xKx3rU11}jDX?N+kn9!IsM5=R^ZVJaY{|XQ zzxn%>Jo`r|w?Lk&Rpll0+_2{7Q~u|COH$qIikkqs^7;!c=lZbUi^y9askRyDAEOL6 ztj$B+jM{U-IvDOTuVQAO;#UUHExgv{AIVuUUWR=B2ifdoRBgZYw7$0u>O4RF&YWb3 z$aku0cBZi)QuZ$`az-L*7;)w95c!LK#MflZb^Jk8^MxbhEwu#PyQgQ?P;fBe%(8lJ zK?z2Ul9y!49sncIV)VofHsj)X^Kw!^3JHJPL*JgD(orJ$^S4$vQV$NW$b1X|bTv_7qF^my^$h4d5S zsOkO}#r%U9FBwimZYqqY5L&!`JU22Mh1%M4b;lAfAkvbC*k5VPFDGB*Px3i@0>X6If1s_HNE$z_9Y%4F|QCFdlq)7S|#Ho!`hlJ}MJ1+@8QuevANg{-+ip>`7>8 z-OY(Xi@VT1e&OYMjv}DCJ9E-fPoRPQFXN@#pDgR;4biQezM-y@6C*Y(>rt(c`hf$o zE6-JOfK;sf!hf%uWc2bVo?3`TWZ5vw)E2DY?UccH(tP?p^jccy$CgxE0?uY%v}Ys- zYTs~3|J55e5QZr;ET4=|p0AXSSiUpV&` zH91B-788|*KFJ*?$}aFAvd0ZcGfD~&>#x}^-Qa*>H=U885e%=#Si581@~SQ(e;Ccy zp*#fgxl>B7{j-6=%|0-(*%au=ef#1AN0?K6GjsE5FZBIhaR1;qhe(mdwDCQhs5(uz zcz0nMHZQTsBAr2iNRp?1RnKF5&=athC&uk6w85Xp=`2`&X-bo0{;uX~pgH`w#J)>% z#rzjYNF^o+Z&^MMET%qvb%QYz)p?u?R!=}Z%~`%tssj;KVm<{?hl0X*PiXL!{U_GzW%I3^33x9E&l!6%*HDU-B45V6_VmR_huARC9yL60+N%wN1cUkK&m>T{*3jzrikt+T&^M{I?|(srGT;7V$C63f? zaxzlU&zKw6&az{5y=_SzpI?-zL09amT62Tj&?EVzNix6zCTcQMhH+TEENgEhQP_}< z2J$|iUdK@aH6sUpNfg(i&PE;in!Gj`eH6Lx>M2R6*abYO@+S=6+upcquO^JRyBNrA ztA&Lxy=`OFBmZ@O*6?Y6tSwB0QT93&hoL^yEOmJM^t1M5eURhP6|@Ugs?_QDcg4f| zNY+MthNGDL54}!O7A_{>UVDDD*_#23drkAM&c{(#>h;epZ!o-1bFRu9 zap5hV<<62O%Yicl+@+68IWnQB(d$H&(a{Y+t$)-0=YSIQd`uIz=+r~K1tM;_F?K*z zlh?gDT#Opm9&@A&bwUv_UkCBU+m=A@elHaed`Sb#v z(R;q;4?m)85BtKnE`x~mml9&LFuZoV7tXvpuNaC1y9*@WbpySlZ(Xf29?xd9_GQUO zEYCEgt1HRRJB4bG4~V9?GS?H&CVX+tL4i*|4rDxR@WUs`?7C*(PuRQ zr=tBfjcW=~vUv@q@kd~&_2&zgUF(2ykiF2p)&r4lENHmOU51pGtn)nf6ht~yUlaki>%@5q{DXUc%Skc!$NJ5T zJ`jFA__ZC+eu7vyxfkneI`#gDY`EVzo~@-z%{}@kA`@Ro?TF{ZvvbD2i@h=eBNwG) zO7y)6IIafg?&%(wYP4H>xlIz0D!0>f^mgOf7FGR8UH4Hln^cmnO)bo~tG{1ecbC&y2cO$Msj!H?Sm zGN3h5rYi9AcWB}0DhqGTLBx`@=hK^ZBGT)}Z+mciq3|}lH2YZ#R3fr-TVP`ZA z3V$TG?Q;5Ag?vf|UK!d0w(jeMcb=H62l`2=OM@ND`%c&H$no+oFswD6ne+H65N4~T<`&OXUHw?wmlQ)he*#}E>++T zph16?u(}{6U>qOP>OQ6gBQMYQww#j#dhCMB*^d1{yL?f`?AHdUek8luT3ZsRw?7*~ z_*bBt@*0IE5TR1p0E6FNmS~h-BjAoi-c-6H4z$AuztisiL}NTc zjhpkl5NRu$Y(d2>phl|Pe{A#-#=S3eb93K7U5Y$U?w-#^o#WlBrq6T$gX8-?OaBTq zZlihmIDIuxH=S7ZeJdXn_i7BU+II`8-6Km&xHrSd34tnof+lqB8~F9U7UKsZ8Eww` zR4M=yML#c?zE6hMa~nSn?{r2>S`lKZXPgN*zMXLo>+GR=!)QVI$Y~(n+z_|P)(M+8 z!z~LJzgD2dZ%x1d$OZyUp!JmmX%Ue>-F2r6VR<+D^Cj*#@?Jn=cWfKeVnrhXViTKA zF53%i(t1s=Nl@{w#gu1M1v<_sXF0V({k@zNe7_BK4RNwgy^<0%Hj8{2??W6e- zwMpDKOl_%B%2G^6#L7E&&Tlq?lJZZd;*6@H@x1oNtA`_D@XzGfHIrnZW%h+z^7*0A z?h{v&3eLkYFWb}K?2XXM`s!&{<82rY3tCn2O&vwoN0fvrOw>bXN zL-c$1N(3Y|5k=(ul}ksM`B8LVYosQ9(;YO9J9GH^AuLbp`8>PdP^jiV<(z;s%=So< z>w#hIe%UQ~#}RqE)h{!RLg+fR@1>R51e9z$zi;CF5F)=mwd3S5b#!f@vSDKM0}8g3 zHyUg_jmR9=oqF!M0QG)B${Kk!Xb)U}@mvt*cXh1ZKh)Io5NNAC{1@0=(O=p9zG{~R zA-QdDOTGUs7)zIM$!irQ;BNPaO>zeTLv_Q?aZh0w-fc>ZHuR6a;!Z%%DCwHw!%5UBlXpZ+ctzhKpw}l3jc3lm z@G0|Ya{L~km9^16g|`7i>AmNRbN|@Ar+9Ovj>g1%#6m&Rw4_ud$2bv32b1cpn+Vt=vdHa~W zP*8iH7Md5!K%{6R;UJ3@{g;8!@aw+fJ{~B%-#J+W9h4eDI$5T(N z&*_r)B}w%Qx6$t;i&&5075G@7X6t?o4*sWZdT7?kvKO;UyNR*EHT!s>Mu(epjX}ib z^Z7pW#h!OC{O5<#csPdZ=>1e3u4n%e?rto%5PqJF?f1FTj`(vJKCQE*p}aws8<7nv zjo$W9uRK<|^_b%W`t`XUQ9IrMjp6u3@Yxz5USCoU36?-)cLNiKIo5xf=L{4ftsz!hwh2Jvsrp(`+Gf9VnmYlxzkH)DMG6=KwIoQ;da@NeBezBLq>FWbfU@qNQb`%xOl4a2e8U_>q? z6(0Zj1U2m8X5aj00Zqu9+p*^jhF_p6-78mIkc2K>6Vrp;?lASMXzv?83hLGi^T|@e z{OdWD%p1>+#i9xa&Ytp_VmvNPDd-;d=-Fnk43mSK zg$kZy^~`0@2)J)belTb>ea7-NGu}gBIjeLdL>S{|9VzIULb`2G9K^nAcWMTOvXDn5W!YbUE)jdKECf+bX{_}*?6Pd{SYGJ;8Gt&!(ct!dJt+0v ze0Vza0V0>(R+inmgot)66gS^tm?Ajt(-^e?^3B}MS@Kpuw+RU(wdVnSRO7~BVj0XL zub*6q6^3V?@A`9B1Ugvz+0|99BZ@W0>KxS@kb3JFILTphMdr&o`8SXis9Sbv6Mtwz zapGQjQXp1WPjdQu{OWE#s5-0SIqibsyWd+Y>78LmwjascO4n`_N!=Medmbm z>;X?M)WP`91x10q%y=|vE!Jr&3B&vv+Vk}?#XH}l1EWHDfl;sg(NgB|Ux7xgh_>PH zd|vwo)OptTEOp$NfMbbw7+A#W3i?z7IWFFNK)`KyQqJY|0^07@AMen(gW5*tejWLH z0=q7mg3B_VKuZPqhGQc%V6eUGw|;gJX6yc(^gP!7{PeKT2h4s4_cO? zDS-owRBiVETpyrnpE>Z17mF8Bx0(-)OJTU;0`PCv{>X|*pOft9K3>pUl{#Nvw-=H1 ze)`$;RHJF~Aufc#l7?8Zk6EqX@T_{k`w*w(e4CXes_hO;*Wfpb8R973Ln}**dcY z!Y^u}si=>K4V|p%#s1PURAH-c!DB_kv$O~bZECxPXTP0# z?D5DUp!Hbsb*Tjb-T%1JI3*E%dAF-}-RU-{-Q?2B@(`=XBgEB#>cJN2eEFx{$(BHR;6?GWMOA5`!2QxdOR3WbOd=n{aKTBj!)(SdSTZ>H>JNtR=#`c@S^cvR z)xNMhr1ha3k=ZYVYlt)LAvEOM5$Zab4`bag)}K@Sf8Z5et0eh7oC|VUAH2NRZ3$F= z@@3&I%le>I%qDU7iDi3LEpm?%AL=&wV{%1*6HL?XHHD;Vp;m$=-udDHWbWIYF1T?K zk?d28KmS$0v$bv8R8x@yQ#r$5`P;C#Ih|fRBbA8pf8?KymJthR#IwGWa<*-h!n0Y> zlbu35p-=PZ+udeZ{Y}d8_cY5>4#0RNW0~cD62?OP4A)ee0<~)AW6Oeih`;!IM;_%n zo^Ae)Q;S{#0e9(E#qg{o(5O#8PQFispE3SE+duZgh>+qDMb9c2e!-ntuT4Tti$(iK z_VvPm!&Q~7YH^TdYc{<5M;weCH#xaA*$Wo;WhF~0tVgxM^j|-GLJ<9()L@&7O`$w#hRSdnK!MP7KDN6POAxQM=3d$aasJ#;5zi4jM@)rvh&3H^mSxW7j= zA{Usr736+t4_1ftTEG2Jr$r#1^#$xA{aNu`=J>kr7OUNTfaQlLTcnCxl9_$Zpg-O@ zx8$%ns^I&`btzR3rd+*5YgaMjeW>%I)uiJsZOBs~kgLrrzj-?98WsU3eq?9vaIl(^S%)!;7bboxo<&3^;x46} z#QG1@2<7uLS6QKFOHAVJ1D~T^djy1f^(HQS&?{o>#b@NFW9g_12-SZ0@ z?`-fELIu`0xU&_ymC|?)lwLgkY{d${o%uV{ zaVmQ<+QJagn*Y9z2{x{wQuE@>5N5s(#z$J^xb9iNK+ExVnap-5^0zkDeSn?E1ed0S zw8CAly?JNTis1tU<&oA1?gp-!ej3#!U&{W zFzcRyu|fL`d*4au({AFbp9ll$qDqRK^xYNjGdh)e{;049(2cVntkGTe*ZXkiwPHL8 zxZFKXNr4T>nUp_BU$_C}ZledJYs=>%Pe<)Gi(|k@iPJc2wg`R72Be2aF+Qt)cO?Rj z2$24*&}4MSBx+-I;8IpD1?tYsZ^QXDKtG`QMy13U&nj0k|71-uFhm=o)iAy~v;>-s z&H|~B*Dtd4Wp*B_IjJAcH;h5M*c!RP=B}D!_xFI z*EY=lj;ODFBxjrUTj)Jw@I1gF6Kdw&f0(Ca;n_5TzaDy|4MRSh zh8^#Pplc7uE{jnF<|W63*v$0+G+hZskq5gL*`=X12{Q08v!4)oSj}3OVmcI9ktB z;sCV2c-|U?AHX2yW^i6>gzE7-tsP-lzgI?tq54BvUSL>rqIkA{cvEuz=J`%8+hgK} zd3~S1L$B_KwVajaFkLTYqL>!2LjGeuI&Sqi&9Ld*8+@Ytuo^k(lBjjGAsGQnp z2}O##4?9l0L!+#4N-$h&#r;J}CnX;0V12KrzuPYLKH~;D=|$&lX^hvQDr?@o;Kl{0 z=B^=DB))+Go%Y2#hCViK={>?1BQjv@!?lw_lkx-{`@ry>q{S8ULy}$?{<|i6#W(`> zj&!ck<65>0Ht!!H&7hGzx6S#-nf?i?sN4R+=h$Va2!5*Z@pld|_V3@w`F+Fj>7?q3eJ+SEs>QMUYWHHnVX<>yI)|sUOHIpPXCHNLCedG98 zB8NxyH#(b>HsnEhf|d7h;8j$o82Q;H0DJGU(Nt0Mo-r61ux!gDFK?p z`#SOGQ7~0LuX-;y2k7-#x^H*!qL%rek%GZ;XlW+Yso~l3{?Vz}NFTTZG#fnE!~;Vh z+{_%x+g<=8q=ke|hXoj3lM&Rs^X-cFV|W}$2wdfWYCWb_*~Jb)hg8D&(LckeX8t#M z@BoJECX1f?HTnaqL!zelML!J1`o?#5?aUg}I}1H)wO{H-k%6vm=H;>r}x0nZ*@O4*xN`=9Ui|Mf3g66d`kw)Zk(NTjt~8pHHnb?Vx8MXl(+ z-G^s;BO@kRR`I_+pOQhSsJ*8Q#p)r3xaN& z%KzK<0mj92VacU`>LDw>j{Y@W!Fz?-%k9s{##QA0*XJhO-u~PEG84y$`jTeP7>=94 z@X`UJz)!pX-;Q&|^J%HSO0(5B0A1^7V`W^@|91Vd9-`dusj&P1`@8?Ud-L@BjfipSokgs8RHZ0d_V%eEP{(*VzD#?=@CFBcf6hWg@^`%PB3~Zp z0>#_U-f08+p-RK+LgWAIeD#fR6`QPCv2M%$+)!UBw>nI8XmoKMok0}C$9wW31n_J! z-VaVboJ8$Y>XWbUh5`MFQuh7uG$6;Fq_Evo1KO@6H}BeIJw~-Yp{-yBv@;@2?tNIj z5#z4CO4B;#xr>^8V;2K=V0|8jDy!BR<}>-aff~vEb*DCF@3imPe3jh~%TIMaKeHxU zMsvk`^}cFWe(S%2R}VYKP9B_iT`s@k^RfQBCKW#Kp##)x=MQZLhRgH)m9P6zdr<`2 zr9$x)*AEulQp`xgT@nqTKJ*Bgdcz1;YH z2+N}`d4!#>I6h3kZ9cI5w2v7uru(C|tbGo&;&Z-J#RI6C8?R+#dK%Bp@A!6_`y}%@ z#qx@Ex5cesK!_hdePtHNa~-S&(mN1Ev8=9fEr#2ne^2VD3l6~6Nilz4zt0N%GxR0C z%-wQ62K6bfXNZ?bp?Z#%grSjCJWI7jc*(cpXmMoi;>B@{&$jiz@ZWLGAK3rB+{y9F z>qAwaIb4(~Ltr@&Gi~;JE3{ntq_MHQ7*RgzEnYkjhvwQdgSJK81Jc3ws8^v4)l|o+ zhFl1UQ>5Ur+B?P7EH6r^nv!c^Sc&Yx}oC5{aAkOXOK_`$LUf5Xw05o<#lyE z3|76X-?DbgigoIppX)q1rwQ%Op-US`Zm7F%vw?(GI8+~(x%qu)5?W8hoR>9>faE8d zO*7xI`|{n{IDMUEVczNb%h1yMXmFv2kF?DSwNE%Xn|hr^r0-{Cp9wMXkZ5S9^SsQE zATWg8$2aX@MRg5#>;@xCQE9;VP<?u&@@i2JDBT=i0rE#$^`l&dS_Vaxe$3^w7*dZZ@rBg(p0;qy4S+b zM*E)9eF|tY|4WMvKZaYT;7^%sx7&z%3GOZO^NlcEduFE&aR4pV&K^i~#&F+@wjC}L z`~T4wQr8l^1^iK8+1RZ?uNG{7ZzAg}nOuOOQyp_sH}|64=6cado0mY9`c?GNM0RJz^rB@HG@V`&We!H?KydsCVbLHi<&*^Vdg&=`T@hLz&Q!&MK(=^=SCP zr9{;DI4e9VVw!;aQ#vYVbp!S9p6@nRpN8()1H(Rfl87X%@?=s)8+C7Gd3xb3KXko* z5P8@P>;FUx-|~rT=S@VKFBCeoK?659q{=yK!ys#*v z5z%$LQgClpA<~aiGw;4;qFSCUh6gq?^8jdS){nMR>nZf~+Qs@Pz5`f%Rp=1gFN&Eaj)R|{pGT9PMz#WVA7e9#Q*QZ9l=TuBeqosGQep-B3+=tBa=1^9z z7f?;3w1cllEc-`(?^chQz#{%wZV%teZq+%Z4#btk$(V($A#vumPy#3z1 zJJkb`?X4}kb$n3Y+V>83iv$cp-~t ziSb+$UeJP4d0ZIR|IEOUTeaq9J|RRxMM8y!-w^466-$kE9wPN6&IcdC0Y&0n>E@VI zh|~o8#Uh#YuXt9z!LJYc9wLgbRlkvjB6@YE?M`J9<|mR%B9doj9kaXGFZO zgd=akp`~Nf&j=cVsBTa+WV>7xbSCdDI#Q4ErZ7r1d4!|30o`iV)YwZKG?k}teW|k% zm5s>ulx96bMCZ7}?P=#>DB@Q+=lz?gqxv_sdiFXhdEw?|KN0|+Jfyd5y8aA?W}Fs_ zWN!csH(7T0J_Q?BFzb8X8>WaXyDN>_?g@R(*N>RKE{C4mYx$NE6`*(9`omRctC#II zwKZF<8W54fxW|898PE1mGEh{(4Hzuz-%1sIfKHVt5$&TZd26WShVk$|ivEfq{gA&>NCJ@wN1WxER2 zzp=*ZO{srB?Yl3d1x33P?@6d!KyAZnvXj5IAhO?$-`eqRc(!9RpVfN{VXA*NH?XA` zx)m1&-`uH(iNV#id|Ftamg&x;H;*R@0^OV2g?{H4s=w=H^up^hB5mIBw`Y7Ko_!Ni zZBoMaD>14`dcp&f=gJhIoBvt%zv7G8Q(qoLoeyusx!enc8MR|m;;q}z9Dmm_9&2?( zbeH&>|DhcP|5|7DWfIF*(2#04=c=PC=51K<;1;Wg+klpx{LFsq2Gq6xY-^`l59$~^ z)<;Ul&RY`UVdkdI<#^VCdai<&%cx1-V`eGL7%du!4Dfbgep%O>A6%j`ZD@&bU|-wm zG?=n~a(GRu0Wfw36}M+$e18KUq9+B_vHU?_P^0h7I~vdizTc^8`ykoz(hWAAGSqaI zPx1QcVxW7)2axnnLXYVQwiT6@Y;vm;we|E3M~SSk>+o!ThKD6W z$^V0&m$=2jjE_Qhxgu4Q;5Y^4>Z=rg^N1FsNVlZ`VV`7Td zHz-RBCEfHkS#iD3UC&A_4;9pvwQYy=WD;t<>EVsDdy7bJ3nyFeIUyphC^=~N5%eXA zS16^h7AA$guJC+03l$x*#vQyvcwGII79*|)7zL#zFr0NTW5d!bSUO^TjrQpgBDKw{)LdGJ$hS5L{h1BM`sZY2ReK zbqcGmf1P`1I?D{}bHFhByCwdrC?*%Js=1QQvv~FszN4hoLjO6>Xt?;J8A9Zp^*eLw4LoMmHp5c!%oB~m9!@|f}Hbi#i@tD~tgBk<4S#y7VL7nFfSoS_C zfiV-=`m@8+s9ewT!z!6y1RP7P>9Lph2{;+qFm)?>A zx!VprhY0(pM6i08ssrcv^vDe>##^T~I>Q!y47x5kcdDy6;@RBaCcaFF1;#dqbK56z zh^%*H=jX!#P;GQPJS(6F^;@NtQia11QAFsEEh!w4o$tN+rNra`LOwoyrUDBtQ2a1Q z(~`%MId3eEFE{dwx-OqPNt( zO?$W52&@WR@m%s%_Y=N5GZ49P^hM$=EZ;>FI>Re!{s~HM?F+SfDFA)#nPUQ^dFZlT zwXT-~!~J)dKODcJ>j1@XZt@9ysDR;jVBBR*`Q+1DRPA{9=>>sf(C1Tl|LV?XFrmTN>^kvJo<-() z+lKnrK)i14q$lqQ4SsuCesUj0q=xQ{*q_!g1)&Lc_I5z~C>Q3KfW>#IXX4jJln9`u z?O)=CB#jVx1EELg#j<^Mq^LNmybX27r4`kcZurmnVc!2C@<}~yVH3~>{%qzCD@LU3 zhYF;@W;DFv@EuZ^G%DIRJajPaHlDR4+5bEr2eisR+2Qft7ti)a{Bn)LRXkg^(}&|l z-?07o)JoiUr3N{=f`kEy_CK;t)kBVZCt2aT4~kt*#do!Q-k&% z#c-fLZVBNp7Hw9%$6$fO$qf(tU@UUzZ)hsUlhmm#)8WCh1&`zWy=LvCEYQE)=29Fy z3_W*}*atO?VVw3+;EUgN)IsbDkSUA7-?&kPm(K7;Dt#h1L>!Q|(Fp(9(KQPk`a zt>-v*njAk294dLLa{D@rcyO}F>6}Bo)Va2yl4d+xqxn0c78bu4ic7p^9jyh!2IA5C z`&gGcoOwr<8%xVW%e&clMTjsWrG<&?aZHYxpD?b z9mDDm8~z~D&zn126qfhltU`cP@$z^%SO>3eIt*mNQ%i#1-B!rM61w>%#TM&dJd6zW z>3gvLu5_`VuT7QxU|eogmE+PW)cN>oK`q-6v^Xu9aQ?RoWcLesCNVIcZQ>)eM^Eq= zqCF&itg!9C<0vKJK51A#m(oiz56dYRVW=`|-^AzhFsf6nJ+@{DXdlEaHy1N;2Z&_c zlOgSP0oot0l61b`474p&>kuy^JZt~K6Eau%fI539#bO<^&JB?*Sd1OBv3hRC>N$Hp z-ys-svT7?_9R`(q{M=UUl84%|pT*GqD5pU_0(FvN$e+{GQ8&-^_xrByMZ^n_ zEG0JH!0gp1{r1ay55T)_4{?p32YS7_7L+Bz z)KTlJJ)uj`6BWm$UAn@aM)b{@8yb{if$HdIdPPMKQ8pagHg)+gB35ze1na0n{WxER z++{8__Rs`n1PP&zSSf+$dQZ`kd6!Y~=mC_vbwK9XTpdgV1WcN4Z%5VTa;aHI{^7H( z7fcbK#rSM`|91HVvtYQ+&WalZUVYuEc*i55h_orZEI=FMyRN&X z!2bGNE~*qM&yUNELB#h@b6dWz;HyR*6<5~0USNaKBWJG3$uWNq&wj8+b|Q8Nda5r~ zvI(y2)6%j1&Au^zVHg^{^7Y+D7NCaxX$=!8#J+#pmwL-apMYC8scz!A0F2sQsG7wY zN{+{$eEH%cOr_Hm&poYzsUU3+(Q_)89GVu+^ZM*s@q3-b^!NG_9+=)3o;mjLitK-0 z2mLBV9gG+i7)$*0DtU9`itk$k?4M=PS)rpNTP-Ww4yJw;r%YdUM@vv`S%LD9g#Cj=eAMr z0B!r?R~KK5$Go%8RAbAJ3OuXkj`)m)T$n8NQjfVWkERkzoJT)eq0!cu_I-UkDB^YC z%qBfNTH+Ux-s2{X%6}NDpVKu2nz@zZh<^w4or>>laNt42v1e4Lg+)X@%5`pa$tpBB zYQWM8kAT#t6QmvS6}44XeslO6hdR4Ft+-lpp|5t(cHaI0ka`@lURjw#8-DQTsU5tq zbn{TMkgqi|^7uAcrti7zA3ijD!@>ijDH+tW6+WnrH#}ws^*7L+!{7g;-AClyxSu|6 z@ThI~UcRktDX2#6ZM$BeBepJSn+gy9=tjjGs`fb5Ss>!k?TKt+ub`QG$h+pq6taDq z>UICtS=6~Uazl8%3ecyNl*6c(-;`U^v+N zt5S(JF$_fhSofwsMGTSn%Z1H+?_ly@K!&d)m8jF&z=VxsHBh~gZ)RXOj0=?S@VpR? zYMkwj;zgPI6VJl0(_~qffrzljpm57J81fNwTQzbM5hJ9xzBt8?XW#z#7Ctr@sPWa2 zo*2(5(8~Y*nohGuRRX%)Mz{D-hsyyrysRAL9c${|DS_FO{errWlO8kk&;;D7#Lu~! zdCc#}oEv$x-hM%(b2z_nDisxnzu;8A5d{@mCRt}xv3jz;GcB4)H+&Jjd~w$N7#kw( z``{26*a6hOP-;luChUCKuWNnb#vU}Pch$LzpC6SPI|<-Qm)DcG`Kv=qESh}RcWm#Q z0yHkEH+GZ~gRR53ypOAI16nHNzIm=;`Mq=ZpUcP<{Ll5m zxJkjIjadA$@7e_^l9&$kw61mgUM!AkayUgc+nM4yj!^4lq(jiu_ak?jpB63aSD)yP zGZCv>l5bP7vBB!{`wvT)v#)3J3&Es^(D#jc+(4h2>odv5a8-2t^{u*%)u?m2uzTP4 z*N7~_p?`Q6#&b{In;ZC3Z2@xM%L^U6ppNOYJsKBvbX?Hyr6=oCYW&fdTJGzMYVAPt zPZhesX#&ICCxTqx<|9g!?;)Ft<3N!ylPmv8AmEM}ziYtqkLcNi*N4qzdUzK7^RzWK z%(xMnw-I|P)*yq&jnOld<-%Y(|H;^=_I%Vhr>^m*GzEyVk4PFNOnHSx550Qf^hg5uy7w0cZ7x>U-tLf66m3IOF}l89YNXw$y7CU+}0X z?z7YqcRX5Z9(!)^<{2tUJ~Vn&B^kBTj>Ln07EGB2NdNhLjet{*IC4v~6p^Zvzp3;4 z0JVL!o|6DJ|J4Vq3$`e(gWtR9OWVVFn4dSu37RUMF#!6u+8ZX8t}w9UMzmQmvo0F- zM~t@;<%XgE)Un}RJ}d3VX46cX{a-;~C|8~934eytPH7iX@U^Im<7RNe(KEmZ)fk`t z)d)-Oq@UgASYi0rCc}XGXQ=Mc?a)EKJ3wxK+e&v6Lfs{$zCnptyofFAdXQn|5bB}! zU+LcX3C*VMeW+7`;aU3zjsN~mnnR?P?MHna1X1T%cgd}5u{syhYj3T}(}uuMX}?g7 zI|K9$wc$ft9x!$|LF4rl+w!^5Vk;cEf~Nx0AMS6n-&pR39(I!{8Rc<2>tmDSV-^@6 zd2_16&i3b6{B$DmcjUrGT|~MVFLKg=h7t=~!k&k+Li}TomTgC*@oby>4!$_alqb}+ z^ZlExfjW3LZ{dUeyEw6ZBzE(l=v#~zjwJdg!DCf9ObMyRJ?zj!<9oJbYaiW1z`f^* z3Rwt%@?G<9Tkft0`n;pt6Q6g8^nUcy!_sp=YiCz{aTlv^C$4w;o#9}INOnp4dUiUZ z&Vo|Ief5t}3O%qf$#w_Oe#>}XNNxx6r#OdgLFO>FeK8}9>z{hHi}c%3nP*Vkc{O_E z=s83Zyu60TX$4LIk+NdAokL9!nM*9;y^S$c&pBw_P|d}@R~8-DI%I=qITK}M_J`@O zqgV5mi3d{hQEhoQhXLg@5YE^+omew~h}TEFC6$-$jmN9c`s`Ve7hl#}UW>ObVtv^c zmy9UC3cjGu3o2}xuY7nd-olMjQg`4#2*4`EciKj_1CV?>GB7>8H# zK~0HfJ1mSx5oO6H^}Gd!!zLGmG>b^{VdGbg)$WyJ1-joKl1A7tCa*F3S&KeYq9wkc z2TDhHQ0pU$p7m^9&>k(7&^jRvbh%G|%DtEE{rh*FtvK+&Fys%aP0@nxxZ??aDv~f~ z!y)!#)e%H`{D7q~XdID$6<+(ZqYcJ4YhHQh+lW5B+w@vUgt>1~Z?ohNX>C4&N?ZkMbXHsAMJjQbe&w4wi=ywd;5>^igN2c#fY18IepMwkas?hcP!^`|ZEZ zqvq}6`l9oSh?L@dJ5U(QtIQFz_vI^jpgGm3V!I7Y`vvv%E4|SOwq4fKYmSE7xS%;7 z?i~4}i@-2%s1)5S3gr7KyiuyJvGua6n>)591m3+S7Jd!h1bTmew#Uw3{B6A{=B;|P z9@H;-WQS_V6k7Z(Su#@+h^pkjIaT&Of$`%P?;PfpMr|KilY@FZQ0L#?$0vjv5H&|8 z=eHvuQrw4YcPBYfwSUKK!v>5;_~W1ELvW-LBFXV|wdN`V<)MwNdf@{AhHQMCm*>IKvx(!-Cm zryF~qw~!t&nKcPS)eofxUzq$@K=WC8aNk)5`k&q3_CCaP#lA_k;CXL%K@CQ1E_u1X zz~azL{If=^s=Ct@;hifJF> zS*}+fnVHu?B+rrTsjb-lTng#jj#tP0_L9W>;jH`dFc)w@uO;Il=BIjmZC||V3R6yZ zCHvC?P{YCXy(~XvP+Mx~i+<6kB2ZCuL<*bTEBTc`lbL=dRE&R(@G}%er1=w9fPa?PquUzrl&{q_%rj!P$h~}%GJ_^aZ4#BE%^nsIDBK*ri8?JO^jVZ04PsZ~ZVEDi8%p9@Y@U-^!vXFKJ@_F(>4 zjDc@|h|4kPU(;B#c-H~76-$S6&tSZ%MD}Z?wPDQuOlVwO!6SW$i9baot2Gn%`$tia z$;abEEG00km>un_oB`e2C#sZBWG%~Y)%udJt5C;KoFc2L0iNX>>x-c4m!WM=Xj^dg zlNE9?=VJOr+}9P6RHTdMq*m^)Hoh6|*PBBT#q|=9Q|=(5z?OHMM@~ZB$xx$~?$4ygaq)`yFh&`8%)pUpIu2ElSsqIi5p|*}D|ZAx?<$$-J;v zRt0)Ixnph%xMA0ea&}YZWPz^Im}<3A0~l*=OirJ@fvU|n)zuN1@pn|0s{VF8R|e{N z;pS~OZH?N0mia%Cet?KA!rxxkYanuKA)~^KSqDeJ8FW|NRW^Y7dS%mNL1oMSA3i%r z9m5AQ><@G|8M>o^QylB}Z{LOL%+-!$;EQ35)jO{<4CA+_-m0v-7-Wh_a3XG>2eWT7 zno@LGtc%`&$oSo5YsZ{ms8^`N9-rxk`ldWtPNwT3qI^L0XdYJA6eb!*ir6Fw#7atI-ye zx37Bng6)VMjPbjnHi!|tn0xF)q zr6H^{2a9Fjgk@G?`Huc~t)FQ9JqYa=GHqQ?X@LKCMTd`;y4Za1zyESMXg_rHTza1q zUkq)m&6njXMbIGs#k||CQLvaflF)r%Cy?c%%~)4scGJZB@=US}#{WV!?G68-(~5>E z)4Of{*kJM}`Eq)AAPLB(rAN4=*`ST%;q?jim1roZbGKjj6lymUqaJFhM+2N=nNFEc zP|tqz^mn;bAl^-XgNsu{<0kP?x#b?SSPR5D zk{0SXI(BRyAJrcvE?9)4`W^hRl zRYJhYF=nvN^%M82j&zPuMZ5$)Ano{scO zAi0OHHNAuJ1XkR6OZ9Hjh7JeEJL+3G|Mz%d{)lp<9;hgey0wfx3uZ+DY0bH8^NqP! z|6d&{ydkXs1X}mS+4N}CrYe7=N2CvkUNZO>d9lluTP+hrhN)-J;q}rKC$S07BL3Q- zFAIxr8Cj`mvmqbDW9#0l>R$N;!-pM(*XaWJ5h#Z+XYsggeV2}@KS5JC_wcMX6`*MD zT)jpl5~5jKcg!UH560$%#5p_c_|7s8H-0oF0zHfjVLxmAeYE8#9fM_2y`$&Xg{}i1(PR^24Ucny3U9kND zYV4f!7QJ4IN>6=v{G?iqlv_VJmywHm1F;#W)6cGL`Gx=jV2Og(9 zBC<|viLV;XejIcwIGurLr$eM%A6#M@AIf@MiM#k!)V9vbspR5A)H+s^-JXc~t>(pW z62q~HOmf#6dq;bXNd=mG|53!puL=U1LMlb%tkIN0N_a_B1C)O&GQK%1r_7o=;C`#? zC3FZY29=p?Le=lh@($eI0|eZC54UgQK&(~F2vS=MB+-lU9j#o*Up8g^N*@Pknjqg4 z&aOfwW4~TIvSM)r)129^yY9^@(3aJ+)Xx+AErPrst6fUM1j9cqh&+>L{Z1R&~XCzCnYnr$x8YB7qRHiF$k! z5tW%qyT8ikL9GrE6VE^VfZE#5tSgQ&(DA|O!&O!`)c+{=q~4pQxDqHmaQnA+C=C(T zG~{3Zt&7^D`saRmX&|5NZ&ycdxdt^>Z4*QHe9%}G#}|*&(?C}6&AjO%3X~xl$Nf+9 zOzkCfHXb`Xu8fASe#&@Za|wuwse&i^WT47MnI_tW>Fs?!`vQOKG3onDNSZUOYG_hz7$|y$;4^F}>GWT*x*Q@)tT{OD7|7_o3{pWPxeW6(A3D~^8|J6$@XNkq&Yh;rHo7-9TK6ZA*n?b;g$n^OOm!Cj zdmoTiWn8xiaz>HI0vnKiHZS~9Fxmt}Nx_)=k+WzxvcFv^ zuMSEGtHmyiIiN8qt;!su6{zob{3cNjAvDS}E&FH07D&8F`KO!I&_J?x;K(ByqHX>p zsd#=Lw10Me+SbFcJJz1Nc}r?9q8+qo?EdxueUO_^)cnAY7BYA9pL+5Y0&ksNyJl<> z4L!P}N)ZVJQck+p<1hqWBcG;Nwlc^0lTQqp2=4uZQciekB(}XlEnfcLuD{_0O5w@P z4K#+n68$Vezj(ksNDA84 zG-}|ehA{Q%+arhE+d#i&u(nQ{z22qew)-PKSPY;^iHFX zp+L(azZ_VN`4PwF7UKCY+M@1TYLU1-+faP$IU98`v1R3r-mO&NXBP%F73*f7&CLPH z++XSmDnFi@BxmVmJ@;!Eq`~{w#J(=xjrix#QP#XO9C>C?vpY^b{VqMJO2E z$%B5aLG=7Nc8^tzOSOyQeT+um9SALpPyv!({l3#m!$64U5&Gm`1Y~x}bd;=v3~A49 z$Gw?{R=elh;#b$A#i!EQ#W^3)@I|G4U%$Txg72|va{)m#Q?$zUD`x?;uA1D|v33WV zYF9UPDaYb7YCaj!+;XP6seaf{)$$vADfSs*4YWFBbXUv8XH>EB>C3*u&=6ZAIeNi){z*^<4h z_v&GC$M@rE^%*B1_J3L55R`<76CN6Ik{3|_-ft=U+r@yi?^0fi%6cd=$T_{?vjGh3 zu&6i~WT3o4kl)St7dGGFF9dg=7wtq<^L<-S-RngD#bKtF+pxIK@Jpfzo+VsBjH-Qg zDHq#khH*V4Q}-jt5oeMa&%FcgXtMh4ccmes5bv9ve{z8$ko?$0`yLt*;HRXG3j%RY z{?a8bO;i-Pw&0bUJtn6HJN*yuTLfZ@{PyvC6KHtoO4Zes<0x5l>tXv3+YzOuULlQz zIj-!rjPsV;JiO2m-7hY`szsUY+Tn|}nQhR1E$qkfXRe6)-kfD4-+5?X&m+@RKZM3q zkIDsR_(HqEY}LX0{fMejCa28Vg{>nm-tw3S&w)%H3k8NfURYebXpp0{-25zk*IPb( z^Q8RJ{UBsI#Z*epGUc}|JI~Ztr)^#hVtFn@XSPXJz2VL$ui}^G1Z^mbL?Kc$x$Ge+j+!`FzoX;_2Ik? zozUGNef#BMPlmk%!#*{`sLK1ewb|qoH0U4hGj;7e(AZUv`Kas#%C5i`;jP&H0OdmM zNoVmv)FbL_`)oiT`g{Y!G8b+rT6f-gHj)7w9?{wEb^;oK!!zYz6Y7m`Hz}Lk1mXM=tQ$A1 zK{eK&B_eE@;zqFV^J4AYEwa!cPB$!5cF7Ke3DygrKnwf7Q2Dm<7DO zgU9VMRMg>BUG}|U;}AvtM|sqa`eV}G1TsNd+qR8IBC#umbR zQ6TJd4f;sL)=S@ji5ukNJpWe1abYItRB?;RCYyT07Ku7_R{{My$?EW z5Y7jP=Rvi`@bKu7T-5Jg5NTB?jq+%p`hSJmF}1HfO(Um?y$ou9vTbS2VTvpkO(yosopT7NCjDHQUN7Jn(d2FNMo zfcyOFXtE+z|3=JJX?B>=>G>ry=@Y=O*a@Y})n4Wi42yAP#c@dw1erEDMMPNS;i z0Oy+x|HLUB|CCi~LqzB4MQo${miX-OxYIszrv^UaamH%>Cg(!`Q+{UgDrN5Xu)Oj% z?a@W471>ZPgdg{IO~d*r|JP7`<2F3*GF`o;YZvM{xo`DW1BSk5C$F$#HV}~n{$>y7 zVtg`%e}Ab8Ut#hyL7N||^RbFt)UC65D1yvnmol}l=J4R-bTv8DQ)F!(G0KBzTutqs zm&#F<8?PC^*B;a_wJO}^YdVk$@&D(CW(tk$zdm`ZIN06evM7YloG}Ut{q&0p~?6SteYArS_Bd zC(o)pWa0rvgLAQ_Ke({{DX+M@u68RX*Ryi^7o&@f5aC6c-;D!1fcAIK8@+u@cq@py z%3U;AQUo;_wpis|T)JOEJ@D7`S78uSdj@!8tI|0bauammUP68vbWn)rLMwHl*hhVX z_0vF9Fu*^MA>50a=1;aZNFyNBxQhi)>|peo;ytfuWGzP_wyA}nrTIYwMHT+-NEmg*QwJ?IN`T~1{ig0g zA}S%R>dH#Oa3+Vk^?%;m;)asTgBu+<9HF&#m}`vymKQ=!<=^n;syd>&l_b0Q+rjj6 z&s;^zKp1XKs!x4JK{S~^4!7+zQTOR*;lkk|sOO6tSLHe{Aj;py`&8);pd@t{{S<~LO-L$vsB^mzD0!c91N!<<`zw>+{D;`NI&4MwX`nm}#CT~> z2_s%aJ@9G6%Ad_ZN~5j3l%IgUyt>5uP@xr#de-@}Yl|a-VEWqg;w?Z3f3RSI{7|>m z#SdR&Fr16oEq>P}LZ-0(sTa`%oIjz-9fDih0>)8M(FV6oNq{h)Cb>SJR zE}OS4_dO6fr-pZmhoRC*{#Jd#6d+FVIIcN*AJ1~^rGm;;JygB+YJ}9qQpUXO*(7#3 z2jvgv|5dELhdQp)2RR)mh`93k%?5{SsC%0t#q{Sla8aCI6LSl@2O(xjZ?tFaMx)h+ zoF^&cs9VB1KU=b6+4c+*4|wbh=R#c{<5r!UXUZQy!X7P?ko;) zf`#-Rxm)X&&NrC&ylXW5K6Xz!M4|+9?&yNK_9LvlB};L>K;A1Fbus)0Lr>wGee%&6 z=;GRc$OogKK3|M4s#^_@ySA2JDh-Qkm{_~=spD)D9w+i;+wn9kA1~ToE0MT96;a=q zcG&PQp!_a}nB+bTAG8rKli|Z#03AYV(`sLo(8%X0OY6X8^9q-~Ym4O#@cDxAzmxmW zr|z@+(QJ!rN0i4_)aDl6PJ4F%$Van#UhPQ4;|x>igh~%IXApbj+ev94inlr!TCNAu ztKK}Bl&xqcp{F_b{V3#rKSLC|hv6?IT~7}3Bi{pp{^%Bk=dYo2gZPn?r=BzRzo$)H zRdUEtv}2R4pBNseGoJgQGnZ-J>ROeAuj&l}5x*_(i&O~eoPEnPBDexgxrdcGQpKAhr7uB1{@7mYC!~R>qo41O=xhC_T-BN%_-**ilOI6eVP@^9PqJ2an4d3HLpvi^r|>mv?4g{{AQD?u?>(-B~nf zWWuS`g5k1GJ*Dm4*tg`bM#BRVf8AciLqZT2tC8y1()UA){I4|B?;&bNgy*UC=4fa* zwOGc5V~+cCb*!JN_4C%w~v zSpS)RT&ECKX|~ZG%3*m7gDR{}zr!=oV7uv;u|{>sxKz>Ca_Q@`?IgM94D|d@9>_QC zF@K@=|241J`9JPcGofkpe|;a@=eS$5$A$aQY+TCJaGDot8_PCd*nr(LlN?meckX(; z^qV%E`&On2i%S~z#8==Rm0EEZ!)i)V=P)~FN@+}4FUXHor8&CjR>_dMs=H(OL>r1;%qZy*rg=N#TB zzYn4`d9%65lbDMm@WAZgwmlg(`CRm%zFuTprZrx7X$=cP^qm)KGi*3Y(|t)y3`yzcc%Z zhl9n_{L-fytmSuI=T(5R?p+v+9@JZ?j%v#tyi4j>S>d81pcw@^xnWsBc~d?0`0 z>zzmSM#g&rwZw1onw0C=(Dxn3xZeD6F4x3jt>oqbz zMNz3!e7gGFeMs(6iqCn?X#eNoth*CfK0@pL>5XJdERHNBv2R9g{bisB!-dbr`%nej zvzHwR;~}e3>)yGbAOV!j4I?A5-e}0)dRm#I-F!6FT}%Ej=>Bc!aJ=#=5X<8P*7x82&-t0H?R&SWn+HvL z$?xKP|CQ0Mns4z1aVSrWyR2!LhHAdaG+yXAfff$DRc>#|hF<&Vdd{7zQSMYvnpyBX z&`-ZQYvYy&v>2KB=*Q}KTm$9yjfYI}N)Y4l_xsKyA|hNjG$F~kqM>)}A~$imkk#Zq z#bULD(+4y|<6+6S#-+&ASx6(r}W|<9t5(5r_yN_b(Hsp^kwkkBQ!4W^ZADu z)BQ6Z$I0ghg>Rv@=1aO|EeAv;vD`{%H%IwLELG}!=nOrv{{$*(M^taZjSqW>P!(x= z=ZK^W8j?zEbXms%q{35O`r$-0B3UR!JcjWLkhTuwJG5J%(|nJ`!vh}!!GBH|&NK6~ zwZ7V3tHhEIBpZc{m-83VWRzUU?`AtFX&9O;?7rk47zjYLhlX6~ovy}$Z8Y%b36-(-Z+!kQ}7+gGuf7Vnl_<ZZZYwgT{&c^ysI5;26{-rSkBEp=wsoM-ly^y z4a}%ofs6qjm%gQZB@5;k?ri8|U>de|`}21dZwa+&X{8 zA1V~=1J)(3#pJ|`+syVkW?yx`I6FJo91OGXxom?J4xyqKitQVRSD{k1q`~9|ztKED zk8AGT-Do)Un}Tp0rayw{sh5%)Umo{oxim1QA@@N zD`#f~ruTYg*2gx-Vmwhjwe@T}j9L-(!&iYPDW+)n{Hd3kdp9%Y`Iv%6J_{Q9J|~eJ zOvCgGcA3sDHl{pJG&;SH<@CjW^x;6l(>9V8g4e}Tvf`0gzlYTylQ*~nCA;j8S}QN= z>}|NkvNeRMyt3k-+ER~}^jB2panF1}nhoM2!fg9Wu>6RU@c#kGkN)4^FR#COr0y5J z(}sG(k-Ma)9E9G#`loy%7?p|$ezJ|&0y!~T=MpcmL+JK?7Rxi2fnwO~qjV}6lm9hO zHm-SE3b_fN_lCYCApKkOypK+DEIl84RMp&&jm0hZ(sm?FW$^*oC$IEW9711fJf5Bk zJ%z{dUbxSF48yssj`UENw8nVUyRJMwwoxn(8X_+|4E^;O4F!KXwWf3}RDXJx2L~`5 z(;lszf9~vC(p%BsLQ$=xivn~sl&(E%{T1l(5AMoWF6qO{Z0A;=UvujrBI54dIP(OH z|1SJ_pZ-O4JpF@r8T`tdFP9Q1pA5nVo4Rz{uOT@g{2Cb(q zGz9)N0^+T$LOe46@VrDgh|xp{@SA*HW>d{SJSb&fIPWSZ<8hjMHY<5N#N_G->#qZ6 ztx)glsW;fZEAR&#&MSE zwxj5;bB^Fj8_ZuvtmeG&ULqYUZ=_>wPQ)u9u?5S$C3^!+%sAd{dOZ*~cpZwob{35o znX%?qEa?|`oN&sv?6nm@Z+d2+uZPLih@7s{hPg3@JzLYi!ib4CZrT18+u#n!DPi$< zmEA)H+lfqiIJ2FgZ?alSZ3UX0zjOU{zdfRocAl?apMmBA?WCRPR}nSoV04$l(mgwx zEED{&NxvUWg;O=Ja6kLs<80J5G-g;L(nW-Ipg11jx&&1-bhj(qTZj0Um#3|uURsLX9ZBS1CeeWph zz3fx*i02>vwjuI^&h!6?FOPc91ac=xgrh+o@jL3N2FfheO;`O~{h+ssPxRuARH)CN zVR^Q<9gkZ{`u%jo01a(Fo>sSES==$yuK-W^qM7rv1_OTWmebQ_`hOeM=1Ahb5crvs zY@*>+il(=ZEZhdcDH5QRb`~2 z^Cdlwp=Vts$1IL6D}UtSa-IAD#`sC@{QO<^3L-r>QuPv2fcD0rt5J#n#MOO2&Gtua z4vj0-#_XQ|4H=4ORSj;w0%G!A_EjtQ!?KJ*0CP;Z=(FAqY)mPbX+Y#9BIB&5=k)BaN)+GVs9 zb9mOF(XkMRPosaBo)0vs?qt8LfXcj-whO#dfK0K(j(MUZ9_O$naCcrR%1;RwXs;Cp z+ChuTr@JwGB&w4C!0kdmOztk;PcTyaM_&~;(Qv8w1B6>o7C+}FAR2F}f9J2aOV5)y zE*-8p6NhTV&-r^;#-a}0(FZ!uo&rTj!E{rN3BtR}m7MH)fhga!!cB=7uHw&C_PXz7 ztCoIuR7R~xwqWvy;BjXxEHekvf%yD)xsuga=!?X;edqiHbbXOkV;`LmHMH7!--&fF zSwv8i%V~sx1mTruxAZ`#uQAJ*eLgBXlxw)Z&l+`$*h&2T`v{dKk7`q0cLIqT`7VjZ zf@brTq9!Vi!`L8S_vw6$XQCS5jdBlH9=MRi6t{-TV}$Qi9sSR^ z3rwE~QLC<4TsE(Vd3~toy}z7miJHIPst`OM3}ZWs=JHIi_+#S5((8Wf9iTHVV3sp! z9rT1w=xr9o@K7FAKI%K!X1i?p-CzHjQ)#(C|E0gdrtB*kY}5ICSnNI4Khfg{*Z1^6 zEq?@Yba)7+?#CG!FJgEl?OM?rYb})js~_+q^5V;WH;liK>QYX+nm_UY zh8NW**|sz6pD43zaZGO;@LNBxG+h8>+rMJ(FJS&-D#eIBEK(Vg5Bid>t-|!0ZZCO{ zJF5`tJ4MjFYmtbC;=*^g9jIiU5A9xknv*3|y&W;%UN?@YcgYIF7dn8DIwIHpWGVg- z$a2kLugCYo`1|1G?Q1PD{b<&<+wy`ikW8xDy3)3w`s=}$&V~q}ii>Y?NbVYRj0hy; z$s{3am13g{S2BE=eXlx~C<#-)PQR$O;|Kb?I5lbfK1jP#7=F$<6y?RvyERD*qJf6` zvNIP&V7O%Z#h>S6i1s;>D1NVh*?GJ#QLf4n%j+-yaT0 zGQFRa8P=#@j>VG|lLYjnNKetY!;cLw>o9+4Z7an|{|8gOFmlF!SY#cxkL0(mP<9#| z1nTHuu7M=2MfH&C!r)k78XtWv@UJg))KEJSUuutv=TzASpL7ni5v>^3sAj)joq$)r(4Vt;^|t+kb25Ne-zm&kC*1CapL_k$M2cIyfsZy;o>I*tjiyaN ztorC+y5%nr`;vLgq%9!Kr~K3R!$r_1$)Zn6bp~Sl;+Ns|kw9#`Fz9_~GgkgvBSWKS z38?W)52t&I4>n#dIF<>U2{dH#@aKyonpk_KPH6@-4Fg4TBU?oD4m>V(|83sH@6fr< zGtR{{g-KouHhj}!_5B2-Upr16+4CK1_eZnI+pkQpa<|5eZaazXPgRL6;o*{XKsfK& z3L|V^HoD2UR8xmhkWDGk}>|#;?#qGcADLRd3}Xem+UYg=GfB({5JsU>FKG2 zx6E-q9sI0*Lw{aDJ^z)J3d!AQ%z}T_HGe*smoZH~IJ)e;0rYd6=7_r5g|&z0K*hsc z4IsJX8;r|YL3R7JC;ee}@HioB5=A!y1#G@eQEU0ZT>o07SiWn_`_r@-UM1?e6_CeA zO=j5oFgcOuXdcY}#5Arr(H((l>wYVboPAuk%HyVcck_BYpP>$orL*I zUA5nhQSc%E*uOHKKz?;KTXCNzjCb7oJ!Y>1i_}@ttDI)k@<=T_>+>Y)E%(~c9pn%7 zBj00E1Us4Qw=_2h^;xo0;T1(p?f4cZoE7Ht>l#`3yYd`R?9ADp{zzwfK2XP7X?PRM zqa1ZKvzy=h9qNXKUr+i<0rjzo^XOtPL#}%c#T9P{5-I=D<|Zcn1g0z>e^lVc@?VLS zXUcv}vjJ)Ik;p(wHk3v>+^yNnbl?TyM zky(oG#v?#9-|y((U|in1jc0i(41ZK7(GAx@l9=L&)5= zvAn@(*m&UhQ4x+p==nnCffH*rf8=IveVK@kjN9D#N^M(Jlr1RP^%k z=>VHUY0z#YE8I9#1k<8+#)a#hp(dkS{JY<>@(%PrqaA}GMCNv$zMdcSY#jS-tap1? z-42F4vu4Pu$HI8aJK32xKcLFK%|&$>w_N$Ck?qb&-@esA%0K8dA$A0-_tJ~E4Ln*v za(^xCCdvbJ8@)qs#-q_x@AX$BYd2u^a@tzH6eXdqxfimw39Ilp$p;06IZoQ}aX2?@CvaKh~beBp=Y>q|XyWxPkg|t8d)au z0*8yX%0KaL8)qh|p?#>t+Tw4sG{VZ^elFb^I{|$?qn{?CX@EDj{-ULC3nX(1f_+yb z9;d2K)p0C_pCYGr8}efKGwrN@6_a*ganOYO@2{E~k6`1ddWXYiIt`D@6bTCnS;XF> zItZ-ZoC;(`^^G_l499>ToT3(Yvk8b@xY&a)A7SO_?whGTr-{cU`MzR1QmjB+}W|bxx!MMWKuLj32V*UK7_&aFI6`K#&gL+*8s#w1~Urm}lVw{t9 zEJLv|%ihm1&a;Hbm}WtSUiKi7W%oVkOuKu#yM)Bl4uV$Q&C5Ighxg+cr&`-?Jnnc< zzheAapuZXHtMz8sn-_dv?wL)+$~{Q9TjGF?KZQT$nApg7pr4O!<+}JC`q>p8rpjXR zi!|JdFV8<(<8d-wxiZ!!P*?m*!lLgER=>$+QMRHsAhUlxH>m0k(W|ne&{}1n1OjO#Y{bHe?0KlU z$G_e!OK|D?#HX){Za35esaWS!Ww!z>{zdGYBr$oYS=2Yaw%r+eOr!mK?u%jdsRbnmn665hlI?h8-Z4b3&#JVmxpDNjUgV9H;E2ui|}Afb3E)Bka56rw6k5u9cn~ z>)_|cvkGQ4TB!OmA*3c0!`CMB*u4%s!;i=D=|7keO#`CMsp(RGE7bN_En4Nrb~IRF z@@vIzF4TVV(Qi(HOEBTCe%5DQD4HRK8vGHO1)BJm>#7*O6Dk!Ba$kLM8&E_Jn;h7E z3W#T#BDt#uke|)4?N_sPKoPvQX>HLEbVZH&Jo(K5BL(Zi>7&j(mM(5(NC zql?HKj|&RDtmdu@?MW&KCh8+-JpQscMlz_Tb^cZFve+9_f2p58OAwu`tGxX98jU{wC9raFRa}gtqn4& zucHp*$4+Xp7@zZKU#1Wx;uu4Z&kj;Z8gKomt}o2?O>g3gHiIj1s8sdo$pU@>sOwxg8yaANmE-a5s4(vaAnq%QzvY7Q zzE59$c$eo9<~MuZw&S{N9Yapqx$Ry2YkehVFQ81VG$!jUT7)CZ(tBBiU=Tl!_ zG5Y~?W9@AVsc=~x?QA13XNXbuS2VN38XJsMqeWT(~pde zuh;F_lY9(2``2mT6_=dxIQcE_+F9En%j$#b4K6193#{It%tHIGSbPd`Z(BFN`F*DM z?s6uvCK21P?^5C=o&;ifcrW#G^5*#_QC#U2u4{wqp`*L?MD0JkYZUoCI-_wUsJUX}S z8}~#fk%^a*sh?!2^@D`#H85W4u445?N15%;CKIjAb*LO4xn_%A0ML&*XH1K>0Nv{$ z!NCs0mmunvR`Ikk){O+qvscMp%;j#Qx*eFpc*Dtf-k2613}1U(|NX)f)gwU9mwofM z=qr$=kK-jIx$!uMW?!jcOQ!NjvGGy1Y0WSl>Htqbu)mkTP_*QS2AZqtK8|D0mwwNX zGg*qfvHh&-*P#r*l}x;9Fh|t;z%8Z$eaGv0Z3dpgRFwyA;Sm;Z+5BSa7%4{>kBiE( z;Pwp!O7(-$91REN`97-oLdv&@x&KMdZTfzg|1$1j9{TC|7=1c~6lH|zFjg0NU+1V3 z5XUc74yo#);w<4amQOIh-8joFa_!!gKyp4@a=0Y}tMB=Q(FLWY`~{#dw7=f5$rL(X z2R}8YY(d?U>IWahTwHFxQIm&QKx~)u?cXzpK%7@Hy!IQ++VOF=B#@A;mi4D2ZwjIIW+As` zrZkiMOwD$ETwg!K^#7!Z6%}I&*I_dLv5x4>4j|s4y`aWW@VKpy-{H%9Vcbin`LPV9 zFA*=^KYINh0uniH1W8pvmCg4;MI5K{ij?&&dG;7S#<=P{4Y%?hqns2r`v=T^ZF-%z zR^Y8S%yNG6uA!Il5D>Lu*`|Kv!C1H5^_#rkQPch~9h_tU9=9i=eN);3S|GR!mayo< zK#qq)QNmKb0#Mc;+b;4Z3doJfY<@kK*GpWt-QGGc5X~H3b!76yWu|&NcN+Kc24Fbg zMCEk1ed?G$yL7DO+|eJ*e$4rO7hZ4S!t`{?s_?dBrS6P%9C7xrGCL+`r>suB`rH6S z;ZHY+D4*Ufsd=wbo z{Ecl6;{(OYD-x3U_Hh=HZOI3oK05^jvWKrDeG!Pf_TyLH97iKh_r!=cV|*u6C%Xle z8dvl?ksRhnnPT*Va(}aE2PWU!{jMb23jpa-^^tEEnc~~Abu&6T6E?YY&j9T$kIoey zU_o6;;$1U7&w%L0iOhwyE1Z@P}cWOWyn**PuJ1tG+2!7D#J)UsU6^0_ncz z(c{?|4t&L6fq71WENV(;-Ap|B28epKvdLM_FkqWxfxpEL-PHm!U2k+?V7_3dpRx*i zm~(&r>2NG2f7C<#qN%n(+(0QAVVi}nVLgf3t4w+7O#9f>K0*7IK1*yqSe3Sxc^}5( zw(xI0FCz=&Ns1;%eLu9<;@hdl+I>*0p7blwQy+sQ9x(TL_w{GO*-UH!VjAWZ54g*pD zaUe+ui!UNqXGYq*e*^fT>-Q(-lVJK3*AbBb>|TS&o#HGrNr0Z4&Obk$9{^%k`fhK| zX(07}Y+Z?%{c)uH#}_E9bz=eJd6+oJ6rJjNU+PoLPe-dPFbb)?u8J-6yNgASoNn*aa!VGlMLZ%x-k za~uP4Vnd8`GTF?clky$vS(+PDWB2^0Uen>Y0`yE55Mj2(F_Rs$Z;3}XN3X5Q2RGlC zMyY~cwJ053>+oPN25A+oL*vfh`x_OXXeX?W;Cj{b0?Tsb%^q}^& za{q>V7;kK^=GG%GKV$dO-L`$_bvlBe?>d*jz~1*jUQ23RJ4Z$VPW}Rss>whv58ziA z$8c>IOP+rnXx*^v_+&bMb5h;B`+9#&_z)?Zn27!>c)fHV36ps}0q$>>^kO_t&Sp`YG7V&HHrkH~45xoE(n7L7 zUl)(t=TC5Y9`;}JiM1n)<%w??hCe(+ytC=8xf$^R(^et(l1F@Yz?iWiieVGBZ zZbx1bpVNR`p2v4VUlS`weHXXEupLkaEo^s7I75>5$xmZ%c!Bs$-EB{N463}FOWfZ1 zPduBH*-y4S0z)n?Ub?JJhItE|rvxb8e7_oM4vUVml-J;K4}@&Y^D(_Bu3GKEhPOwd&+utQW|I}P zcSk6WG-CPJsj>HkE=UYu^W$vd5W0U78;468Gneb10*%`1{N))_TrShTJ233L=To-= zP(maRwdqAeifXiUtQdA)lZ2D4_6`!^^=j5*WMU-tA(^7GRBY)E5kSkOoDDD zJhq3V{X6L=I8A~4Aw=@R5pS5d!>htuw&X{Es$-e))V`yrS4{RrYAcr4GfpN6?AgT+ z1B!PgLT=ncLqGZXB7W#Y*Rvg*7p1C!toS9a=__L$)mptzzxfDd{1`G;(b0q+b)_hy zIzc>c=bl%ObfTc=*NCln`f4HXTn zhhOW-G35jOU;m=6lnyEHpk?`5mzS4f+sxXtOgz~@Fe#~1xwRRO+rL83uci@^iOJ&@ z?=YT@&Zp!~uY?V#r-=M&Q1Ul6-luDIZp1L*h5#``qn2HciGSw5{xb9`#|u6Zb}09Y zsgIhdJX8BS$)aD+F0jMgtq*5QVvCplU)cJ`!}Gipj6M83Q)k4?FF&CFo%_enb|78I zI?vstrOaBC=`lQO1Z1<-@>j%_fxw>PWYs@|`WHJY$18YHEyXk7M$=b3?)5RzoNocj ztcSd2#_hVX@lRa~uZ5~n-=ND*?HGGB^cGECHs=D;ucpaA+>d~~ds}c+$FljV|Ns64 z@+${DlNL;V38{z5R}nFP8F|IFHlx{P^TB|W{M0p1X*25gtUT1q&1koyLFf70vuJo+ z^2YH;@(_JEWN?djAdn08oMYQxj+)-esm>;-0GX|D?L!AC)MFbRmD!LCv@U|h%=K+R zEZk(nO~m{Jl#5#*$Vm4A@%zuPm1~@!WAa<**3N4XVZM;8QE3mv`HjA*0|+`AM`|?W zS)qELho)f53~K-T@ucMV1w>yJpBR;x4waDy8YeasL#M#@zzxOYgHwea)*f4MY4+ zh1```K3^+4YjEts8u+vO7U#f?IUsSb4|{df6^~PlyHOl<1c(c{(og=`=ZSi~uO7Q% z__8Itcb(x43WSKY^8DPJzXBx z?y~3f5%BPT@!^7o8|qyzoKPOQ50B%~*PO~ZhsSLux~PgB1M;WFm$%3M@rNF2?$?h@ z0MavOz8{Bw069R%XhGv2{Pr?Vc8#losF?lyu=d$FG-&^a&2@|&4cfNfFmfpbvP>;i z%U3b3BaKiM=(f-WD=}CGtm-=4A=JJ}_v6~jn7^odq{4PqrULrU<(!zf z-T@IS94=d%XF^Z2u+ygPqKxw{VpO|*5=fb5fL{KlVQqH9b3vsH0S` z)zSNbQpG!acIqW+OVK(r@$M&7c%GrQ6Xd_|NGo-+i|d`^c~kyJ=KXT)s@y-!Z?k)?FyeZd{t8$@$4J{ z^1Yw9$O>UTfmQM9< z*R${eO8bxqUo>N$Gq#C|*fjv5^_ogjH5Ga%69%40ZG#zMk6&|dUjv2JK8&s61(1|V z`3gSVz~gS?3tMd%-;LyK=*{#5vgqhUVd-NaB~GPpReS*B*_U2zePIvek$3kmSGPObmvbC4qZvonKU#n4wW%~yym^O-f+iK^#%%ZyEw&UFkH^UvJ1_c3`T!|gMQ|>L1&BAa zl{MX&;@q%(iobgLodQ!lAetfi_;3Hc+ZV{pcz*N(!5lVcS?f4@3vOwKX&<)N2{Ey{@OAik^nuxfF9CZ_+ef@r4J- z*I$yhN?(Q()MxEO(^WvZq;ER6&jeK^js_?0ZpGu)PuB{$#saxLE25E01P%UHV|gd$ z17we>2Vr%7%4`?OswS#ifl}`xDZ}*wwW%K(F4z|il%Yi#lP)oa9sB)wrII!3=+Rj> zc<2;#9bH3uLXiXV+=Q_E8*@axa(0&wy$;B6=Q_FHV{uFI1zxW=aUH?qPUQYM9k2Y3(Cm}$Vpv^_5y#=yXUN`yiHs~4a&*00ap!z*W_>WJEqE-&uZ^SKq zKw7<4@w-SVntYek_1Fi}p)lSr@pZS$ z7Bq08AXfC{3N*fzTl2YGJL=|kPTegThQ}4Du!`4wMMHgVhLSa(pmv{K{vL6;z)st2ZPVp9EsW`80tY$I#*+dyuBA1R~&`f3|=iJZ@Uy8qK9g znN40RZfD0D7;cbG%i4yAo~n*PTvjmDJR`U~N|^+Tr=4EAe;-Wmsa;RpdjMwMNf(4!ZiasT;%#27;|(h;712=cb0Yy010ZKA@^fdY1I0`^HSI@qLvG@KJ zuBRHI6K}%4+wcaWc2vt+ajIea)}ywrOu-=3_Wb4H$;nmNzGcfwK5HldT?575`I&g= zx@&$+`93qA*wE~U&pWmqgU*0CEB#`I-C}=vy?-6bF^%KSgo1nAkXwu3z(Lg6vpVOOW z)*GswcZlFgwm^FQz*T+k(m9Vx#jO&5@Mi(}(!`6-QVI}_v`V}(`+)x2uOe(C;~YCl zEqL|j9_pHmT&PI7hR3mNpBca%hmQV?px3F5KrWeBn=beRNcno!79>?j_KfB<+24!E zwEmky3JGXfH%Q_B|L_>LRN4Vdr)t?uYMdwhKqEbwL9rBTiS!u(&~TkNEE4RsKMf zyS>>==`-q%IfuKfh55hw66!Big{r_5TfvQa%e74NH6^#YJ;N~(k7J!X#gXmE7zh22 z-x8ON3q%9umvZ-npN5~SPH11+V6^oA;&l~o-rNg8&5t9~yZj$AeOHe}R(XrEBHG!J zhVu0PGhQ{&D2r#cH5uo)(49xN299Wm{^|T)fi5&d2@aobeuR4Na;K$RG4W9-v&oz~ z>8&olY`H{Vm7!LD5*i7bdK~2Y4#)k5F>rczPY`R;R%zVra4p?sgx*b+Y|8fCBJ&GeK<9k^0y zCANbw9w<&rZh8eI-fK_vCosQvBovE$Y5R!D!|lL=HR9eVZ_&sxq_hg^K3pP3JKshX z8D%UAPA-UCtoAgm$qdc@<~;tGj|&!R#X1#@k3riD9ReZ$Vu)4%Me$ZEhT4&U@_* z^Qc_6!wrZ}6%YGu;%3;JsuvBbqS1gsaMnu>b)da}o_f=Up@*3Sr~4cXg4&>~Nn%1F zsIO)2naF4&bZr!5&ps)RX8OAh#0O!x4+CQgyEWU~@hm#F?;E`Q(Sjjy+Yd1hG@$WT zIP2jELH@mwZQSUrC+qZcQ!!GAwJ#Ux~of^?%iQkvicY$P2h!|$cK;3g1yP~cx z-Gig<`N}LW<4>qa-E=g}jY;oPW_$T@O&`@4NVhF-xu0K$<`46xq@`5@ou1Y6-o+k| zlUf|z@!=BCio{n2ef>}TI&7b`n5O?;*#&enVScTUTZn3PShhHNNq<#ljeV9nj^U<2 zQ{8aZADz{RmfE<}_Y5D9-^izi>=c38Mh&@Qvm2^kx_119j{c)F;@R1R9 z`<|Bj5mrDVhFWNPVmL!X_f}EoEW8+c@5d_^mBT<6*Kjr}TatT3U_E_FEUy8NyF$sT z;iMtzh;Vh>mVf*Wx2%LutQyAS#0zQdz8GFzXSztk+LgR$8u(A=RV?Lcp$u22*mi+ zk(U#8z&Mw`@exG~2bUN&qWNO`@_IUn{rV1xn^HiQ4?XK=5db4IN=`SYjDeh&KXX6- z8Y0y8UfR9?DHg%Jop^iXz z_u9W?zu<8j*1Fo8D`CG)o*w?5vku5L+qKEYZy0)ra*lcPeN>)(PW*28c~p3GSEXR| zQ8aw20|>TFKtIO^t&@JxIbdiQwMGKl&&h9(irkt3igf=bsrph>25a&UTD(i3r|XmDiEuC`y_wv z`$w*}mCI&f_q(c2saY??cWLUZAb4zffD12n&1~Xul+e*GG@` z7+xPU4*k)&T{i^tfg~dBZ}F}ikE18MX560!k{6+2_zJcTI$GY|ymo)-z7Gw?@vS+# zgP|vUcxhU{UK&xqIi+$x6$8?^#}#TS5Qd?KMws+(7-m zsHtTpnBOPwee0&?UrYDAsNZes#LH_pP$v)yp4yh8CQYmH&3Atw+Kg*=Yuzx6+eOvA zW&Z+IMH#~nIU`VC=dS8F9w{L4JWPGNLmT@2mAt?0Vd(qU?cP2c*rCjlnAEb<3gasz z>pnF&e7YWWjd;B}`gjcW`3fpX)KsDY?mX861}3P#{TBC*7mRUSsdwPh@$abH`_PGs zW4};$q4AA`Vj)-{J#pNoUc}g6TI61Q-HT`%@xQpLQ_xg1n$_m&MFY^IyHi;cb+~kl zOxf>%-g=J0n&2V`ZCt5Iy`l&;cMrZgqkkC{6ure6++7Kj-Juh5k29dx_+TO~`!DJU zAPu*_+6~Eq20DxW_K3Dh+?s=M3AJi_Cuek=0E(vg?e8W__rpLwaqnm3X+I#H_;p~h zsslAfYaYvvj6?0xkvePFZ2_W1flF*D7xRAcfL333d;vNp?(w}a902ku8>q?q=N|OJ zwHX&TDtuq8RCIO2cyn8nc6_|EIt~cQmL_wCN-+L+n(N@kJBVt)QZwAU63Fg(ReV3a zP}%hwi5ePHeh|>&tL{qdufW=|-6=3{YZi1x#qNZKDMq`@4Yp}Ep$=)j+DPtPApSD> zc2Ly-I#uo+F#awK#Pnf#rCqB~gUEaQt%q3N8);y7nO`>#9=FYPrA=uvCdXo)1>a<> zfH-kxz|%|#=#9MQZ(g$j$%M1U>&$z26COhH>~WQO3bChe7}O=GA?k|7saWZizA3M@aWDuymhDPi)9 zqLK#B4uy}Zh?;eQD)adc8r->srE5A7=YacquuE*K9UX21DJ)ttC<1m)tW zH|BTjMnsnrH>N!JQ27JT&dXVQp#E~U^i$zcRH`k0pnA_M=(JZjcVFKN#`_FzaCk33 z)%vF+A$I3bM~b$Wyp${Sgl(OUJS_kuvU-GuGB+B^KRTpVxWt2s%DFR+*ronKlxqHj zTqSu_MhIFHDaHm}0WA9NAs7$#@2sb1Z?kFv#rPQO@5ft#lEHT&Fv$}Q-Tfl)QJoH* z`k5c%Ui)Z#7fhd+$Pd3B2}u^W zEQPi(;SDj>m*Ts>G$ph%FXrHTAh&ywKN@nN)q|M0wK+jg>BA!h!7b2XIeFU7iWx6~FwjrF?{o+zJmw@@zc9@Y9_PfhQExR9 z{xJ|)uG&eN%V58=H;3Qj)PQ#m;iLNcOYyi+TYd3x!82{>jFQRz#43nrk4o0g#JBCo$&I~ANa0ZD;e;c2dA=-$U& zZ;U#gLDk3GcimJR@wmdxQAGE0sB>0V>)cZYJf=}Qv-;NTrS%15 z=b&Tc9Splr%3L28xx^0up)bPY?Kl{Eo3@RC!e6Gm6il9U9G^t9C!v1v&G@TU`50+)BZ~1r z3KG_J>5@VXN$KWf^%1k-+M2Jj|R%yG#37Trg(AY z^_w6wU8*jDs8_dl^j`UehAwa7wfs@P{QGp{yGE~H0?BGj;VGp4tvXc%#ck8-ZCcPWFGWnA=yg#77hF9>lIkt%Mt?nBSy%f?FtSY@NucFYS zd%DMO3qqEVTd=gGGaC0kQ=oNH0*2MUU-GQ^3m+Pz%cE}uE3=-kFy&|Yh^D+(Cf&Au zpuFPy%O|8goM?1QWJ1(b6&l|9=-8z_rl`wrLwbnj6VxX?o_j`o7}{Fn&vNb>g~l{K ze%-MNC>0={6gi`XD16f=4AcNkv$9cGtBEiWf8u#3={D3h1n;cl#^Q(uxR1@}Uf+e9 zDr#T;s!M0=(|zGi*^GU6dgD&=kpLk3JO0)&dji#TkzYOaOE|u0qIXB2TJ8%J<|Izc zihh7b-n9-+CC*7d&`*wfDxoqgy zdGq-5Yc}MlSde~Yb}9b=jkO)BilJ!$u^YF|{>}^ta?-gCS zwRQK|CI1Xe4cIuYe-2nb?f2wY*Ce6A>dO<~35S3zCQ;$OeI*ci_u7`mNkITt&?rs0 z3J8N~?E*g;`s{^au?S8rsP=3-tRl{&2cnYOzkMgQR>62(CBKrb4;q_GBU1u@pp@+! zc2!HWL3JKwQYiEq>f1ZFXF-k=4LVnh?Tk5r26uduUZfHsJLU{tsNoDm6ut^7dpF4# z|AC+68>A|)t$HU>2cM4gn* z&B5b>Zx?r0XQ8nLx!;P%R{&*H?%+`F7?42TYF|AbqG~)2T4vk_N@9;C+dFF*oBe*8 zy~iC$l9Y1&D-64?=}xZ$cN)|_XVo+qj)Bg8rIXv_Ghte*qHUam2U=HTh{vTTqOxBK z<=&KfW!7lx213Fdke$?=1fAAu6l z=WJ+t4|-oIAE$jI0a4IIC830sxjq~HGLXmkKuT~16Q+Nb&R0aeVR-kI2J=02-~AuI z2Cnr&XLfA`_jxQXzhhxTtW{?_BAz4-c>HoiHQ}zV+fHks*1O|tm4nKF!cG$Yd1Og{ zMZ`1F)}%W-5g|6}^P#6-z^k!Z;oY=9D!U{bduOL6P`33Rod0_q>LFiX#C8@H9-k9k z7g>jbc0T_j>)L=iE^D?{cLMaENMIMfRElzQx%LJ6<)W6Sx1C1SSQ+-K7*E)GhQ9go zUTqm$F!VZ=6K=x?IF-$-c4Oi#9%s-Nl5>#=9hq!bn+BqR7*9UPw$BtuI+NCq){ely z&z+a*9Z#b{4t_`3C(nTVP-(T>Re+Y%;4KbvHOg!jd)M@2G1kfU-b$8-d5Eg2yPo#T zoS_fT{al-r35$)7_h+$h1=0?+1PKFKRKT@WZG}V#kk-Yp)*YBoX5-*L!+w_O9D%8W z;|)>oV_<})r!~oDg2(N=oVL+i^nbsH$K9fcDtxZTe*2ibKiXZ#Bqsy=spAheZUoBW z1EJe{wqx(FsWn)!p0R&jdDOA{_-mjqqy^^I-NfTeT`#Omw8Z07s4G@vro+Ue2Ayxp zy3oh-T=X(~Dr%B$7}#j+3iP@guDP5uh|1D`Zp%6+sOrsYI9GlHh>!ModVF?(w3YX> z#ft1;V&9D8b{>YE*WS+EsInGj9`z(VZN}orNJsJ>C{>jLx#zptw2Ic!{~h`N8qX|L zz*NL^Os?+|{tX%p%^_MzEb%`=osQc_q7E6A+H_k7f7Hg~cvUwV87$>ZpdOWwj5HR^ zk4cW-rE&i421M22?Mkd|fUfngx23+HhH7EIIgclHsE6YwpGb}lkRy(L3_tUq@vo@$ zMi*pq76HLLH)d}S(>>DCd$T3}x4scDoG@BVsrwG$cR+s=o5^m^5sYS|cyN7qI*Lcs^M}j}Eo)JL3=eI=- zk~DxOytn4-wY;$|qdY zI|1ot`o7DB7tmni%j|?KM%;6;Aj`4e z_&CuF#F_?KElV z8yTypr>L|4tj9N^H4ux)PLq60_^$scmlG|lA#aW?Jivsbq|7?iYV*O!8k65__l|elFx=vv7e%WEtb7si;)s~AUOAfd_En$WkKw!w zYwhwK=kEj3NS=@P*W-x3@Y25L_cb7C_Zm+LZ$*Q*BV0-!&i7;=bFcFsQWRO~Fx!8$Xfg z)Q9C=)@AidJW~4Sey3|y$hu@D7_azibMc%R!eA0H%)luizB*?5YNiQQ^Eyg- zm8zp+=Z=SW`YxkB&EdQz+jD66j7{EX`U%u!>0jQV$JjSz2Lct9G59cBBD-l%{ z$%N~FZ3fzw0}25F82@^gNWRg_dlE2^?kV<2j1vv*uM7I!xDFEA>Hb5(Se_1PbWFK! zZ3>XI@{g{1{t$?XQ?iSb>M%u~*LbB?2U+EUPgGAnWm*T-yjq(_vdM_5bwltS8x@Uh zbapJ5#&~MzaW{Luzr6rtUF`w$=bK^f?aE}APY6u}$Fo_l-3#>Y>OmODgT7HWBQ3iy z)JxyKO83!Hegf*Q8;U%Oq|wmA<<^nSf-n+jwoR27sLWOs$^O1`PMOW$&VyhdsmzA+ z3te|L3w~^439>JW$K#~3tzK4g0p-W=SdQrekhqU^93Q&Jw14$_JJ;$+pGO%-k`iaC zexM$Kz1Lj1=22_ueP`oGThL%ni-M`h!hh`tK(_J~icvlCzyFWNY1NK>K9|Di=ay-{ z@JKW|TCz3$852G?bQ^D2GgK~vsQkSPd>KNhJw{%^npFenY0jIqBPyU{xckYK$a3ga zJowIRjSY~_O-oz}!a9;L* z(02N&km(65-*rUmhp`*WRd}2MCoEk`@VEyl1r=8-p|U|%SL)hD=y)V+t8z6JNY_85 z2K~&3QL#8r^GFsvZnJ<;O4kA!imXtw+l%qq5I)B`_wFvhet(pj5pkA5zfD#}Z)uxg z*c)-C7n4Mw%JlbTNgg9i?)s0fX!$)1z4TA3-ugcRB5<&sT)YSryy`AT!-t5NJh*bq z1d9(GXj!*ct2hisZZ^r+Xoo=Uy9=B4D`5HmRYtdu=Y=fsl>!m(6FVlnG~TG`n&;I} zVT_-K=v_s;9MA)V#CipZUl?8m`Ke25otqV+D#eFh&BXHai93umTniQOxSaEso{zr( z(n?J$sGJ2_s?1s|5I9z84y1-x z`ZKNPv3e5&zioVB0t*|XG){eL0a8NAP0zsFO!7i#5{`Q(!Z=UK89NG^i?I53m$tqO ziUpF0UA%t*Q~U=W_sY)RyPxU4kEwim(k_cw=U8Y7JCnUWUmcGVweYnU2!iTgQh(W= z-UA|wwTuZT4^}Q`aFW?zF0>a~zTJP53);U^Y2OCapoO?Ited+5_zzQwS(Ct*%p1LMUd%)|a6-EPnd>?if@1NfQAhKZ-8F^j`nX*Nu{)JuB%m#|2d&ocN}F z=^zz`Tiu$K^;og?%RRrDmtX)r=kGe4lw3MTP;r{;!unS+KvGg$*WbDx2IOuPd2bfN z<0KBzPwQne+9kgA`3B5i)4N+aIN(|UCNE+72UC8H0I_?+-HIZNABueJ$KNY?B3S)< zV+CC)dO!lVyMYIa@wio7E}vcVv3A7M9XAd?#N$lr_8flQ38ZYLlv-DBAb-Von0WBx zaV*s2>P{CR>pi({8AgVNJbq1=n0TP+a@(F}!`55hcJ)4iYeGOMQ5E`F%zUp$)?3pZ z&l?J)v*wvU(xst;R?_svcq7mpDg{#ok7E6*G&SU_CjiMLctckIi)H1gJyd~ug4&4c zYn;zT4c}(S?co%g7|ahuKD(7ymAfB?=Y@9bXVpV}SY*&F=@t;}wqC;R!f*?Ie2tPU z&|lK8VOY<=ePW;ihzWKVSnp@!al#w*a5c;N4Y6|GHJ;wphutd-gvz|Jy!#1=_q#`x z{a-`4e)`KhXAi@lr`qpM?VpA2uUe0esidM2k-m;>YbPMzn-BjpkqNW;;rkyuGum%> zZs=eyy+R%cYX+W44s|g-uYMNQ5xfzhq3+p5Jxh%5henK5e8QCr^r5~B zL2G{k(S6$T;d_j?s3lxS%IwOw|LQ06y63gJ+&#kr4TfEjv;0{8TJJ=J&(Mx&pcnAp z(*4a3Bhg3J3BJ6G$5|OnUe;;{(o?TjXH2krfEtUP9X!;1jPrHXo_@oV483|=O0s`A zkUXyiXo)fD2Y8&g_TBJ9BS32t;mgv;@(y}A4&46wG7iXLGb$ZvSMazEG9@e5zsKX` zvjxHqwgN>*!hSGJ07z;wBL^ihd~|Y7-F0atEZ>X5`RF_4m^8{0aN;>5{t6ABKCV~t znF(J5BCkLFBPXB>X|mEL(?^Lw|5?&$`N0v0?*%u8Ii{k4?*TJPCuu6+1`jO~s-u zs2gD{$MeMYzZJZ_zpM&)%3SH9u0<#UyrL?Bbd)z(>f<8iEg z{*5_A=+sYE5NaAlzegi{pCn=Vfuz?*cZ>^bFzoYnKVzhYfmX+v{ypd)T&?44Cz4Mv z@ylV~UzC%pEh&MKYsp*c_5J|i(6tGZ(@eNGc$_liv!xrMp(h_}BvN*we)&J$G}lqA zpC`_Q>7HTqYsFr*A;U!=hv)QjHrWE{#cNK#{5K3c@mJ5rw0uFRPlY| zFN(S=s`6T6B!SA|fBug@hI>O73(#j#!+6!&-W+;)BK`>KJ>fGrrd@{BJ1TPjpkCAe zZZ8@$r^?=behm5>gtRt^d!nJYF1CkkSb;3O%Stmm0;txWFfc_2((gyY^WXh}yuoDr z$aN7QON{T^>BTtbSk?%qN7pc)%aOU6<)!Mr0o0p!W#lGMMW*y7yp(eGL%qK6Vmbb7FbDS17^ewQc|q z<#nnQ?_&8*IeR2(@eKR_WDbu)z*ihYFS&F65g!vTDv+XPE?zo$7c$~koQ#r+hg9R$ zK8+970cCIIw{%`6y&IyBME}`(TN^04g$Mj^4ls@T-{gvww=SCCaRxyKkNFWQ-m;Z@ z$7hT$u6ARu8!Z;Q=N?xze5u*r00U}!zu25KMs2wv9pQOPaSy2U>7OyL^%xF$&D$4~ ziLYF+{_D4z&L8~-^sk3a-eiix%r3rlS@w^Cw69mUDUV?X-o)SA`#Tpp&U7T0Jv)xN zw({Khdi5nk-=5rLzO4;<(luiboLtMW^A0A+^`BgJoQYK|`jJr>&j{s|-mcQzNIdSo zt-&$tZBW(pX~5|!hW|oIxyIL?se;F?u3!~0Te^QlT>>oM9x5`88`M>MW^!$6LGL_5 zYw#RPelR>v`tFXMPEu%e!=xl9%WCwOZ{s?9`v@RDN@I&z-2ue@cl`6L&za|yOQrf| z=s)+ETu0?gzA@+PQ}377Uq7*=hoMnjRpsV_)6iEZcJ*9{2h40pOo;0k!`j0qWJEMx z4P*5b@bRbVebZ(l_y&;^u-t6-$hxLo<|&1RFmo6&D5_x1>WFGi@Qk!^f|BXNOD?V>v#(b(*T-4~EPSGHB{je=>} z0p25J{y_5fjqzF6fVE>csb*UJ5gsS&G?Kb1YB{-Mo>$=vAv@X$7~$FcO`L-1aYRx+ zt*jA?ht9ck{H~VjO2+=Mm!)-cKG3bUf>>TQ)%Vn!$651_;db%6juIG=~c34F}dj59{fY3m+}6V zGWQF6hv3(08J}=-c|<&mE1vACfKHjbr`B%WFq1O&#q&@c9(Q1X6l3iN6b?7N&o3`9 z_HSFh$Sx5)F7zXzcvmzKB{H@cUEj`Fr&|{+7DAwUf%o=pPYkz;c19-i$N-jiG35GN z@I}kgJtHhmY>oQ6J`zYfJ16ccox$UFF6vrz2jOui5{4>oE$M-%*~`hv+=YpE5+=eX zo#%W7+soyNdA4qT& z(vkn!kH?WO_Z`aJ2z>&3RQM+)81{+QF~jEqF!O9xtNa>fKH~<>rght|GU3)hFI7da zo!*LuR*3ZRkTAS!QcUVk9kl_db@k!9+$4?yar?Vo{8U53<;A&GpS*y4uGp#EQy-Pi znl%eGzX3XZn^!2bL(hrClP^~?;cf!?(rT8r^GosSXwW=?JEfcz+A}QQT5XHP<3#@a z-FET6LtbgcYi_A50mVsYYq>yr`+(l9*Vah-kZ z1^$TRaqj#5hlQ~CaUwlc(@^><5bNCEV&fKP0T^WbYC+(bpQZ_)k z%&rrq%8b)Vo};Yj`pW_Y-^y_vH@8l4n2QB6@r43u!10fX3zmoBTT|Q7JRk^D71V3C zCG$X9%vkqAe;;GNefms_@B=EhuRZkl!+MxL*0gmE>j-pg3jg)uUNuZOs>>?63Igd+ z?*X+kC5Zd1nDY9?86a;Jt9kTV4z>R|@if0d2(?|Bdl0F!WYjwk&9P=XrO-X)z49&ae9{J?ZxuQ+ih>|*fC!N^d2=gsctzSA8@GqV$A5@qubeh ze3n4Zx!PqAc@I3}6rNUFzlHg)J0;&JsQ>|4z2!@v#N%wqg%7@4qvkR9qjZa4bwk^9q1u84MiIVfle_b zJzBkuF(1K;JrVw>vv3FZOcrLhkyFYCC#j(@CVc3RQcN;qo#(wRv%%tVX+2k?jxa1c z(rJ15>si%6YP!VfyWot*c2s40{1|1D2l57Mvm54&`EhcLDR?Z1=?y(EeD2pXp{oJ>y)yz*}<`%TMXmcQb!+ z#uBI-%mq6J`G9;iQ|O%UFGQ96;m1M$$gq=zTWz*lWAaLWo~F?t4^y_yxFnD9Ut`cdyZr{rO%c;Pk9^3+s|s`982vY5_MhMa%7w^FYf`kuE*pK^MN$K}S(y}7t_ zuA_<*`pug?PD6dby1Qd5v3$N?>+E{yQ);NREVDhJ*b|X;C)I_vP=M@`^u;cOp%2-b zuKBvN879Q8AC2w%3iQePmd&63FxJtN#qO|57;s3GstquP+U|V@T;DG!v$Z@vXk3>7 zLQo zvD6{$4~f|O{*5Igb(lYotV>@LbsY0!q#Te6IUD8(d`v(yq&9L!MuVHb&1UvM?wfZ6uz7%D3MehLe zMq7!a*Vkj^ZF+GeXwPk+@aI2_c)*3o>Fg7S7W;wR7x(oGxdx_*LF2Q}U!cxmZaa>c z?F_v=lCfY7f#~w%RjYG89;c*Ne{rJ>5a#<1`;7m?OESuRdTXc<5MEepo-=$6b5m>n zeBAt*VOLnE?pz=&Hy?=Jnlq;29i zX=b`W=m?tA;$_ymDF?lTGOI&?>@&7U-B+1${;n{UOm_rgdCxX&sy!aJYyI?yD|Vkt zO3M+w{Nuv1a^JW4=TTN`+oj)`oK{~i@nCuz`DW+Vh~3QoaH7b=zn11)B)ya^`WFCd+~yiGFM3P|N=dVKEO1X6IwsJ6x~ zJkG-M-n|ISk2-yCcLh$Pmg)P@23GFL;U#+$=67|M7n8(czJC196$K23x36>??fu0A zjP)g^6cy+KkI52gE-0-BOSufpIba1wN5smp`^^} z=j|A-BGHy_OW~2h&L*BhN2) zqVC3oL)0rBn_2c%TfgmBKcuYLb`l0?db2H}u z@4P#$b8`=nKlGhDv7Mnu3$vz}f1d$b%--WTPXy$Gh(Wy?hTVP4{-Vt0kN%Tq7*V+rYF1h7 z0J=YZ}SX-e0knLxs+?*0i{-P+j$qrg?oMWDgo1xjVA~C7S6U)i_cL1{VTzwY)&A2^Zj;Px(yAl^5ECozzsy> z#z3QcMX+$?^Uc3(Sezm0`Kk`d)tH~)S9$up@uOfoZgq0MnwKM*+B4-Fy|A?ZD&tN{ z{QMF8pLo@%#PIc<6^IwUZ|U+K+Q;M$)7jC~$GDd|e?`}Etha%FRqKoG@ z`oNs^Dl-X9Zb&{AI`=D^7fp7s=U6pixZdRVTei-0eFuV>iT69s0w~?{H{mkhCm{IT zyX~oyg2`89;*F?TEN->Mc|$;vgWj@olh<-^L%C}pkcN!k9;pdIP1>9HJATc?+V8!Y zC%J|xuWniS$tUmEJhG`0YjzZMHBaaT~q5@5A<74>q^`rQ74DW-!pnn z%B)YaS}gS{5c$WxwL_eSXu#7{ww3QDniN;5eEuRE^?M(Roh#gh8hgLqW`7-ss2O)^ zx82A<6LF&nRMIdss%SbTZoPnp*4|1UO3zeg*>3R-e+m$-rJ_tBaOvDu#^u=L5_B)4 z#r#JK)yMjjS?P_m2%huGtUoD^m`^F9 z=xoRIoMNleeCRng!X7|+i8ouk_bB=xgcs-1}Kduyqw0{?y@gx%wp<^4Vwo5|+o)uk*67dDn3jEv~OTcxvho zD&W0{R$RvN^@$)W!OMaDt}&1**($k&lMR1u6ZlA99T9C?_@9>gXQ*n1{g-6C1Rkf# z-=*m`4V?$hh};)AjRrkULim!F^6Vhz6%kc_w?m^J6B;XyhC3Mnfwjxm-LA8Fr41 zTl2Uu;G?Z%b6yBSy9|9!6W;{gV!tDN{ZmkWPfN{I+ATEaec6A!Mipipm4Ew6`UBZn zY~8H~3d*<-mlKbDZBb@Ru&hrMIfObnS*6BobJ0w9V)XI9B|r*v^mA%&M9re?d$|ra z0ExB4@MQxNUnv^hA?LkfPZk>7L^rSLQb7YRYA*>p&!Mih!|(U7{(#Pso=>k&EamN? zuA7Z(-w9WsCc%3T>u?AS7)2$^&Um7tlY%?#v}MuY$KTb|*iQ_-kWJ&E1?JZz$5=is z+fxqYr>*(o&NgVYpRcu$U!UX@ z)u3wCkL&l{H3ian{dUhBS2Ws@mQ(Q|5RcQOS|!h8@u`$I#m`rZB_ZmCTaVNF9U1de z8l>>r5_-qy=Q3iA(D2sGx90*ff%wF~|M)W-tX*7gbgA1M(5q%Hy6*$#H&xKQ64|tm zh)Qlhpg$SG@UuJqw#T_;9sS?^TRLah#D=QhZ?r%?Hjr7ty=@G$l zqz}8C6PDr#QK7nl=N?6-codlCX}^@NauGUHgS(zTa{-dR=S;UD=I82+x=uP;ei({R zZqyd^*@61aWCVAtWQvzY4R#;wa+B;4<>l|Ln_teL{zK2620p=f!8#?&PPeus1HnAz z3fC0I@AXxE%|cPYS=5*MDQ^D%9S0cEf@AFh1^%(q%=J9zE^KJ%${ETzTTDMBKgNHn z`p$)D;vS)vJdYUm^kuO<7SU+1!}OSpia*LcQ!2ymwo#c)e_mhs*LgJfvsCKp6J_W) zR!T0?+5>24eRXN-Vm$;qsRq6^T7iapiq1UDb%gw0A@ni#HR|}Z z?`@t>FlyMgYs}oZ9#N$atF@OK;&JJFsbM&GAbx%l=yH7$4cyFlHB9P46CY$1yao-S zch%9Fly7Bd(z)>o|3((3aT+|?H+06`0*GB4-HJ6BukO#HA7ygvD$wvSZMQdNFB(zz z-My#sJdp9p*Tmw`b6IFsM(Pkb-6dANSs5mjcSvt`(t&>``4*i(_;g8FgFl zG;I9Nfkr*XR8nKifaEA(cH24)>TvU?sz06t!hzRD?Vp2yc9K=F^hp@%$&|D9)8v9a zi@d((p|g;*G9l-1;$P(z;!TR*54)n?ulGMWN@Dkh9rOD83q!)upmMKHUG6QWbv?Y| z(xp2w#t=dOc3l1<2h^|&P~AA1!B?H{z4WT3I5jj-5WkzX%L@(ao*k4cO-HTB=|iS2 zmVY$N^}xu@CK(MM{d!IRs6HBdQ{sys*awtDWj9s^8A7AmLTrg{sftFLZV?;iFMOCb=Jo}!{A?nfaI}I&uP`g+Az3#?+X!hzlAq@=A za;e^yu%`iQGFz9Ozpn9uH&s?fki+Ti{B|G{H4}zn_HuQiNwldcF8Kz^4KxTbe`5#( z#pu&3f)D&oqe<-H%L2-SEU8YeIr!P$8Q+v&504dEnm3#sRc2i{I_g;*jtWX^E*|)U zN4@k@n}14AVC$sHVjGn$0CldU329d|_G=#f=LMa-OV8oLLA70P?SQ=N^PABPG$4&M zF7##dp>CTpW+^z}DAz9KWGrk{>YYGn?_PP=J(6AB=5Q zY5*b7@digf=HDWQ?&u%e^8iM~oA+H8K7^<(V(ryDOL&&bY&A_p*#{!fcd}l7%xojm zcvG(oB?~*op&_Mwwx>xLZY#+)RO#SDE36(n_KGgbbu=uiY|R_v4kW#(*=@l~_t|JD znRd~)9FxDsxU((cP3nlMk-J6ft`v|=?>sye%Y*|0{Rr?nO%)IXjefo5G`svqCp>J3x@0Q0HbUprHb>%R@^E0e(H-1+*6B_rn zO6_3!kxAP^#zs0#VX9T9fu^R3?Yo-~s&c+kNCRm?WbHN3%03!odwT&gv{%eYupY*C zx%t=j8|{D&53$=KRb^mD;yaNqlLX@*n(nJHKf|LDFTY;9W55cqX&}k*bKl$W;#~94 zqnBg>I&)_NecFE5RR7R%0Rvi(op;3;IsAV0p30lVLZtt^oGdKKyK7Yzh29E94WwfBycJ7Bwd`?yocHKcwrbg2l9fcC7w_EC{z zP-}YNh0}#rWTrRetFBA|)})h)XiLz)&Nb4b^Dn78(Fu8ZWbosy$~L4O56PEwss_%+ zSANP%9})U%>LAR}T}sI5Eg!$I*=hgO8=M*Ei=MNl&*RzoGbS=yke=qB()5m}P8Ld^ zDdvQ|m4Lbqy2ZS(24J?I5EzluBJ)vw-wiYqJm!q}^3Fe7RwC{D&DjG>U7@z+_kJpQ z+J8xQe!EL^CU~!JK35%T$o+iSV-CJu$bfF`?!O}qTD*E-Ls<>_@DXy|_}uqc5eaES z7P{*!Os?n%6AB03EZSJ*gI@|ZH2=85Q-4d$H}`3?uyi_E;De`7q=q?koN2zc zGq4DXIWstmXAtqUwj*s>FXsgUM`nk|`~nB0b^pku=o0+R%tOOP4!+#HESa=nQTAie zkNAn7u0%Po1(=536&s$s=Ds(gy(f|~b`APg4r=Gmgh| z_%p4~W=SANfw6TLLdP^68>>(nQZ_T0;v*1Jbm%R4R)fWFo_ z5qB2%A&q@YaPqAKGL%PeUG}&P?Ab?JPQ(vPpLgN-cb+=HK0Nl!PwW8FN3OGz6%-)b zcyrV#aweH{aeVbE%4(#|Tz&rJ$N)CpPkzv{-4NInbLovAwVDs}!1SuNQo?ks}M7_l{ds z+fE_9E^~hW;40*qoBnbz_d~|(pT<)CG^8?0ZZa-&uxaae&AUq6d>&2(Q?!*E=fZfJ z;5R?pSh9dqZu66q)AF1AV4XqUKB#l3xgq)|9`g7?d$itqQAo9GuRaUf16}Wr?#zxQ zaOsR$WVNfQ@zAwRnej190qNLz(E5E9j?0fRAG%2b%X(9KN52HdT>UYn|9%2mRDvW6 zrG${y(b?xUs0GctH|HKFc+a4vHQD*=5qoT+7Mu3Sm;zmH+p~}dCn==gj?bRAoq;T? z@mComTcOp-%;L$ZCHO6>z29GHHJ02F;9rs=3oXlcC6nX|ab)lC+q0Ls=MdXZXE{(0 z5is%>Sx3;LGS8q#xep(Y5?9&-d8=@(T`9 zdHC^QFfk~&xuy>p54WY4?WqNp+myj(T=0ZqC%82cfX}_VUC~5L(Br=lT z9lv=;7=~yAvww96!{nT2?jP=WAVVz1sw@2nS>SPFVBo=2=#23BIp@+bWG>C!YT3_$ z_MeBoGt?ddYsLKT3f3TW34cC6B6$RAh2JiHnaLva_sQ&O7R#8v-(`mywEdC($~bZP z2URHP?~Yn7zmhCqmN9?dK^=^C(QZjK(}L=x8T2c^e*$fL)LKJ+BEH?ea+kh;*>-sR zS#PUMJkeLR|IE#3#d>F?%N!L*J=qTp3%l!r?k&KEkplnmYAG_``W?TPDL%oW@`3HK z9c!T_`>lGN;cKMtclEm6Sc!D0*R#U|iF=y8`FF5l643{?HNa;9n08Wgk6dvgUlplw zQKZ0w|KvU6*;nT3TkHNHfoIFx8@5eKgvb7c2D!cM>x)=+Y{CDtkEf6v z?N_CSFm=8^x`wT@vs)ZlfBYoshIHP&blq;({ir)I3KU`#?!L<)#P6| zL?F$NcKpz+1<3RiK02`33F&(;owHFS`d8BUaytc8QP_8rTTn)}SgrVLG16|vU+9~+ z6QYg;m48*ygqO-78$bGf_;ZMhJI8*Fv{j;x)==;AYVp={dyzp2 ziTx@+t+y1eo6^5Dkp))Y9Gx+PM{Y3s+PL(GG%k(?2F7jAXgnqOom$xz&s>^vFkZpr zHs9?%Sfc*UgB|$Kyf62A>5oq>CG`H7E2xZJ>~p_0W~Vcdo*%;@VrRf618KrhF2o91E7k&zeNK zKP8kd@B&MiwttpyKLG<(n+se|3`5>#8_7W98ek7SuG|JwSeYtMHWeSAzR%wkhpY;j z-lxbUVzF+ZV3-r~keEZGd1hzh3 z#o3xFkH3fLDh{e0SR~j!#`3j;zaK~>jZ9^n`gU>u3Pud3=B&@&+X+~?V|?G6l4J~0 z{ycPW8@J8|+m+{hN|W7&t;HJ4$G>v%{V>?0ug1x&#J=@ji{77|9?$(4*PdQi3Q6w* z4maON+MV=ECtt;5W8<>7b|UkE9pEb1ayo=UsxHqjF@E|>0@BkiiFPK-x9Ho2@wySJ|p8LLw^ZF77813xVSG}v7*Xf|aXvFPIT zx+p)mK^8!IuW#YjRhp3I>)+FEau+%SDwE1jUxd2hsZ>2{D{LxX5~mRt4+EJC-qcvs z0qgae*XJdPx`qk&kRT21AfOa%mX`W)icIRDp8Eb%1)CjSb{0Ba!WPrfvV}Ewq2;s3 zfxut}v;+!fV~8!VPEI|TbxjUxt)1ED&Sv89Kv98^8Ihm*6*1;8w(}mA?S%{OXR;tI zq~Xo6vqz@wrrG90$wHXsE|u);z~nhU)E`sl-B<2}&PONC51d?q-C+~{EB9tlNR7pR zf9^Vo)E9H+Hy=$T_UV`^=ccj~b~{I{Cae4t4+W{E%!qj}Hj??Dy{PfdEdzEi`-+N* zC{)Cz1+UFkf{x7XJF>Rlg7&ntlghR^z{=}BcD_RlnNQoq4{qB5owM&Q_nun~os|;G zZ6b4^)4YG|K%qFa-|^rp_`439+m}s*P7?l?fX1s=_d*CfYo|wsxXQ`h6uw;xQ+~UK zA}xlbKmXPe=-hFihI{SNNI;QRPcH^0b0RyeT5}Utg3yG*>4C0E&eZ#gwa_((!8R%4)+b{5WpmF(7uzweH|eeM94@~G(#3-| zIk#N^R;I7ip%vVEG+^yi-_Md-iS6fKEt`4zAxzX&?;Cqe@Wz+67JDbyzr-4aVS#OD z7XNcx*!18!=&RI}!jDR*djAGG&JH8} z#^V@SIfCbqK5OKghm|@uooOii>R^LO{)(IRjR<~`fn_xcLx%re=dQgsj}_w1X+M0p zb>iiJfmlj4dXwhGx0`$9n9_ zN!GZ7)YdY^SrikVak1{beK1#x8~226DW4^~%OsI*JYj2ci0B{B{%YKJgDOoXEo<<< z`;f@DF{RI}UEyE^-MzAZpVvBKo6Wk*YcmL309|r~ugp{&s?QG6<1H$Xy4LJz!LtFR z`!1|mCba-MzgY7+05Lu)VaTsZRN%iYqegwvc_JX18R4gGqeo&hK z7CDML8c&UlAbsBc{-qzU5puBhx8FXEew?Z*wR)BaNM|S=N$RV^$T}sZO{2@P>%^jg z)0=tmp46z<2UuAjV6f?l7)ePU*aF*?>Y@m~;3|i&Evx=Ac-F1WU!Xx?pl=f(PSw6hDK_Tnzh z-i`KTfxg0nCDCi)%874xkiQj~;%L*JJuM$=6C`vF9z(jrj)Eq`Kuaz@7Vy{-67`+Gky(q_Eqk?eqC{y8(E$U;2huKfD`_v55==&WCL zwcyCrY5g`|xS;PR)TjiAXx}%ckXD$EFG%6y;Gnj8JW|=G80mG3MA&^qU(FVK-E%q% z*vQa#c=0DRAL-j%(ie#9Al0zH^fyMsXdr88)^|N*{=Dg1_j1~v?zu~{_LIjvhq2Y0 z`V@eJwnJy~OvwUQt_ufd5dBBGmyj-9kmKO^?xDbWE2O6V&#-+ya~A#UoFeVOxE^Dn zI@GUsopauTryh~achJpPBz_js`}gK1>+C^#PDmh);H@O{9T)RqTWp2b_5n}B+jmT_ z>%`}(M?aBKFyC@sd@^uU6@If`ieppGC%L`(e~>z`NKfpN15#IR6ZOluOl+2x_WpE`7vQb4&g7J4gZ$BmaM!{CpkL*L{*k*+Utm(96H%*Xxs&<(d}WH{Zb zCmr%4&e`G6@!qG+)935YhpeLuD5N9HytVC5AU#$f*zh?I?hHRjhc0kvXJK8SY1Rq64CJVTKO7Qw>f&B&JYVsdb zfYUC2J!`ff(stk0?1)=GEw52YvD*reW_c&;(&|~jDrwnpc+eb6RW|*yVZ24=jU(n) zfGf91cc;x;)oTe;O%7kK$_gQUG|TGs=l^H_AJw8~;l|LtIo;ysinam`XBXQ6u)&E z$pw&Ks$0Iw^%}9hA(irjzhr@V(a~>f43~c$qmw@cl5ZFz<-jagt6c<-J9X^W&Eg78 zr22jdl{HX+W~cR{%cG|-blcW>%^K5w_JGiaUh5B(^kXK&$pZbqvO!(Su9`Ju?*>_PSg4-oj5L#t!|Tf)G;ZV zM{MQJm8()nW+o3y`Bx)zf%f?Y^9?AZEt=W-IT6q=JiqVVx^kq8E>g7Dioyo+>y1af z34BFKq(Mtf&{ya>@btODE)%2$Ej+eALlp+sA3VK$A@BX$p;D@Lvep;6Oizo5o$AHz zwk8qsW1??U8Q*N0Me|}{_wPS6*JKXT_a+8xe8xiN#%ms=BRV)PuCn;)WdUHX%>G^$ zH|^(tPq3KzjqCS8793@rUh^OMm|gd_q{@er1!nX#SSOAEXKkhVd3^$hMN19A)9_k?NLGOXM_i8xVDZB8`CQYQv&-mo^orMv>iW|zKUqW+`eW1K- zKXxxp=?%UkgN#!ef{&6eA|q0Y>UP~2udlszn_pT2>4J;>lL~fX-Ati|@#a8$D;)B0 zd(5;vUJ>FiJJkl9dUu<<>IBalUC1|i>75fW_()AK>Zm$0*QC@waMQ#_eTkT)J6DnM zbRx>7ZQ7qdlQ1JTfDh@OE5xsDnLT~)#mJTBar2l^e_hmKcd{mNe=U>vQ6e9Rw3d}V zHuY?1XlnfRoYafMc{MHjQf@*wV@74{Jzr#OH~)p72ir9LB~_97ddHI;b8kYU1=;fDXfiUl8V`uP-~h*bRK9iT zW*A@er4e51LRaI0x5^a_z@E_cK$};NeYiQ| z{t5ik_u{F(@k<3jm8+Gnuf>yGnC{NHWSa?b%>Xf;<{xoKS@ zk36y%xnEqecK~(Cx9`U1^RRTFof+npjZBh8ea){qP*oMUlfU%~uzQ{dvm}#|Zl(~S za3==orgq!+O^=AwETLv>84XOG;5}FcAx1+#!aIJlx@cmnTDp6srit2 z?uP8pZDr_ijxHeYtq10b@ZxlbdSqUa(kXhvji&>>_|2$T!78NBIe2inpb-{vJTk)! z;;=fv-tesKM{G|XbLd!QfpqWk^&?S*NK3Kx{^nqbRHZ|U^oysa?N0wg$9$}T^CIQp zquArq{x4NxM&>(UpQt8R|4ap@99de~VA@~H$SRJ0H5)irdXMbiZH5(N``0*U7f;JW z&(#Hy=D^zI-ngIS4!!BuKaxhc_al^st^IuDjxa_QvKW(Jh&~&%jV?O{?FqZGKKx@x z&L_M7b-j4@BkQ|Rn!y4h-_z9b>73Jrmq)%z?_y}o6@%fT@7RA(6w+Q{ z)@zx0NQ;}Z^yuoBNK@5u-o0o)()wGQ-iRt-&n!s~@5lSGx$bfJ6rViO12{P$qaqZN zjgqy=*hzq}+I<_u21j*~$+h-Kjn6yLZ$5yu-d}cwMMS*3{I;*_ zuN5B96|Iu*v$+vHm3BrgDfNb-zQI##Tjyh%hctb=W(iWQZ0L@DJCVwFTD*X_qJIQEZ@@9H?@? z@t^tPf7`3`U$4BoVKGvhR|q-w5&g<&S~s0(Ho1gSJ>J%@DF`YE?TqQ5zVvgYWgkAc`EbK&0lDi>@B z>PXl~9)iKaM*G9sCQ!WQDyyP06jZ;th}yxpzXV9QfiPR!%T z`4hX|yZ|CK-vU|vpyCJ(cH-vW_EjxZW)=rQA&)n2o%6>>q z&Th0gWr0&OYWKOOm=SV0{8any$0KCEN5!MwtVYOxu_{e^*9&YG8C3L`&kut^j|TR= z=lVg|yxxzrY9_&d&eYD$jnOhh`W9{VY#~#~D^l_)(WU@LXokPjZ65oFke5a6d-h%Z zfs8bfr{~Tshnjh6RY9NMBhBTyo?YX6V7=he7YNZq+6hmqU+L3+$5G+%-Xlc)GyUoD zH(pVpz*&0IRp>(@Ozz%xc7AdR23FfnVFBT%vzTb^Hh20SX3ZI7J?n+K)z;@W#$Lh_ zjrch`RMmhr^~CCrK+UxOXngeAXfghMBa^9l><=`)tIgkcM+*91MmqO=aG{X;)Yi}Z zsRb=IRjlVjhV75%YKuQY zx|iB$j|exPLKYBr-O}Ob3+?o6RH4v+@|RE+)=bKEvp{0tr~# z%`Vod-{Fxv&H|Sp^X%Xh^p+WHUKsI;=W}fi zhog3!tOnMeKVb?lzd+}y^^cylbs$|OmbJdk~5FFH^buT6g+AFUX3F@>!`}?d%4jU1f53;q|cFSY4>H|feWJ1rfMFl_0 zg||R&@w$p{1~$M-y%kuu_$FCkuSUKny%G90M4o?qa(ez+pEjm%-3Oe~txmLN{^@(| zIlmmy_Zd1%%iB3134Nwc(RyM($oyB;&i<14h}3H(_UgfOh%b~J*jurkiz_FS;%}7me!u}X(o%X6X6tiCU| z_c_wT-Y6_{xlPEK^T7NH?<6w+r2Q4;!2~S%ky|9{>Juj&X=}_H!z9XpC9q9Qi(&wi`7RQ%Gh(6lWn6Cr zCj;v1lWa=lu1@P`z0%H}rPTB4F>^6e)Hx5LoL&=1$feL56aK z=U0& z_#D}q_XEp>&hG2`Scgp#d}lje{D6+)18Op|n!wr^aFJS=1f7lpWYedYfYoC2Jc7x@ zq-P6j3|Mj4e5ZRzcGF|%%$!J>7@B_1rvbWBIt@5mUih`jG$HN&T(SDpn@F2nFGn&m z1ZwN0uiebEJbG&Q*+gUdvR^#sdVs-hDZcxGICrbws`@k&ykLx*cS1BS=T6&mtukg| zemu`*?(S8H_NsyUO%Aix%T164&fY(JXLdH^uiNM#o#p|p9t(VJ^lSd5KY?{gMx#%K zNfz+^$`{2fgyD4;=B^FvwVH-U7eOkRmH2ZzCfkV9eC_GjEr?%tR2)-fL%{oFDjCOy?(9E)a`1B zE~M@L5gY;?>%@ku-Wx;x%PINsvQ1EdOF!*B%;g!!dfG#QlggX0Ibxzey8Qxn^0g19 z(dg6XY-Zi-#Xi$^fsemoBGHGDu`}TNvR#6}I(_q+tm96yfZ%c)ldm(7ep8jMp@-0A z5IT5dMl{kJs$MVSpM*}i)dIAZBS?LGM66|WTAowlO7l~;1GQa*-n3>{3l<{A`Koy+Nyr%ulnDSnR;UJW#w4l1elx>9Qy!O8Yk9X3|tNQPh8)h zG~m`7V%ZsWarqzZNDUCv`#vm5A*IO(82!3|^q9z~&i)82_u2KU;^>CydwXx;o2Pp) zr9;i`fTAeAe&%-cv&{6p@^Hgx@q-tjE2w`Ft8X6+^O=nDPe@@xY|gNc@(*A&eY)mk zU`X6UR?qVmo2wJ@6|RP}+$6{Xo?-sqpAq%ilhiqz2j8=xr#44;lynK1e2_W6R|uJf z-xsF}DL~e}V^_YnIU#+SjDXe4vGZ2uh+xvtO?KNk7q$6FX=hdewnF%R3^ zN@oUK;o>5u?ULaYxl=9B5;wI^a#kP=K2077mN)oWM6s}Pv!^45jnOeWRTx)+?o9shfKmNw<)PWugSb~A)#xLx;3ixs1<=*V31sT zm3@?v8g98~O+plu`tLh+EZh|8m1fWU6{U>r`LA*o%W>=Zp(Um8oOStW;yzTJcrKb3 zi;*N-=hh!*u>PCm&kquXz}as-6_-4KO*^QUTNXxQ$InecXBKdAGmulW>{E=D9I}fy zEtCGc3WmHd%MO)_AnnHeDM?l=u*)>Jkt*YmHbsZ-1Ox0gZoq(T>Cli8^U^r^C6-bq zr9-<{AwzuCIZLAs{5@yZI|VlrV4Z@m`Uy+0VcYRN>S-sC5nKP{VKu=EL33f1*IMnv zx<5T#ACFf6E8ToYR2dDash{}|ms?ERuW|GfF?~p#6xdv-u^Y=~zRB~^*9T78>0`Ms zyO1v06O%4JA8CR9{GYO~AzkfRwO{WQd^j_TWudqZX_m!Iz0!hd|H&z%X+#kQnbKR1 zJzj{33*+>te*Y04e=WbayCMj?-?wk~T}a>ysad{`9p*7uY1dR8ePk(8+n$!jGM3^` zm0N+&IyNA4RbTwvFaE@HHqY_6S0sk?!Zp3OuATtS#%LFvl@Um9{+)LUe?$As{Ku=4 z&mrxy^e_4GUr<|;8?qnzp_wtq!8~Lb%P&9jdbH;uv;;@2Sb5+TOx@q)J~o$&zkDw@AmOr_IFDD5T>vkJNNU0b5_|%%H9TG`_4C`Yy`b zw~Su;eD+GixsY*4X;TN^dFb2sn(f;wj*0yJ`>(v6w#UwH_%)*DiIvd>JgaMuOI7=1%7U-anH2%z5i4TrBOa?@x zLEkfWg#F5oNUz?dMxJ#P%l0&7jGuT+_%)In9cNu$4xF;QS-NY!;n48Y$1NQMzcN!7 zPB?lI`5cyPrF)9xCHT8mJM;SbVxD>abl@x#(x3J(zAm#1y;+I8-tkOZ2w30Zbt^6j0Y{}MO3<>)GIhs0HqtVaacz+cG&V_WaXgs_ z>@B`iMh2UJQ}pVb>Q*LSga_kD(~UwKt;`#EL_qFY=x+{>$3{@D*-V4xzn=IxQ+z-c=e z-mTq=^vsTJ0_8JcNVey)W${^HXC8cUO@AC(96U^Asc?%-W7;<;iXTR8X3aK_1J^2|H13`WJ$#_PaYf- z&$`!LGOZ)&EJsGf-i1=RT|CdNVP9WrGDOr3wftJOXAGVeVoTCnl}uom#U zG`$W&dj7{{EqgnmWk}+f#lAO4Klbuy6*T}_$g{M*rRpO+d4K-kw|;2l?1PZ9&}l!} zWWyY(Vq}8giSN67fpdDrNAbuMVBY@7Z*FxMo7Xn@&FQ;}QD&QtpIRy~?XUKV<#sH^ z${T;~$94K(Mdh!E4N(@rK9HkkdOQl6w@LE5+2td1qh)!o-7V;QKO9nGcZ1m1=UWxu zeAOfioFCsCpmP~GH6Ob(>>_}3^|_!9Jqp`8r}P|`6_N$K7Z;}t?IZJ@vop8$%tZPo zIazyw2JF0^!YZ0eCJVS3OFu2WG<|Na1-bnWgihCaH(Gw30k-N~-G$%&qwnGA^QqoP zYq42GYT0=kqE8=VO6obEb}ewUXXTzRw1p7o%i>GKhRA|y-YGiz|FoZw)+xi?_W(jrXW^1?lOxw#>l$NjfPT_vP_msU$%&bwQk1gnV@rby;II35lxmIW) zJt#V2f8|4@m1Nv?<~t3QuSM%a<^1vTW#NRKABwU4r^3iPt8>Knw6D?Tl-)(zBbTKc zZu226RZMK+;iVp(#Z)KYgLuY zmJ3oC7Y@usM$S*;zzH2-&57x)I#7vKU+pPpHz*@L_r^Qd2}x)*s_;2eIlWKc zW!}sW)IgfGa(L_44`hMwR>#lGB5*1!Z;|vT2Zf2>ofcR3iVH*C!3C?A9t$P(fWZ8Q zxXU9jl)v-mwH5Zr=&0}RSsVq$mg-B??vkO#arZfT@o^Y>czj0Oa2jylPKv0LO`uDK z%3l5VA@tqcTq*GUDl(R)zh;?bkOg+1>p1w0;2os4W-!m3RpQYr%@4M%j4KGjo=rQ& z%~CHTP2#%kE4`yg`%!<2zF`E|p0BejT&;-pA%870p~hi{@-fPze~i6qH#(+7GO4H-5`FYaqQ-(=`RlrT3Cd`$i!_SpKK}U zY!zGg^)$g-!c2Y<7a81(RV(?1ZtS7LRB-ZYr`Tyb*ZrY=T-K^-dumgFcMw;u(a%o) zme;5R=1U3wyuEV(TSIL)HVNjyw)r!7ESu;LKg@6PS3qe$cG7I?>wUJ71?1jUJYS-X zwC3B}kH!(W5a#)iu!rRlSXS|K(q8rnR+*oR`7oQP+iGq+IQnj<42AUW)A2<<-;wbq zur5iK4**#}roUMBjUzVd)Bf@Eynv-U8ersaP|&OIFR>{<{iCKlfy1ua@X|Q;^-CyG z)XClGHwqlJTG{dt0-4*ar3IXv!#I@GN>mX?H?#x597j~ zPiFCDKz*U8cEjw~(DgLoO@JQJcY=DqP2=LiA)pSrToP5JApOU%x3AnDq({7t+aCKI zSe$1@2ZOIn`(d5|Mp3_z#+dyrIf}@Ga=!VyvWA_2qhwZXDEbt-ydTFvO$*Sr#)>j0 zh`ut+9iL|}+}j6}v44i=`CkH#S@ETbZDG)L!|RajPERtwG3ly8*_Z$K_mEMk;PG-$ z7^$<~ZQrwP5NR)7!`R8`bH9RR#(ZcEU9+xL7)ixLm+LyYpB3An+w%UR1D_3%8QvaS z{pcbxCid@b+*(K`xl0{Y(B2Apf#yHoFN}xoJHz*M;|V_Z%zIsjZfZS2nyp3%YjZwy zsTmktXB|lG7WxA#YO~?3-|2f?_7s1CWj0#nSl2_xC|jPHh+{20~r$U)Ad!O2DSOQ|$-Y$e;+63OW(~ALEzo)gB>Rq)Sbe zZGKGfxzaz#If{~W{A(Wvv%z_JkjaQ;{_G~0mq-7otv z_~V(h$iI(}&R~wsZA*aOmt}2Y8#iH6bL+#fFOQLS+i~Np_9_);FT}k<0^#FJLg;OJaUv^|;%iDK~ zG9r>lTd0zxQODa?veT*htfVgjJJMy)F=stA?f9(KcbWz4J3{ss($s-{r1pmMvE4Xz zx3l1G%=T$JM&_cRGWR?J<;gP%(d{uvI~U@xf9_lS5#w}x+1|z2-xHB9eLEh;EZS?F zbM8QM_@vP8HTu977!{v8(F;RHD^!H1?Gzd=?k@J_)~91z&QqttJpi1eGcB`!*pmfH z>V)Q)*&%)J_U8CCqR80faL-xm6tF<_jmgSFqzQNjiii&(L)hbU*3aY9etoWV$3;hc zEheIrb37B9Vnnq%wsSC6sHN9-?;fBU&2Ejiszj=n!JV0Bs*tf=()!%@zc>{1{8{wt zcw{`PUmPJp@EulI?a~Z6z>kc_V*^DgNl1@UJr_22+P;63DgC8=7qHI`pHgn>pVqVE ze+nPSQ%H6DSC~d#!QqsSyN?=+&egZf9sg8Q*i^1&Yx9Ac~$82OsnhIFEd~V1sKmi%0Oy?z@8<~>yR$om!_3v z1T{a&=E29g_!H<>SWgQ+za2PJq-`|YU07kGF~!gSg2!ECfk*i^$B&;R3pB8NWU{z@Pp0qfuQwn6%?FOrnPsBy zWPp*u%29lJd<@_Bg0DM z&*lVkWVUxc>;9?_CF?KNZW){&ckrxi70V4s-#nz1c0F*~t~#T4=-wTqp8wMP@rV)9 zKmN#8{pE#p!@QN{m-~<&Gg%lin~l_xMp^GozkpM(J4!Q^#pLxjGwSvT@|>3sMJG8O zq0xjsi23-_(`Nt~CnziJq_!hN=6A>~N*=H`ykOEwe-P_D5_j_Wb#C8MWGqpyPF+Oc z5t?TwokL?GvcUaK8-#AzPW#_+22U%wJgzYI3BwH<-(ZvCN4T^COQ>G&l3?HI4s5p9Y3Q}X-^NB}PnxfUfj`i^1C=E^PX<@lDzLL;B zQ87Ekz5=OJXS28dR)dy#b1e1ug#laWVZl_{DP$H{t=62SM;172w^Z0+1R1q5r*j^O zO^^R=Q$X!Z=rZU!v)YHt>q{26K6RTxe}?p38M|FKRzYX&$7?wc3H~-}!an)_x8=y3 zv3%ay?XR)%O7UpU_z|2EbMRTIv=Mtw5Bph!{zYnB*j0Uw191FT>JQ(#PZoG9n#ozp zkF@szGs6A-k?wMJ&xc0@KRjoXrGiNBPh==liZ6?ZQApSRE;k$0AoCr5f7~+mGg2?= zdPMdS{7GG}9;ki(d>5K`kQVyY2|%4;3xBUn&$Jyj)ZSvLfD@bgWMxu_`oj=nsgJB z-X;oc^(9g%Sw%P{<2)l=T6$W~7w>v`eERMj)evEAwSwA0@wJE797bB=^Hu3$J&^f< zEEIZsCUI^$H|08BByfq;71keToAU6AOxpo}Yr2i5?c?hGy9ewUz?i7(oc;O;R5K^^ znc_tM`37p5aly;$Shb+2+a%yMCjNPUGfF!G>89)A{)Qq_`x(M1zcnz#BxpmFnj_NQ z?P!r3nm*6J{|;S?%0b$s>cwD62;8I(AHG~h#j?XW<~4L8563)W`(@x3fy-k&2uq1J z-h>tE-(oByw?fIw1Njd`N`Qq%-_PtF!X_t%*2J;HNS>wBtEHmVFuS&XOaZN8^-c zYs3@%8C&_!mz^eWhrep8oq8@RKz;w5oXs{3kRxn%`wVm;^GW2FF(rL$>Mj`0n&$(o zZ#!r2-J}UQzYiXX?bv{AuZu!g3QzBs?#Fh2`3Vfw|7P)Q^B|T+urt<{bRxYwvMp16 z8*~;oYqU$QB=nHY3Qg&<<;V#4b#m&FCdT)^>i+iguOTb%rT?(eRbXo#b3JxK1IkVX z2|LPsMtW-N#Y*je;#6R0t$|JERti#cn&+lh-zWU8W6vm+MFejuO>Sr4-8KA3UoLJ? zQN-0xP)fSeeo<2$7*uE11z!j|hRxbPD|_xR(##!a|90aLeo^CHg-h>v>T9|GuM+DT zPUze+J)WDfNn_LYsrFEzuN&cShlMuEvZF?^Der)i2IDz&-P_<7|L!-^Y8*YTM(#nz z{PhNnH`U?&`sKcb$_mROcB1NAd`kd1xt{m10l!u`;On2#hzp<PkAu?&!n>QkEC$Q*F zkA~J0;ytufU5NyFqVE#@%-0O_ze-5g+HvyvMj{Wy`YCZyKRy8J$j>`E<4Up8TF%mG zZ8bE!CY_w`ybVXJdsi}^5Pbl;=UrvDEDqo~ht#~aFO0f*>JW(iZmv86_qjL`$o{$U z@-gAy2*{nau*Db<2vlIBtCp{8=+gYQ>In=n~M?3lC76mIsk=?Y=(fQ(iS= z+#nER7JGi#yITZ$nn%&Vi(?$;{ z?rL)^9#sL`!aX~J&lJXQnGy$@ygdOgUQM>$OPDs<9iW zY((aUp=A{Nd+;Z&ChQu)k4nhj{_|VUWH})HuJ-0+VRK~Il?1Y;Y@k_2FHhQYHhLZR zQZQb|6VHbnPR7F0DW3S}|MVwPjIxrR6e>g00sp@erhkzk|9pJY_)KUzOYZ#PZ9i={ z82ak4xj09d%oMN{eMjJ~x;k3CCT}M}aZzW-u4(&os(XuFh|M}^)r%E1is9*73}f#H z_o($hN5(6q%B%sF!Je8q`+ZjRr) z-sOfjCUe_gIvnV$aVb7*lU){_%WWI?{HT-Kt0o@*S|SDJBB z;62h7-CJGwM+T|ac19M@EhpsS%a5h`>I7df^~~(~IeXoq&*<#UrBwyMu}@z%mHK+rUrJ5H zNK&rYm?!eA7-(y|;6loW1P0B=au>Y_; z^iMI7et>g6eE|ztRn)`O^AgxDHgWm*P92{4bQwE*EmYyydUov3fu{3290nmu>ERNg;IwS}RU1g`UYHU+0w-0h9LOq02&oSC&J%MO{31 z6SN!_UvQb(2KDs5J+e%BWKMj(5a!4x3kcmzp6IlJ!O86o%4ar0zs##hZR2q0vKyhD z>f+^*W91cwSH~>_P7C>Zy!JDm^|Ld%IaQMLW7-bakn&o}nD%46tg9oR6LK9bO_z9@ z2hDc5W}2#8{x=F~zny*Bfn207enOqJZU+u~=hqCEuZ z5}7REDR#rDhYz}ZPo}bZ17XZnUHI~|Jus{m`?Dy&655j13Kzd9fzd-*K4~(?p{r*8 z0`)Jqk?JeF=UomH8{Vkg-b6hQH9B(1BsWRuN^aY;HS;vI*ku+brxqb|dzp25=VR!* zrS!-=)ESw_4GZKhw7_5cfw+|1GH43j-7!PP7&_-Kmj7vV3~84K7Vj(G4a~7(?aK#d z0&BC-#y16$NIN>X8XofSN)md?XXOg1seM?IW# zf--BL4I4#}1%w}c^!&aA!xSZEDdGT@%C?9eTzLvQtq&d8{QIAM;wdCY=GMlr7t{Xn z88xA1f`7VhjnRg*widF$yr(LzGzrLDtje?sSqysK$;HSbfR)2 zHofr~krXn=G}V*!e|HFjOjb>|}`VFKd?ISJI7K0Y~ z2u^1RPyZCiQH|zX*C`8w7#g6vg1|vmz7uu{o@GQ9_>{goZ=8h8%`Xf7Hnvh#XH>xdZtRrBXMAA4MOQ zuE`gU7UXpPJSU;!h5fOU!zXSR@vIx;;F8_Dy>A0sf?wM6uPBAYiEnK>ZVXNLI;3)# zv$5!wzps;eCX}psG$V?g3!KZ|^%Gb3!N6=U(Kn|fVQ@C>QkwW6g>=*{JAz>fjTNVQ zuSgz*p}%vt%vT72!R@mmWPB2k9+i_m>~jrTP7hhh>0QOtvx9&7R98UPks;|7o9P&2 zKSmc4n~iO`^IWW6ufva(c@o;0rs$N{mL}j942^z%i{`oQM>;jBNbE%fv?^52n|ES0 z^tfKGwo;vsk*RGDAN_R3#Qwc@vri$E%vN_E`rwOnCBtVWvRr=x*gFr4eYs$O1tyQe zZq4TMsE`GM40XSLX-(8IW2_GT7gchePX|Ho$(qXgi-W!Q(=u zEiN#A{tKh|B%Ewwa-i1!v0?bB3s`mRzMM976bc;UN`|jj5b|>Sxmn~YRcMrYe*D$q zl}LA(;b4Y0vFZNOnxrXz;=J6Db92}q0n9xfDN9r)w~1IG?eQ#= zA4(2bWqNGY+~+fp9xdh-Dky|gu5#z`lo+;Nja0C5bw=906w>NvideCrzGot54@Q|E z*{pQZ7OCa*pI?u40y}8up}t4qNPku0TCqSGSR13s{rb~>R-2t|)a^E;h6xq585zNl z<5fQEMcHJMkgt?ppDofKt!Mw*tv_uS=)K);NF@67%Hz1qbhnGl8l*zs-N5ejMjzDE(=gD|GyDp5PU zADQ#IWR#=7AYJg~k+{W6fb)LMLz{dmPX1EME@8Srx9>T7y(g}~alVq%-b2O`rNZZH zzQyA3-qT)+X0xHQJGuFm+(BrVSikYZ=~A9`?uPOG?u+*UOJ}X;iu01d?si*vy!Isi z`Yt1*oB9VspQS9jmirZ0VJA1WWjz9xZ%~auI)Q`CT>UdhUrGh?y%zYbk^d(z2(Mka zNo&dKT?B5X;=BF*WJfWioAH;3xMl&zv2OTutUGotTyLs!k_^337fgmqkHe5kk#Wbq zr&w`eTQy&6I2N5U-=VB|8@kli7Y%v8fz}9v`-16&M0r)0$#_!WcNazQrxr(~k7Ms+TJ{-1>e@xV-IL(fVaD zeC!jIGb{@AajU-!{|(0SX0o@&rJiZK`_ihG3=VWHx%zwiWeH%-upg9{+=C-#HJfvz zk0Embryn2uR~{>}z>l^Wej`+9IZ{mhEzjjAfZnd&fg9`VkWr_%vSC98FzF>}yH2i0 z>Qm~A4^RyKBiFlLY@3a=TtiO>-4bMeGM!z1ikClw<2cUPc{LLT-?pq=XXJ*(R1fWj zSv++eNI(7Rh?E%=LbF zw;rat2R&`a=M(bv+UigHC4%>o?zAVzbbklZx6AEnl55BOLk=|m)6rP{bFKx&fCN## zW{kt3)<`!L4QrgE1g&=?lV7a9j`aQG`Gdk|q5Hn$=PNVTV$-qrZwq61>u`rD2OFvm z%p>G%%dGaoI|10XBVeBYQ!;jHpV@nKcoWh-b(R$$eTTHS{8!&!&44PmHEN!z`zRzQ z!QMMHxlnH^C>YxH1K7H&YBZ&8PWw4)H5`0jW7W>6EB>TgIQ1g8YliS03hCgrF4dYA z?6$Qxdi$&Zip&&WjM%J%-W#NC6VvCExy(FUGkpOtHyqj2KSl7x)VpkS(-~Wh%-p+d z-RAea`=E9Gy|Ymi(wkDFwT2HIFb?S}Bj0}hJcako_n#?peh%#U+Z^sIhCs`bg9hbWg(##Y``@p! z+JdRJe3Ioe-eXCF87pIT99DkO*k*D16qcolpKRGk#4+nGaWZ!{yuwoED8K9=w@)9^ z_W$Y3UEzxKU7oWaFZbk`2SeD)Cu$|ZH&>b(e8=Wu6PbUcImVW$3H?qv`+v;Jg{m|I zEfw?!b|&-JS&{$X|DN-7UG2;JK-*j(qqSogyI)P%#cr1a?~uK;P5#qz|Mg7g-2Px- z24`lT{(Tlxts{=t%{76c#k%}4cipGg;eF1<(*a0}4_$vuYFdwfS+XqK?jW>N?X<)d ze*J%0BHmJY&^~O4@O)w<*XA<*A z32DKKx*V9y2|ce!`6bFvAiTD}=STb(d%Z+wklsndb| z!=P(+`;ULimy^Zk%-q1~eY_pKQ|TdeOH0`nN$Vo*XZdTNga7;NePTyw4LswlJf`QqW2Az@mp5XXtGtIS5G4UCs)=}g{)Oi3 zUEPFz&AF@|)A9eTH~j2jkuDGgoQglW%SLx#vx&R$x`k#)->X2O#J?fdLDFXQ(U}#~ z{t|!6+wwI?pSR{!uuTn?f7aI;4auDLPo!>5-2VXG{P(&^SB!!6dMGo??gWm``)Yuj zp97~=H{<6DU!;v1B;YEd9&V~SsZnKD1aw8We;(Yk7SbjagRV)NL*ycl+_R6yfRj(^ zBMUDAj?2YHq_kg_a)dxg9!awDPVVwBkkNiLLNB{r+6Y1xZulv)guw%m`#RKV+ zkh#z*M(V0Aue|r&%h8kP;)S8oyOlxff(DPW83nXAL+3dWFkVOuzeZ5 zV(2b6zDD6Qewr1SVTKHWqA5R*@5t~zbRbir52y0JG|~gPaeSnmX!#I0gQqVqS-~WA28TlPcRYMe$e6uRSAu*VyZsaA zoSS-wi77tEew>{1U!PY>PE#+q!h=^Q?B+XPI&SRcJ^x;k#q(69oZt>#_FKl`fbGMH*+cHwoa(1t zp16xb^84^MTTLh3T6Zk~u1vL9&drFZB; zmb>ATa+E{H%7ttXmf8Z`{uLD{h2ZW}3lO&qqjGwQHN~ znGKLu%3>p35Uo9R8{OA73njhEpHTUD`YYCfVdUA@{YBF;7)*FP{Q2 zzJ2uN-_$*jw)|v$!Ttbb-Fvg{o%t1{QPQ3zWPXLwU+K-p;XJ&^kf$EgdH@x;olL>b=069q{8aXi#x2A>KlbqX zX79T^cu1sfJ;$^&0*=3diKUok5>Dm$9{tOPw>fWl z;#vQEf2-)ZY>OQ{eM0_^@5$yslg}x9?l?8#lYEW62dUOYUD8?kFa^77_xZO&_r5D5 zrzzsd+`8)Qof0z&=}FL4@#<;&<>7+yS0Vfql4pOWoeF`cq{T+gz804Xla=T)bh935 z?}XZPzVQFwjt}ZLs5FS*=Iw7xThes6;LsbMzX$bQ`a9!}<^Op8G@ZQ_u6zW4F{)j@kh-sPc|X-3iVpG%L1fd(ztwXgsZQi_xA3JHE#>J?~7Kg@ZAEZ zV3_Ziqn6As$o(1=k}67!r$e~w(EZIk{QLi14wOH6JRarkFFj=V>U@?LQBVGR#vtcs z`~bG3$!<3Kl1(91aqjO8$^0L8ouvTF{rT(Ygqb**wQ>CFx#hsQ)aSQ%(_`-E z>UwUQeq{30XYk%%Q;NT7y*&N@{<&|J4VrP8CwX{od7oQcdQjsg7l%vcf1Oxm@caz+ zNZO~|{lkmvtBZCZZz

    uEJn>Qs5EnKcp7Enl=H90}qt!Za0wm>qq@N=*($#;E&%(D{+yFhmKepcXz7g z-}%jB&;F11saJR8^a-^`&?WfI*kSWxvVabM!3xy?C{+?p*J!^0Bd>LhqXZs9mw>*| zs<(eAd^^4yau(ZBNDo)V`MAXbJ7=p@NcBeO6Wn#7ZqgMv2cLx+#|C4C)~Q8;LGRJS z#bM@popOBjDe=a*_jUB!_Ia{|w1ilHi5HvGVh-@^^VCc8R;@2*fPLoB$xK-_tezv9 zv}+}S7vXHAeZF^25ZHRp11;t&LG6-)9|w-!hfL2OqlwEFPP=sYYJaO1)L)s1U>Tmn zsm& zqASoP`nxY6j==L4`ZqeNfdEa>_d9` z_2*gph zwQQdG^_{I;(=$2?!(3+w9oatv*bnHgsR0+Tsi4Ztebp^!cbip}8oCObR@4ir>Uaa= zYLJn4rzCXEVm#OG4uht7x>oEL%aJ;9?$en!)?~i@ksAu7bFs0?L3!0oC(Mhm2_O4% z1Pi<02<(c9!^E$$_ihW_g<}2X@(-^6g2_IYn>z66zKW9gFBh6#T-gF5E>t{)|uiEzxX;T@plkJMYi8l@1*A;@arV|!Q zs;(50yMEcvIjgZL!|r0~iYw6VI5}yz>;rVE_1@B(*FolQuv#?H5>6Hjchvv~1%zKe->JH_K|JlhipYDCfkI2(R zDJ11{Q%ibD*xe=3&#}~l*0om`7Uyi{`Fy)0<&mnB4~3L=?Mi|@(VvLZ8fvsw`Y+Nw z)8C(7)Q-$t`6Hib|52xK$iaJm)E=ar-ML8Vmprka=Ws3gS3iYxShl*cdkc0eY@xDc zvap*R>bLJiI5f*Y;@q>cKzf9hsxAE{(ig=%`+i0pX}0WfHHI>96bp8)m+&GBRIPD8 z^D`J}Prq%gu$;D+YK!ePREYjEOmXj{f>UgmSUwT{%b$qbvGtv)^M7XpXCE%iHJH{1 zK~n3TRtrJhyV*WvLIK!*#WLIXp~kelpZ1~Xr~~_p@5oiVJ6JZ)|MP-*+&Wvb0LA_X zU%Uk*8ZP~)IrkRtdC2Pd65=oyvNt##-{*J|$GsYd@jeWa->6 z!w47()bLs4XM>DIl1uLvYI^#vj%g0m-X=MAgSdUr2u~l7K*Yk}+5>&NkU7ZiCsoM~Bfefr& zJHtzg0>}cJ;}^oQOtQdXt(PICrNCA`LM89{CvPdzp0Dz5_)v=UT~Yfai@4_;yH^be zz4j*hT{GqwE2k^JMy8Eaq1LPW#5(TM$~ero9qDG$P7VDlfU1%j`yu!vP>lsn=Xr|a zRQ%AmWXxXZdM)`fsxF^G`Vvy@HK`5k{Vb&=y>0)eazhr#%VrPl2tC?z)}9^9DLPZ}dndhJ56PEOX-ZL*nyxGUe{6>_H_OkPb+`ih4>VpX z%HQx`KVQBocYw2-$6vzAmGWC`-xD~xNLQznh;XhxqmV=wB^Q?R`1ep}kr24|x*qR* zf=1kPq-JAd*IrlM++Wz6l(Ez!gUGwKzMA@A`X?28R~-EAbBZ5(=s#|>oT2kPXYes| z^yTHAf6k|+uOftUu<3uiUR@sS@wAm=STOtZkGOUJ!6RVbR!$zjfYe)m{+``&1*lv8 z94!9HnZAco=2og3Azd~ii@b8-v|U=UZLW1Q_TN>_Y4O_qf4X1)-}(vc{17_h(jv0J z^}Eb{D{0XBb@8CxW#S z3gyG=j$MnAMQX?C-$t5ufiqeouR%#is@9^N1^14T1xi?c?P32f9?09f{5zrHY}JPM zrNjRpkGr+!?Q9{n&;Q&XYHIz#v$50mZnU3K@*VE}K-Vspk@@FZfmKzpr_1=i;$4V) zS5)T2oIY-ThC(X)-fj}zfGu~fi;vp4<7BJ<<|#cv>@N$n?(k2=+OL_{4J;8#sk&+J z;{P{szctmWM|@u3l)~xC)cK>(ecL4Iomv*;dn;+uhxLK|w7i2HOYmmVZVo;6_U*tP zJV1ZD{{VD-S9^BMhfN_R&db z=IU*-zy;bChj?yCmDgx>F*2=}bW1i8 z{uL(9esj+yGOAcw@3!-*vJR6xaf%G*?f^7LG$bu(U z-!jS`$DS|&?WBOs*vev`D7>LhAtlM{1vg(n`W;D~Cw&ODi*4U0XA$_IiP?))Ro01O z$&T3fs~m13&7c2PY_bQAE3IF$;6orXdV<0p%-as>7f-G`{EX<+KsT^FFZL}_K>F^k@Yz?cBi;Uvjq(%^zUY5`PoA*VG(L6@%f*_(0 zMT$m!hE(=XlERmAqz;}F4n54}JH(zfXS3(5?!xYy13uF>C%T#DKG{v@$-_b0#rG{G zHUGT#fVp-}*om!Nej4DI-Jy|b?bx4l%ED3Tda%-p@06@?`)0p1hwr_m##q zX!pG}tS|HA|90K~Kh}Q(#r+9g%KtSl|6k`rmD{yx>f3xQKYiG3(n$dFQY}4gG9!={ zc5;?qdO6ZtBOCiKs^g1$e&_UU|HpApq*dxrJJzoJ@9Sw6N%udV^RNF?NHZ!mAJ7%C zJ2=Z&W|t+;I^-;5O3!cL)~O)fYkmDg-`V(nAx+S&lOH=~4xNp?#m$T1)EU3W(q&=D zJmX}&)n@{_E!crlhf1-V|9zo|$sAzmt$E>foWReQ&Y)ZEm3d4SnKg6kA8&$}eKMlQ zf72vU$5m-QH>>B<5-7c@w|(g-!EZ2XS-9D;vj9Jcd4!hvJ%%nnt>B)1ZXGjJFKyYB z(q{t=ceZP<7_jI5eN6S2v)NfExZ(fiH!D_x(3cUWA@Ne&xI6T$d9^86*9g<*ZOuE7 zHUp|ZW}cJG%fmK&w)?qaDmK5#Y5hH|syp0&#E`p6~iTAW(I zOc4q22j==e)r`)fo`v9p&o+vE+DGBf=W~{N>yIgi^t04Gj$n^sOix|P zD~!`PtsgD87xUNeD9ye~#Od!T=X{AFb@S}|1m8g;&sZrK+8fq;>F^B--=8Uwhr(ZR zBx`6&dFEEAnlVRYfoBCyB}A4;^qWF2ZH4o`fpIcPNhZZp_XK6ejZLJXv)wpxdxO(n zDKhjH{Jmjd)Pr3*%d@XF>;`r~m6x0I-+YP;14}TJ}HwfDheF8GY6{X~&8Yx0$73Nlv zUts8@L5=~X8G=5>=aT>F45fuZRx$Q zlvfb=jXLM^k2e)60Vh6>Y0$(E9s6fXuFp$A_EhWE^aCfL=}q7@vhi!|JFsXTYoZsL zDhavluXIgC6w;t4HV1>5+TYS8C+%}o5+0~j53a`gz_f*Je)?W^MBc-Fs-Ii?w+ znuW}lrT0$jbM0~HE6}(pX!(}=xmdc%y6z!GEPogtH<7;y`%g7n=DOPRJdb`M{K*lq z6Igt!fAo630hE)r<$Nj}gH$ilQnQ~#znC(Kasz=Uu}E{zY(DY(JJ0?MzgllNgC>t< zOShd2R2;yDxH$rw+zPR6EPbQB4cGnx`cr=eTN7^I6J$NL3TzFP!Fu(vuqT&mpix5k zh@7h}(0ZOQs6V)V9E=|RP%FmH!nW!pkr+}Hh5x!#pkWs`u7zQX2j%&+H{dAWeq5`i zi~ZlrR#A{Qe#4O_i&X63_Or&4w(EX+8QjkyyLt50d=eW+)vj-yosj}fX`SZjvdeJl z>4Q*XE;iFL_drvCXx8uU>V z%b#ZO_ARbC9tI~vmht?(HdFIul!hTr+?mySZbu1D8d(Ij)%F2%#R0unH>{wv>D_mK zH*Q@5&~m(dem$us?os~l+t!56z>eX$odMl_nB+ovs3U#{iVMcHiv??;*GGTZ%tB)* zIX2VdU8*BRfX|zhI$;6%cUUKWoZ1eNj-x~3St}tY;$t{1cRv;>q%v*w=uoq;EbIP! z0&iXU&23ewlPV5%1eGqv zZk!DJ4j%oj^!GQ&TAf+^EW=!M|phXrJBLtFgDIlIO! zA>FcKeZAx*?7sS;il_nhYGm%BgE69MFL#@+3x}v>LQ|jI~D_r z?1X0@aq%}~erdPv^exSh>CjaDaKnm!eV;p@_yA@|*ia*OE;4U-30jALEAAXP@sKwj z-cU~3ICCvepIoT9d-w6WxFy*4fv@I+>4|6tNdA%>sl}O@y6|=>k-|!N{)&)zbj)w?5kS!5{# z#IJ|p*b+7+Wm~FA5%7sguy6J9I%1*O?gXV>Q6%d zpaN&qJ_#qkSqjf=yoWRo@naDxy!Es-s!`{9C`7z>Wcl2l0*jN#^qzbnpKgF@?&an! zzs_Na&!~konHL|DEEd=)D-sKxg2M_6YVTqHEa!Bgv{Fd(-E1G4$*sSG^rPOR2^L~J z-*32H?7U%240?joL!;#HV2er#+k(hH0Nt?DcwmqL{lzXPyo$MfN@0p}dZUqp4aAGK zWS&2!0u|(dgxVKG{b%2*p$`H`*u5Q#u3f3t@`Y^m9)q=;{-Yn1=F0Rx-N8`0hpA*E zy@=Rnj>U`IRBqe^X@x;Eo_EJXBmdHFsgX(Se&o9I&f!)ZFA*$#%+|x+k^|N&bGi3F zmd(0gBq?=~M-J#0#+T|;^Tg$mA$LCiaO^8&2j5iM_=($>3>blX>$T4B$MV!k=N+Lu z@lLWp;z)_j$3y>;^Z)fHh2PEgLF+c&K0OukVV6Cvd2pbR5qo7v#8v`7()-B#jN@;j zE~|IOUqvD38I&0g+1^w6f)G6CLm!*pfLBf~o^bQ{n?~%9M)cj=h1_$60}(?~w|v#1 z=om%QzrqijN8X5b8*y<@z&i6HW3g!!R(&KZJvng_`V2ZBZ0#ias&&}QSUDT>izaImQaL_ ze=E+@e~8g%osPu{^Z3Ka99-e8eEuku@5#F=@>NOdf zlLiZZ+yy##%zl*3<39r}diIQGt#mBWp2!o8v4)J{eOIhk5%a1#Dp_~5Y7RvtarfQ@ zCfei~Y&Ee$t9Q`kvO(iP$`pkZovwEHb0Sovm&aH)EaiEAS?!%i$-c^v5U@n>?65v} zTqzq{XudzLz^_#k;fL2Qgr;)??`~P2#uEEQU)6T@K&!|qn@exF`wdiq)pIKQ6DTB; z^#r@-58U9>TB7itWiK zc>jBM%)j2oBJ9xE#6dZBynVnHF12VSX zsrs^miz7s~)qdGn1@5_^%xpQd>&q#E-aT?QhI+l= zm@CjDRes&Mj5x<5S?6~KsFYAd4y-Ds%pv0RqwhGwqU2B5vOL+ks> zeTxp+^2qamq3o5D-gXqBk7;&~6oud$ioF~2pipKl4et#ocEW0}-m9}Y-0x8YU&ZCV zlaT_>=F^KyT#sXI-s36RD#XIW`8lrhx%CM+RFf?6J8>p-O>{lkIsY?K1oTrkh0aBq zmHxIgs~D)ZFx@b!+yE17gLwINZr=cCXnx5J2Tob=kf2>hn7($4N7Hx2?L z{T}8ku2~?XKT*?HqZ#WY(*#yN3&LXI-6T$BK904Vad%raA1e$j50h)Uk!@sNzAN7r zhh!2T`>bP8NcA&3lGLJb*ie4{pWF@@A@vB!mJ_^Btp-w`^FMKb@g(TMraEwMMKa|lalxF$bcUH-FZ55U9r>b z_MfM|U$Ez=z13cN9#%ec3@lWbjhr`bF3N@C&^1;)>sI;)?2G*?A{|J_V2j$ujxaqO zm7e#mad{-%D(KaIa3&r4TlL&lZNHDji@!Z$U$Dbsl22OrR18p)k6fgAKgDWkV=rGl zCNwWtqiKD$9oPxr%qi1?Nm6{tE!P$BsfwDKWHcY@UeB7PZ8J&8dC28PqXV{Bw@^C3 z_vL=1;<3j%%Z@@ZXN_U@y&jw>-EiUQ+!rv>7H6^Gyf&8T7S?_!yTVYeI93h~_b`ZYR~XZ&%MzD$5PZ%Rwcm=g zwYfYL&|H20UR#wWna{ka$jtBG@zfdO^jl9=dGJ|4d-gt!wV3EPK5^CaP;62$M!TL} zkZDEab?Qlr!rx00c2h!a+39%4nPk4rbFP-CiSoRUG*%&UVC_Hit}u9lp*?u( z+(ai4wf_0}o??%23lSciBt^jgcv9&DN$BY}ULpUC>qkKwdB6u$v8!ulG#f*$@Uh%8dSsyPoH?l^z6a=+*QTi)AoP8=r^#8_ z`cY_AaegV6X$j?`4Q(~%v2Z>-AZq)PZiqSa)@IdD;=D0Q>1Bpp;$*&>D=~puOrTk= zEm6T9z3{B z==|(_TOyT+LsAtEB{ts_q6h|M$H&jUf_YOCL7N3@C^Ju-KVscF3uA*fDvoL`!7{%9 zT|lD0yU!cmUciq1!XFtiv5;t&8Dja2i<6}AG2>QT{5=!K@{WnWc*xU#80l$VI}T>@ z#H*m08k@59?uHP+taxFMM#i6?F26}&G(w=%)}4SY+dhD zYT*Yh_a44fT$f1^;*bA)sxAsThE~rH8s_3CC^OT9Vmt- zi9F_?3+BAMYrT&~t~TXmVCCn_@6Q#mFogeu?!Ja!*l@^yQ(qxJHqWhY^L0`P(f_T_{E$3T=Mnauvr&w^O+xY$m0{aS686&4p1DNFVQ&2<+l7k=9O2iw z&l5+r-(vcM&5}7!YOu1`H|FMZqMvD1dgQ{73wB|R=EGVuyY(0;o8TX!dKL5IM}-O& z68JLu`T8iIz(wf)e0JsD^fgEy|8dXiB{yD<75Tq*ggC@tMV+QtT7D{)pHbQ2)%yo1 zq0Y9aBK{C^Xez&1zt)fyES5JiW$_Rekp0nGp zMilR&%m@$KJ8~4^T;_7YHhvP+9~%2)VgZ<2QlO?6-GnW@1}jNnVo3K>5};StGi?_Lv{pwd7q2b{@hubvUKpnky!K1A4d5yH?9=fSmJU!kU-3 zaYN`>_RB?NDiS9hGsLS5xp94r)etL@n#-*(fu2=23#Bei;$WF_bvkndJ5M;odL666 z((%afySK!#Ok_z+eOf-0-1fg_u!qYpj^jp6;ti9tF!@%vm0=PW=Lfxx!%yDO(vY@G zRCnjf|KMe+U;CaZ{}S6eyZpW7hLB^n_>fB`!3UB3$9ch|T?$3W$~Ym{nQOO0Wzgvi z%j8HXG%pafmmtn@=bx}erDu3>=RiI8_xuZAE}jgV1P(^FnqPuaIP&?S>=tM$<-4Ca zMf5jne(~x1&j})~#U2}Xyq#-=fjdQ%wuXMjD(6R}i?83{Xm*A=@)Nw5JsCUqXz$R& z?8kynf}Rp~K-Yoy@VSHNOH+A5GdSwxVqof7tkU){f?<{0Yj@7M4dsdRcOHxF1x~~0 zNrvE6Xup4b-1*cNp83#bdby}g#^Ipiz0)T937)EvLEmCMsm<^^?!M=3sf{>byLa?b ztrHIL`?x9H6M~+|Nd2C=%h+zJHnNsu4IKLC`s-HL$)v5Jm0taH?BC|L?Pb<67WN7=@?Rhx#A@m=x)0rvr9!HMvJ#A#^2azgI)!*@R2NVLu z`f%+3rvKST^-}_ZHH7}y@%XKv&_b+rz@GvV<-h27Sptq!eK-ByRak&Nn#jrB z(uCA=?;b(DmsR^*9is1HvvuQjME%X(uHA7bw zJ>uqV@L9R`uctxX(0kO>>r*{9zDVJFoaYvB`Z@G|xFTk&;zJRcB^aY2&5Qp{I`h=$ zi(M!*HdS@EgpNXDme$2@@BgFE@e%_C`MlrI&Y!dQsdo{k*r%ez$t5_&G%sAcas||d zS4Y3z_D_5QHg|6&?n* z9eYUvXWw@E>H+_6sgRbxY~M?zMVNp3s|x=G9=t4!{yOG*m9q=!wLKcbxqGqLw?cBn zq7VA5R%Ikwr(s*Nl!CdABNoqA99AlugZa`y-Yyb5vHrD-@)r3yQ1`fUx8Fi;y+8IR zuMgMWvJ(bfGWN6t6=4TGYunO;520R8E$@A3Glk!KZ}+V$PcbauPP|H19SrO8N6XuC z`AQ+}y~pUQDWZSXsJ7-RrOgD7d40=)^{cmz<8ZwCZGY=xXn1q^%7G;AIzq>qYfJPU zRziQtv0l$Rhmq3 zmUaC9#Ei}8)4g$@340P5 zI@=n|n%ERlPuspgo3&7Ex0Tj1$hCujkttU2eE2K0-A@moGKK@((|PhJ*naQ2 z^U{oRp7+re`%Zj2M#Q80;YCADE5X-3c#!IJMV;u&$0}i+{$76uYmW}C^XukahjOn? z@>xr{_!fxzX1Pzq>@`%AUd9rB6h8K`pP>#(P~J7rk^03K3!2j|O7KVH-^o$i>t~}0y&H8nFOM-j{?8?ARf&QmR+->$vf0=-CtyVK-Sy0~RywTwec>lihk-SQ6sH7#yzyCZ1-GfIeD2asr7~hvtGX4pP^EyW( zr+u;*I9B(v{SQxIqwj*h8{A$&mz{_2#lrVMKOQDGhy*^NSho8`U+5TqIqyNh6@8qazUoV0U$ob3kPMziddm0tqgB1CWa-9^kI7po^F zeZkTMzAaWuL&$kIc`t*k$EVb=XM%e;o&RJ$`&#A)QPiczDLmHGZnhGu6-Et}u=syG09DJ|ifJegNmHRINJOB5lHFbs9`u*rg)ua-1A7A%D5tm_i zNVDA{5AHfp_{#$mo;$c>kzwa4OW6lFAYU_ZP00(U9f|( zh>V0v!;_qA|Dm@E)_$nsXS4vFtn+(cYbf@4<_Wtu--FU8clUjodWST-$=9R&M13oN zq|5NHwe39e^lN6lV0hvr&?Xi#3*FP?SYlme!+4<_nK~v(PyK3zAy$&u<%qGUg9e zb~h2fYY`6%{W6>Z%;hU>y*`~KleXope7NH|P+IN%UMXxL^Eb)wJg#AdeKU?P<#d|B zSj?Na_F7LUxe>7GNBClxOz5#bugX2|(9z<&{L~FD-vTry?hc@AAaFx{EjnMq6u5c{ z>vE(YUv=T~3qbb)K7N1sc@(~hd)W$BL>=&naA9Gcj-|kCS6X%Onh4MOwRw2^+*~*l zOK(1_KmEcPnQI%LRjnrYLP`r&XCFJntyiJ&UwkLHbPvG`6|B{@=(l494ql%#UUl*S zvHtfp{HS5vyb$y}9S}}X?u82KX&cVhDvF5LcTKe_CJayv+~3U6!z%l|Du?!S`A)F0 zY4Mc?2YH~mOPsg(y$>4lYt>i1m&blRN6K}+32eXUqVq#Q13FG;2bD`OfgxMlZ#zoh zrl{XoghPu0a32(FFU2(l9yDeQA%dIu7mCf2Hq8_*oT;=EfH<3Wlm1TOVwTJx0tFKm^c(e0#9z?s&n<=}T10|&eFCAr?u>RZ9*cGF$I958; z?>NrOkD9N);^TH@SM0qMT;IXZ!77JcO*H~@VA#~y?!p#A?mJw}TcX4MT_3|L-r@FK z{4_Lw>l}ahsB>CxovGME>0^XJMM0mtFsRN8Fu$tVf5`RfQ*?V12ecSv?ebk(#QY6J%H~uNxEV&*jn-_j{QX6ZO;-9y|lI3H2w=Gkyg%t9#_q&W}RnZp8r? zgCNXYa!UEKIJYkc=Dc>T_%z7HRltD0QKhVb!av{FS9Ri7@sD@VK6LP4ZCDaiL{4_h zlyC&Rg_!JT`lP@evPsgb)!q~MRQ`GUzU&B1 zhSs5X@2|#BK(}z~cbl8fF_Lq+!$Z1?!uK+`sAA;b@-{ZSngJn$SW~EUe60zg&w3n3 z9Nc=PAd!EoS;VvD*nc5%B>i_UcHFpOTykp#^w_%wt+tAzkSt%$eCMM;gvI(#VuqT={{Xj|xl4yF4hQCC2A|$6Urr_b_Wu`B&Ked{SKWDSsN}Hda>w|9F;cBa`b->X7{PcCx zeCUn;=+>Ul42(yP*=8cC*z6|%?dVc61g_TUzQrMUSDTFcT0Nr=VOu@h_`-oX&~$Cr zvNUgJ{28%})Hq%VnMN~r&e>oIMeC&wbYJv9M(}lo8Y3#uFASPw=MKgPlJ%(1U?Ik8c2+eoDawuc6wQRW*GaoUm0r* zkHldOe^P-%FjhWv*Gal61p_WM>;dnY_-nOyD9MMwb2c@$r`@t$i?w%$-ke_G11)u7 z&lX=S!EWudUdMIAq5grS!)yu>pGx`VQWGO&3|VupSVieSL~5BlEAK70P9NJt*0fAG z*F%-_l}!ouJaI}KYTsz1vd|B(@At*di_SPf4N$$Y_$zif76v4TJi&n=%fy>|`Js*0 zBlX7iD-=AqqTe{2gGD#QkMF8-0MGebPkk6{K~~>LSI=)FOx*F%%zgU-U>H={X`fKS zvBy0c+jEJ$c%&IWRr#bhl-*X%^64XRJ5gKM4;F|jLW!pNZl(A{=&3a{v>CF;`nPA+ zTI(N!zn1g=KK+$At#9miW}VkXPT2eN%lQrQT#D@};o05TkoYaT#4`r!v_=7aR5U;Pb6snSqsVCQnM|0Gcj`Ku7I(e1u%F~U*nzHE-ZEnx|#?|P&IZcKswPE zJ>@D{hiZuXc;KbW(o;6-*kq_t)F%G}sU9Pe`MnK5Wjgm}?+(Vav-C%I17-oe;bMku zIv+&*{VuD2P!Wo6H}8EJehsTbjLh;&God5+uOOX7#G#T~-Q8AYY(Z*UUm4$H-nd)+ z(X;2ZjuCOmo=M53#INyCzGm&F@Fzq+b3f^{RH|}0h3`oS`0H;1&a=-phD(TggzR5F z8#jL^;*OIu6+PnaFNOAypS#P=I*Ic!_;s7h%K)HvuaMa1uZERd{~nI;*^P|3l=S0E zWwGdX$l=`s{LpM`xTk&83O$|qW6v*HjwrJ_q4={UMT8w3H(F^8MTd(tw{!$zv#)>e z@JfL0YRk6ESJ$1jtw-g#5-PiZ;*o1iTt5krxHdM>%%?Oj` z^0GmtfXY!yx_yuL-%`2%ftFFbIYORQf#oAvX)`BtQ_-|A=1Pv9S0 zkGQ{0l1T&lGp{1;H}@guiqqK>jS?8hFa9F;?@4HV>L}N8`wr0W8Sm0d{{^h2=UZZk z_yqU)ozAi6*SP0Ex$Cv?vNL8dvbA;jyx*13<(_l>T?{vWgDk#70|LMI;Z)Od%ZV5+ zz62_&L^!kVErQy&ItitwpYZ-0uj~_IS71y*tXWBg&~J?F)i<7TW@2CUaq*|0>?oqT zhn%`PS0Qat%7Lc?ymc+{hN4LxvqgbwCr(-t{bagNWJPaxufMgOR_ z^9tLbJn-zI2^-OO?s4RJdB5vBPM+NvSasn$b`@telzbxASLIyTNa4*ypo9n?Jf=Xz zQ|%4Y;fczM*f`J}wk>9!m@mc;?i%t^`&lIBJ16UVFP$Z>bKhtGr;8dy9<`?7TinwU zNHQ^Tp!qV7sKsv2n|=hE{w76;xjSLamO6N7WDMh*=nsl-qvA_pXqV`SUg%n=&`elQ zBZbiU%EQ;?4L^>J2XVWaQtKViLAUM(!`fR=WA<8N#^@D9AJfP+x<&Xp}S@kJ5g>aGVjGk%Z=pMjAom7jA-UBo))mOJTqkjPsHdvjKv?;&(0dfg(l*-ue*lCYO$ zOfqR@%Fw&TwT7kZ8y#|_@t8|HR!Y?mU$L}DHfAHwhQl;vtVSPi=L9dD`96zXU z*~7jH*Gt56x4xuF-Js?xkn13`<=a<2Y!vY?w+p25T`=qIJq3P7f|s1t@Y~_y+g^C? zr~5cc_+Py1>QHlKE(hw>zBxOcW5U`iwl4#BU4%h()}|}9+y7tQDOQj}^!9krc()ErkPT{&Kv@j&NV!)Ff`MzLl;XH-UVJ5KLO{BZrgIFw%q)bv%H zqJD3#aszCcE`_D#3X!aPolwV8>Q)j()B)x#w#m-@5|(9<`Yr75`9JljV01Y}kQB{> zi;=v+%eg)H<%p|{JPYDvM&r`!l;hZ{kaJ~E{Ws{nYEe3N{47+YK3Vh%qv}6E>vJ{~ z5Bvs2JJpI<0{E+Xx&qjcajfUwUjo-sEu&wRd76kX^G-R0Ul!R86^mK^`@ZfV)=N;cjA#PjxWuD+ z-mF!6kVp=d79sE$P>1w?5FtlDQ5$u|MY6;nE-> z`8XLTcCDqcaGQmwECWGD45Jva($cp}$O<)92=Oya(%o)_>o5x(jD+@(5<; z5IO_3-#Cu)SeC(5S;?ELnbweU_1MGi!U@b)J}5;)X@Wx8@U_+XgsyAOvi6ep`fJ$v z*3`egvjP5W(qOC%oX583mZUmm7W_+LyOt{b6e~6!=H9~h9?C56O~7;#WVHOIR9rHL zmX{(iOR?Lq`)@W||KHU-7;C3L<^ezZaZ31+e{|d{Xp&*Qw>WDE6aAYxKWsjTlq2CO zBx_o%>Q7o2a;U^vO`h64su!`{rDdDNpZ&PR)1u5fz=2akKOPl0O=IKb2o|R~0x#1Y zpd}moGYJObBsRLYs4 zAM`rw|K*o7uA0&m80eRKyLIa;j4k7?|s>>*nJGr zx5=y2IV3=cd{A)F0actS+Gp>e`!lb4_W225kJqh+{vVS_VH%jz1xc)4<25@D zL_SZcXR~YExeaTcU3uMf^ab^PCf=K#XBp_lu4b_U{Y+;dUwXAOY5vE*IOvi*i9rjO zkG!~R*g)|8Cx6B(GH9J5(Ttkhc{4@$v(Gn;Mg84P;NEMs$_K4wi8!wKFiGc6jwsgA zsIr&eO~8PJ``aH+=B`?2a~m&LfF>0miZ!Bt=VtsA|E=RPx5O7iB${)Ymn78I;LNuC zt-(A(SbJjhO4~Z>{b1FBi#|%F2o-NN$O77}7YjRHGkD;10Xz+!+p!1wlP9)2V$sBY5Y+q*IC}ctCg5^7fPJfke zE%rU*DH43R6>^T6S}WJe|M&aU-|hBha5JwW`kG~54zSybtivApqt)YugpTOIW(JCZ z=3}f%;SbDRV-IzwZi{nt`(nHp9aqb^D;6uIAKsMA0^=-mTC;4_I0Q=aLOw+9Ex^gB zW7e~C$QcF zTC@C9M8EPvyb$xBduC?XsX2LfYi>4HFHX?Ot5EBuB>IYq%y+)dkU>83O!8I`R7g># zYZtX(`tQX__%Aw<;_V!DLwNUgd{Mx{oO^~l) zzSr>2!m8&{tnyc9W6&xd2NC~EyngwToe#&E>c5+tb^VJE1I~^bQLyX!fZuqjB6jrh zykBhD0#zBGoOVqY1BJz~pF2DQ>wfs|{mc`I%cqr3X1yiB1gBu(0sH@{rv?$aE(T?q z#5td`Y3Rr1#$0Uq^O8KdlmdM&Ri+ssSE2Irh|T2OYTxVjfw6VgE5_LWj3J_5p})UCVh#hbxzP(A}dEb5k;a6d&#x&y*6gDeDh0uD2p9uuO7F2ya|5KE_^k! zBJ!DE8H&BG2lz;I20OHpo2a~9eDG{UrSKpPq|oZINl2e2{!dSck#lUDEPa37j*fB< z$Sc^m?@8hp$l)6q{Zw8GHS3-CzF>0238_yu-wlcSdGImy4f$b`kSw%Ehu*LY$K?e? zbES(g_1*2(w0o=Zh0~sVXZY4A;VGRb$j$tDY1Hm<*QN3kEPWz^$}erjUt{?tD= zf`_5yX5_P?+thpj$Uo&fim&hi#pT}kO`-Eh@v2}+U?%(lA5N~n_w~+ss4x{WIY4I}h2q|@VTWuIi)1dZ+>vCM7?*_t zz6_XZ5YzJV$i?~d<2ed9b6~=4$Cv9*s$k&x5lPV&YQ90D<5%m=|Lq3fIRtETr>S)X zS^7=W5*Ob;gHEB&V~6+rpKpRZyK~$!mg^fGP7r9pnr*}HrGzd*`m02dqZORkW_IA1 z(yk&HIuU-!B|q-K4+6Gu;>npunrcBVcW^OQY|)rOpdi_D8S z*%~|I;w6IJ_rG;N*AgVoQ#jK zV3Ni1(v9hEoE92%ZWv0(46E%L3dbloV<9?Alk*m8!b4-7IGdqu;Pvva6kZtXJ2onm zU=3saTH0P0vY_vYK7pYc4My4_2a9? zp+altJC&(bybY-PeqfeQ;48L?)7%_n=7(cr908-ub6+_;Ut-#uSB!6DegpY^)8nyQLfGl9k@C&= z-#Xu(7Q7+%F|;yP9tw6Fhk<9?w<@r+5aaShAO-}e{%TNg*FlN#qB)FH5+}C2U$^T1 zOP$Q4h6V_q9m=Bs_HVnFp&vdm3#(G`VK8{#3{I@i!;Y0$sZPoQ3~Jop`P1Pdl7AfC z+OGBz+8KOn*5BcT-Ci!{=wBZ z4nGbM`WK%`MoCeqbt>@v^7K-1yd`vLc?^s0It^9A4QUhmsQqRnzYlP~ejpl0Ih!B0 zu+hPwWNwmx(*;<3{V_Hv+7u@C%A8GSq}HdfWW90Uj?3qfd}M9!#vwV-G|rElzc6Ddlkp-8QDuEZ^B2aZC2-p zh&-USHp6@^*D4+p@xDaE1@#ZBbepm2<44?kmJ5bHTX)b}%wmPP)8m|?y|VPfCIaS6 zPl@NDJsTpkH3285AI5q{X+hh>xr*m8R9*qZ@?KtwOH;ybrgOU6p6bHrsBQ+s`LZsS5l$-`4ifRg zqt23OL(5t1BB{I_sIS)Wznngd-A#vewmor#`so4B(+XdSaXZ4jJ#fnxq6#Y21+02BV3 zJjX(9v6N<4Y~Uo#q)ILZ6#$N_Nm=?E9qSu3e#mnH!^pE=?Q4Y-df!US*n-=9pxxstsfnLm=?7sIWKZM>>j@^Fz`a)#i;=i=n+p1x$5 z+VAT$)R=dmx4Ts3|P>+x$!RVYwoC-#bE+2Ppv_~8%Rd~wCBm;Xw3CH4Em zp3fFvcE5nht9}(uLqwlqB-ZhPlI9m|c(KeoQH?+({Z~>=^ zwR;Jkd^y8==8>-JkZpaXSNX*1c`!wa{nozuf9u%P8+MIc>MQy0*Gcpunta)BPQ&Dz zFe#Q4LVuWXRQhK4&S>ZfJ!}#v91ryY=e?)h`(dbGa!l}6Jfx`^OC1{gw_n}P6Y@W%L!}FE($}{2IP_%kS)O?yiFsS)+1*K@ zn6u6LB#Uk$w)OQguUmLTVvvlfOsL9%j#-)a@~X$Mh1W2c?%N;eZF=JOf#U=&pFt*;3e&mtGBdM}zUec5|fcCnsH6cnHnQ(rsJ}6;QKp#^2mF9&7C! zXk$zNS9abpyL9_8Z?1ZMWH`hLfY-BAYT5->M-VPx+m zH5dThn2VPTaO!B^a(bu&v^UGsggR3DH;~=Tk?tn9nopBTL*Bb!R}QHg2ln1jTiq`m zIx}bKFt|#e9!AB-#9zQFoG|tZ-Q2Tuch$Op@w6c6-J3RG-{AeH?{*RU(3pvbRmrJh zs1I%r2%Xy@OUto2Axnx{Uy`L+^B|-AmchTc+m~+L1a;m*SHR{kx<9r+QIJg1pPF==z@+J#$4%q!a8~xxT==prRH{YRn~PK9Ui4A_-a4_Cz!%K74~PoAl7yO( zby?m&9wKSF;V~t&2dWJl3jHO*|Lv1m+iZDbamJ=gWIx>)WJQKQEd3YH<#g@7!)THR z{Ua78YuyQaz|vLiP>vfvV2Jy&V{>vIbU$joepm5dTzvGRdPN84LU>#u|U zAw<8b`oZ(K(t#`-c>G}|xRt2$#SQ-`ES^b#u}$g}u^C5j6Q(^h_>s^Xp2>;4tJ!fG zyG0MXUGbrV&da%1TWQ!a$Dw+iUa}}o_2jcM`TmRltny^d&%J_(RH0w*-#C$I1BFR7 z?0wiZtMiG0mgp0VTyZ`0%kw?t3me5ec=ZJP&}yh6iy7-!9UOMeS3`B}ih9iPm{s$s zZ9#C~qjegP>9ZJ+Vt)`P>bcpw6>6a}S@h~Y5*1g1%Xw-$##pH91QJj8Z0UPQUFXpB zU^rX$qPQ$=M-#2*tsofrTk|9Dr!}A)`8W?n zw+!Z+am=Y;ttpO6iD<8W%+OQa-gx9Z7isxm_&POP$u7w;L|Lg=Bt!2PlSrA zHs`N>qLF1{>2UNLCG-hW1)d$?u5ZWE^7R|tRtUZ5oPs|!sICL!YYh5#nXU3g2J*LS z?8!H}A*pr25x)nHpRd1g7ADSxYWw~g`geWC zhm(CgVNkT-gae^>htmqphon&nMy5kE1lW(^m_yg18_5oaHrW3RCz)Y)0lj)}_J8;^ z(!RB=nILp2J7*4EJ%pOj!J@YF%A+$dKfU|To6i<7e%tu7{BvU%uKYDUpCJHahmMOs z;uivMYn{!m!lkmbf?f}T{^c8!yKcJKNUw$NNE(ZCcQ?RH(%bRT%>uA=>VU&8?!B@! z7iUFJa>YZ~F3_-lN%&^WIoz8!y>=MO_qD&;AGX@JbLdi}p9=>U);s?xtrEV2LskC1 zdp>^sU+a{JyO+fpQzmF3;nwj{=H1kJjqSDMk8f39;-D68U+Y7}gX9#8M9GLU=ze37 zytI>wKLsa_z!`aYMjTh1JEK}kjaxBK_nV}{7Io-0Xn4Ba_#~9xwU28(dj#jZJ*cLeiGtWJ_M4S|Gerh0U;|*#(9J^DwAdRmSYeHV1HNJZT6Yjdy z>Q3mw@U89vhS3Hnt`o2w+4TrJ7U&xN(C{$0*Z@TirxKU(&Y zp@BiZb(5i--*Eg#%I-f|XE5I1@rdlv-7xv4Q(F_kpBoQt zXIwLjp6rD95W&Q6kEys#SXK^;clsg%l!O%s`n zyu}9U&#;_i{+5b4XE>%WcP8V+V>eF0t#de=#^OI~RF6&ToHe>7igCa@=2x$*8fGir z1CkgU&hKp+ee5J6OE=Kpv&Qr}j=QLfhNly}hg$Y4dTy6j>0KbD5cvvQZou4aYp=9` z6j}x52w`lN(px-+;ydqgjVQGEQ?B&F7C2c#TCC zN+heW26hD8+$3qaS}&!XV?sBA>jk8f_Ogs+jY9Z!uJ?hR#Ch-U&EdJNgV=9fAKsHW zG^)Sqddb&nhG$oZeR{;1$#lMf(Dk15pm@HKyo~+3#y zQ@cLLuh;0r&ie}MxJ3=1{Iyd?^ff9!5h7c4UkZ!A+WmD(eUK7i zHP9K;k4>qyLKC~5LO`QPm)w(49KVrlZB}+0x+2%bGAX5E+^a=^i(XI>!&_Bwp#(;k zUOL(|J;9-k2?w_jxK~`-tnI%ht{SI4n7n&GMUC%(3?DnRbEy8EFcrZ#T=H;L{ThEJ ze(BSmSb$t3W7f6{M0`D!Ep@SFM+{EhWgpR8Z;AEkeWHP^Sy-ZERcO(q17xXAX1a=S z?A)l~G80Acu;yG}hR`Vz=Woy2{j>e9l>h#}vJ5*KO4FMppim}CwD1qL?*UzcbMG~7 z5jqM>cXk&^d!EAaDb??P9D8A&#*)#$p5OtsJ+mcUseT9LFFRE=oiD)H9#z>=`~xLg zb}ffxrDa*20xXWtq`|P6k-jMR5b=MmPVV&Za)eQ@y55sJTXBht<@i|!FYF@qDq1&@ zV921WGLqF9s;UznJIfOKjh!>6lv&5 z#u(1DmfQ=yZUhV59u)~CiP)X;g7J+P0ze{&dngQ`a(BUbxf_p+aTZAa8yAaAd&`W9m^ z___0~<1-B}Tx4z=N}REVx`WDj0|M0bgYEClr#VgfAZQnw_=HjOYpjkB6F#;n3;HiG z-JVOF_yU>M?=`!_E%$1WCI>^^|$ zO_}FNg2C#U*WP6+j2M+P~`MXvTIt3l$B2`C8 z4a9gU57|`S?#5Y^Xc|%Eua(l6XY|M~FV+#8PI$R=(@dcJ(@~+qbV! zNEdKJt%@4&Lg)33`-MvAu=j+{pLgQKIJ5?doervw!7iDfjO}ZvI04A~)Unwp!V~)Q zrCzaa$bs(IJ$#v}1V6r?pYrvAeiM#QcpTcW-~g4*TQol^T!-Amlg2JWIykmDUEf~J z0E@a9@5{10T6G__iE$Q_dH>={xiEU2b?2~?y}ej|pB}WF?HG5~9D)uD2}Yr_1YfqI z@kJ-|0D*^}(AP4dKJA6{^Yzo^zNKK{+^V3>r&l_ zN$ENNV_{d z1WN|!nj-fRdLnHfQX4Dpe1}e(=FD09bex(pZQC0C2XcGQ{a&73faXWuW(E^!FvLL1 z`rCs#4@fi%8CUxn@}WL*p8r=aJ25_gl|$FwB||&A<=gzPi4d*uc=k_DpkxcETGs)8ctVvVXcRZyOUTlEW3W@QLJNuF69^8 zmzjxtt=9gP{JxbKsEZ37bF>(N&i3HV%eNzNs%AZ>(V1oZ)68hS;miOuzuTGpbRU+pw106MWPmPs{Z``fS)fF7EU3{?_fLowq>QpT zo4}-{!cF?ZMrgOn<(6fk)(fykWjJ$+j)>c9w3I%Zh@Zp6JvZNjuOxImTTpv2J&Mzs z=ZfI#rB(B_`Fj-0V$8qiCaA=_bf4fAEbXx@ZLIqWQzH`d1tJ8GYgT>#xg+ijnAOFg zm^HJE#j$1U`i(A;Xpg9>@+2o<$m{TsAXh@ix^;i%Qb~amj_Sn(+xAlNpCnq2)a+yP zS=jCum?Pl826M6nO!s97U%if~*aqCIJjTe zr%j$$`e0A?NZk*9BA%c9ZIjYLpA8kfF-Q#PSB$8?va?aK8bN4$+l)<7Zw{T3`>bc zVFM@IOq`-K{E7BUaCsgKwG4fXqTGZJcHx~zzB_IILr=3K!eY|Z1Y%jL3&uP@VDhHg zC=>Z<7%^$f2;#DbKf}%2N;A*lNNW(gB{zXjT>kr{dgqC^IDV3R%5Nd}-?`M;;oQ?QDAw)0z*@|IOWwLu?;1xo!4gL*7aX)21$@fL+^_G*+CP?~oh$EDn7^dwFDh z7-0DF(@*z>-C;10O)|*nI1Z^eU6T{ngACb9Yq1Nfa7;LI$1ouH2nUX)WZP!lvc%Ti zDjNn1RDe=>W>5O!I$Tt)h@3v-k4xGW>hBmUq2u?#T6&4=Bo-UL=kJBov4i8pH>i9E z;~Vt?Pr?Z(c+%3OM@#rWjaQj`JK#>>peHir3=17BVcy~-z1?L(4`$?x=YET8W0+^P zux|bF92}-i&~u1X!oK4BVz#rQI6OPOm+n7!pbb|&C%zqVhL*^5`Jbl#xN>xkZQ&}_ z2N64EHWjv{8bj}$b<^>U|$ss4MJv`Wn6b+@xHe+WPY0hJqef<`?(_7Vh2|flESIao7 z#{Y!K7l)4=jON6NqSK~bSy!mHo66D{B2co&kVROjDI#bM`i^bMLaf@eJ^ z`#pqBM-^-DOC1%k2!#2IC_?K^#K-lvQhgTeRGbzxrKkjZbrN{AE~)IyC!dmV$j#}w z+T%>jSs1&OvCrh+^Q3)um*Fl>TAK2c*VALyiKV*xLezQ%i9Yt++n>?JK+fH}W6_EX zjjHj=qO)vR9vXAO`R)(+*nE?ViIebGW?~o?El`Ssd8_=jYs}W-SXW-~r=K^mPf{`2 z>@yX=fs>}ajO~XGz(O16qw*PM9R5>Q)Nd*Y)w4H!q9ylW+sKYfXG@7X-lSN1SM2_c zFumb{;O;CXT-kFdc9y3NW~4?H4tFqNciHg$+SdyI^6+xU&1K^_WAVw)m~9Vqa7s)6 z+u5*+HDKtWs1(+*ObR_+V1r44#5I+CRG&teb{G1p@=go7^Udf zpuhXD@pSUntR)4U`BNV|Eini+`7fi^IK~2{fNNL&T`EsUmTpvFZ@rNV^yQr5i*7jw zL%ZAj^J%`|UtN18%ZPRusvO|hu`>*NQ}68#btH82N1C3zb2=46qFHD7tuJ6TetMMR z*W%8Sj~!h}f>Lej*upu_bz8U=2fiL1sqf$=u0J5{_`09eemZv4X4WcPmd8HrB1 z%{X32i%-pn`AI&ls(F6g6dIo1DD@Lzg*fMq1KW)V-|RB+qap`r)nRP=)lY1V1=#$LSY$Kf`9k+Pr6SJS)QZ^(Z&VOW+5# z_{2nAZ6o%>o*p}iDt&W^d!EVv{yM=^U-bI4mEGCZ4n z;*?&v$cAGY(BH@xSfDI|)%+q~Ba)uO?{9O@WGja8&ymC7MH9L(@P~n~ve*K0V$(yH z-(7(2C%&^y6NFxD+lkY^JboO4qP%{s<>YmklPl`^@+{%!PWO=WsQ_aKP9$zU`%Z8k zhwd5A9~6@amWbT&d-vcY$EW$y}ECw%)jrIWy+7!jUf48w#nk3 z5>YC@4>OVs#BalafAQQs@iX0q*lG~*b>2~o$cJCN3-cP8hB3#Z2i}bmdIMQ}5q#&> z?U#9o8pwF?*-IPjww+ zhp~5>PkvmlfQgZSCj;gYI9+wC_P7;+cj!yseXKETEq0Vn8BFZB4fV6h=SGF7{+-ZS z)-|(mn$TnIOcg6W5l`Sw2ZL^y{kcSe7VkYJCPNm`Yr@~>G)(Asr%LSL4RbsT_2SEO z@FDv$`FX&115%Cb^L4E(r5uy$pw`aaV_je2ats3TEy0{)1)eQ7y zro{gH_o$yQh?0NgP4s7>sdFaj<@4tM{v6D1D%lm=cMW{MHHds)oPy!tu4`ZR{lSje z{I6n6)OsY0w>+-8c4rkX6SE|FH>mzt-6tKg(&vn$C3GszZ+o3O|AhvI98bRqihYmO z57aj8m-oe~U4QDooKS(W^{)li8ZBahv6#mJ(;S>$z7cW3r4S2zybc`I-HkKJ^S19emf0(%ZLX|ar`WA6#!D{jR?FjKqMKZkV?s*MjC zFmNBh0lh>qt7#&iNNaH0&*4gk9rMq=jZ|{LsF3brs4b!U)|_`NrzW!(W}e+uFLLO@ zcG(>X%e}lLx|)8wN8Nv6`m9h~RJ1nC>fcx&(NuutN3Yy0z3~hvr%x~zy&i@E>F`DJ znHZ=@WchI4s2TgKzCIDXfLIW|DfJP>07`uQCgskuK$-@J5d)tgjBu8JIF(lh9XZK| z&Y!%oYP@>}Jfs(nRbz#=|2=2fPdIaTfpcqyH;(6>*CF#I;6l@xbs_---oKMuT2*+O z$V-Z{X6+0djnTVWRy4gDXYkJehR`oGlunZbPPa`%1% z$XjpbM;EaRt2CADuEo8@iiQ+&s~UBGhuo3q99P~I=&%kV$BGktUCN+|QmF51>=nN4 z=207n@%HXLd&-D9c%cOUZK20~(37UHwoHBo;uR;%>#c2}_p9dc!RbC2e|%oAUML+x z!>_KJ>uZ7Ldgte?0xmFYTYW(>wHV^Lf5yCc7z`sSbDQ$}+>wk^8(6Imz_R>hUc+iV zSsGqzTQg?DKWDs(&0_I=33LUCTc-A}z{*gS?>b!$=&x4xog1yfso(4uGaq*2Yzx

    tA9|CF0CZe9b!hJQW$B@jwOG6#olyoCme8(ZgtRlWxQ@}q}GT(=Ou zu=6@3%$0mdqBlEue>+7G=2hob?AeNlaTYnwJ4~}m=LshyHEPdGK7f_vT9pX_qE9k^ zJ31)Sng_=&yG2a<6MYxbF2l68hDjVaqr!WBlOc?~zw%e~RXr3~+XvN^7yipmNS3C&;iS~FEhHKex1Zpm3k)XspDX%J^u4>J zOlQioK4D{F;x78JiB;n;oz!yX<{b@a$<{M@z4bmM9b2g6lp}l%y3_3G%oP_QTD5v^ z;aU}DSEM{Safu1Y+Gn2kbYB2UQnFoX0MDxH&-rYt>z4K({+l-vkMc_|K%89S%LE|e z$@HHzbXK~P_&ZL_^OX)&KMsd4ao)eBtp;5$ZR&?kd!wS~z;mniRcL3g#^@JPaV6#KFAQ_}%m3FuvuvkPII^ z@jT3XcF19luJHG%n@7AE5$6oXnCPelIA9M&i|?wD^}q8)jcZWJ1zYd@6%MyIh5j9SWs3a< zSXG%Hb)4ohkdp2yIw0w9;GL6=Kb~& zHD0b8kdmCL%)CtKOboNCMQ3*r@xk~|=%lJ!4K(mPi4hL(grUMNos13jF!t?{pa{(X zE?*N`b4QfGxA%^mu%L69!?|g`h~p_wa5(Pon?pYJ(BL4s_gPFmHXf%-F68MW(b|~& z?3*WW1!E^pzd2>|1MAo&j9$H+#0uv4mnTcTNi@8nM{4yO;H(yMmyz09-24VioY7K8~ zpj%$`_JM>-=p&VA_wmzWowW$dL=;iq9^C7z!)Qq8#g8{EGdp!<K~!CsrN{mKW7Wtol#yjA<62h*H94_#&3u298Wrd=(6o-aF&?iSSR&Xu9T5TB?TacdZ&m54^{r>(Wd6-_T;$Gljvqh7&*@ z8r-8TPR4TOkgISu`@h!*)<*wYegq9a`zO`TTpw}mKN7RfSkTNIcB-YYnuhNU;E0U}9*svvBy0yMW78Y;RYwsVs zg;SeEu7_1q&j$#iZyLI~dle6Z`u(|T7l!)OhcLcfQmG3FpSM|aspmE2j#yz)qGqJ> z6<4loYUg!ci>1dalvLbyQ-8m*T$}Y?*9#KkeqI@dvlJNn&0g-(8v@l2OExhL6ZO7~ z9cGy}dpF^VO6`TXm+eS?dqC%0!|HluuMzLQ-)V#|OvmW?sIs?<*j@fJHeGoh>RsqW z?;j+5JF5svD%lUUz?$BU&pqd z`2ff{{!z)!ox4t=y_nJ4A>9lU+mD8=+ezTMC~F_Ck+RBy1zM9&8L?-eKkVph zu7>XzcUqR?8)Xx$&~T1Fh~5pE)~6V@*;4xyFdd`SH;sgj=tNTcKs1vNv^y>DQa{#% z-NI3eT~&*$SLS=25yf(`1D!gT1r{f4%`6DC^@=Ao|@lUqtN5ocs>Bhz)S5yAGf zx7IOo4iG@H_xvMq7FPVloiDtl_GKaGOS^;Nra73XS=atmOALB4ZWw(3EsO1-@ncan z?tjhCRnH~R>EakiR6dlEj*0x$rs^HyD4(frcGoKyRq>Ww_nGiP8i-o{qsMj`8l>qY z(X<%51FM|fqgUaZR>whiY?U9lSYzu+!>xuIRK5hXPaP=Osjde3TMkWEB)!1}e{HFs zyR$HBjkUbkTEaKGYiM0bL}V)z+wcFzoJjam)`_t_vT@45p}*3$on5Ce;UvBFiak}I z5AwotyKu<`TDuRwtJrh~rtVHRzb(Fxqm<+$*SxB5c%=A+h0GFqJ8F9ecCPXVhbEh) z(-{&}UIPA$WuvQI2z^e{87T$0ah$Zqy>m7!LL16V6ByJ@%W=F|SHp(#3_H%9-@kX8 z9IgZ#Yz*DB3rnRB>F@I=?yG0`f?Dl?ozU3kX3Tq%$|HjgL6t|p^Vy;7;||)D=Tu#6 zOf&z;s;F)FKgYSJb5J48j|D3~sA_2aJFmO`{<+bc@CR%SCzgIH68@s(yJx;PTqAri z2hN83?!2`Y2Oc=jK6Mhr?y3Op5Y8Fw)%jthluh`Zb=Ln7O6IP@S-YJntZRt8t5-Jm za)I2x`B0a$bN5a{uemws)?(uD$5rFn;(YxyOJNh_1q*-RaHsN@v9olKvkU7K)*hr; zA99`0i!Hr(KDvAS6b@~eTBo?Mh|SqhSbdENyO$2BxO>`Qy-8}3`Sbtq-E*b2+$(bh znjiMiMC_P=A)AB!Lt8HarS*MZU|1+Nve8W6uqE_&T68K!&Lm|)yU=ib?6zN+B~`oM z;v=E=o=0==iZ_Q2R0_VhXRX~2^-?l(5!Z}Jw0d7H-dx-Qsm;@3M)sOWmcR7tnT-cb zdxcyvbi9Yz=KfVmJOut`;voIyAa_%!p@a)%kBUN3``W?z5GvjUr!U26ylODQJU>yS zKUDy+l?R_pG!i)JGJdCt$7buG{Gx)*l*nOlUg^$hSWDEu1~PeD_IUow8>+_D7=sA^ ztq$>fnSPzbyzBg)?|7l%FxGaS->)4x2V^z_PlI-QoXXOy`nVw;^~aUej^(gJ39Fln zmiixAhUxqA+xRP>=aDk2)OPB);iP3r*zSl}T+DLxUrCO}iII-pN{u<3;j4C*?%9ML z70G%MiVaY`r~OHm0#Og^eXHZen^E(@pazvGQQ9dq5k)d?^Dv6OSE;Wjy2enqgj4+&L1fE zpBDS5r(xUGaG*;Pht6HQn`f_XgpA8Kf?pnY$BwR6zQ6%3EEx_Ils?u1ITt==yqco= zJ(3vCRnS`UXFw}MyrucS$3tI_ zf9LsqtMHguWm(U%mznSxri7iYBR`A948xbp0{z`MeecpSR?#+0GTrYaRQzw>Ddwx$ zqaBaSiz+xhz(I+#u**ancXIX}CqMzw!$b z1f55}y)?nhggeoX7Lp({>srM|*-y}!&h=a2ln{))IOAA<{1r|%w6*)6X2KrYuMq=v z&yb?*ep7MVXGmQB`QhEnELMS%t9Mcvv}7_FdJ3q*wD1r(vaMr>=xi`4HG4f#2lB95Hk0IM=N&&tSNI+~K(LYTfF|EPnQ9G(Fhz z#xd>FHEST}Wj#FA-SzLBqpcYXB6P=_HOSH*?N;fFQs3`7o!l7p@fg&uNr@KR_Ah_1 zAh*id?SpPZ-L-EX?S!5WN8eX}P=}I59))@?0++tHlC8;ej2(K9Fl(wY)^|+T<4!Wd-n(}~JIgo0D5+u2e}F91dRx zS)X4y$i7ZRa;cz`a*sPq`D~#1o_O`DE41RvA#Uf3CSeZ-tPFnA=?bnt$ z?PF}%$MZH=6Q^-}>xLwUxRGB;6vnLjf@9xSa_X*7U|%~jr8^4`#{c>A&BTZrnS(TQ^wZA zoQ_qg`m7(csd^W0KSAIG%iK12Ed^zh7?O=AzxT>PPgUc`_|-bK@Tu);tMz*d%&g58 zp1;G1P3EU%kEqDNK$T2C&FKcn9%Yq%DOCtnp{9|~8wR0r*Nsp}G2276DYzj=Y>$J3=txax8C(77{JUN5mx`1HdEbkuQ>rM2-t zbaONXYe*jonErS{SN}_iK)Z`ry>;HdZ)ZN__v~`Kok8GNx^K4S#oRfDA8)SdwBxaW z#R^Uu{XYfJ**&&6l=ube`1acb%iqJQ8}F2_Rcwcu-kxWllUZRvZ^PE4o@s16tVFk% zkVvBc9;-cCLe$^LqSBF)pA$&*Xa0(cj?zJsfjGtN%LrB|E+4uzABBCh2e!RdehKYc z%%7WbZGxh;l=r*WapBaxIqxscQ!q%XbF(+KhW_0&$8q=w_Px6Ac=-6REbXn9q!+9r zNKV$ipJMPYGc$SL{aTk7JGyh{4rW}(p5>9IOOD5(fU_g~seBGjYs4Hi-b(b9+IAh9 zKDD3N*XCxnQ(C@GkZ3A0INu~I!h-(RT(ZSC?9c9ERP2u>(K-t8Y-j6&=`F>7-*u6o zFX@bv?d4ZE6>XT7@JkvCdi+crf?cqv;cZqy{v%krdzb6Tqw^%1o!22C%N08w989X% z-G*INy(Z5&3H<2a)0HY@y%DJ64S6wMAPzXVi|yv{Z7j{bAn@UV6)sLxZ2n`>3(MvK zX-lUjbC|PZ@a@6M9e+SEFQq znMng>ke9vn=wleqRbg3|mko0*$1R^T6Mi%0&Yq#!-UMHM_Ie1M7U;v7v&uG4a0qLd zc6H^uuP{H^DvrtMd)lQ!9> zhM#%Npm%WIb4#5JlCN>?OyXUE>@`oe!wqJvd&qgr`6Rb2jS|n)kEb4(e$32vxbr16 z-5HLpD-*`9jGIR<-X~+F3{4ByQ-AE#U-Qf79ijIW=dpG=(pmz4Nv z&-v&>PP1739?fd(d?Dn%<|I|mlf)=dSv6a}SE%OXH%TNR3e40`!Uo_gZgAf$~a2;k%BOIJs_zmVw+goKn}z zoP6zpWIq2s?xGnIjco-irs?5wnCiY~H~uHT7)Ly=nvRsM!n;GytjWjpTyL1l%m{yN zv!9sH-%D$K-n8TBXSMFfxl}#}4v^{j)JEKlcDg|wN1Nc0XysWiyyLDbJ~H}ukL`t09C)ZyEkt7E!osn-upr+>8FH=01Q);u%cCp+SE zme(2gN?Jf~$jdh)U76VJai!+;x2Kqso=F$#eFM9=ej4B8{`dOU<~7sgPV9c=vdnf% z5xyoVFOb(=!;w1w<>(|Wp!8_YbKMz-(IKCfZJOOMb^Pb$nu{UO|MXO5?$Aplhk7J` zUUNs5+0c{wm^v*EjXt8!S2+yv_r9<^$z{bRxMAI@n+&~cS6J@MbV0pL^!Ws* zO}WMgYw=qSNSu6k3dUUUjh0ostkE zbSAMMeN1}>yR2{f#a#EmBzsMglQ|zywwGLWS>K6mn;%^37nHz|ceIb99>!z+f$8%x z?o$6ff3UMM-t5uW|Ink?zkTc4AtQ{aVcPz@pXlF{^WW1B9RG&hR-X)|7w(c6I6s{T z%lC&_?!QqZ(FBiVoK-B#ltC0a`L8~6XCQp3x-zRylduw+*2nP+d|85$O%1R1R}lP% z3C8DoQo=-Ee!9ITE|Av7-nkl$Lt?ez@dc-s}Cs{>L}$n+YCP(|U}h zO^XQ+4HKax&h*zlqgU7`nein(il{5k)n7V#{|6`j+w6S!--wcEx8L_y`*{PG#!bF1 zm(s)dU)LtdIzyP0EX>SLTFn!j=du2NsH_|2AG>{cI8lRS7N_Vx-rBJ8z&o(_XDMXI z)`y*?Z6MK}xBI*GqX0I|?k*9to`KSlaVFguqE9^as!k}TmLEo*<&3o3EW+3x_ej;# zzi{sHDa+A)2Z7u_Ya;Za6Pl`r?(Kcg1WTiV>-z8^I88P_`E~F;R5cXc-QpYoeI3_4 zl)Bq6D)vV)>9QG#mW}L7BTj+dri~lQ`U##ssUYF)_jOd97N%WQQ=qI}gu;lai>K+S zI1Cca)Vb%|{MazJjZ;pAqKeaJ?Z%{xrqFZ$p!9;iEHv&*5;><>3f&ULKbLN8#16q= ztsISHn7m8x8F8-$@vYeK@iJ-Xz0Eenpz$2XpHA(D&LaF>mReS8p9bw)-|f1#j@pO8 z8S=_PQOxA3ajud#rj-f)mmjt!UEH8;56O?CE+0JLx@z9LwaFdmY9sWdx@bRfeR3oE zM88dT_E?txATj*Z3`>r_110*hX|m;Gxb#|)=M_H*#Rx(YSrQAMt_MAIs?y zH(S1%f$YRaMzZ~J7#9lZRgip!ohRDPaH|lyk!32KqSYZ>IKa={DyPu{F7)mS^MW@q zT{qKqU@RZ9f8?4lytKhKxoXxb#ULDiy~5X4oR2HlwR+~3DA2|3eWmmDAoj=!z8j56 z$2M=J99x(3B!;5RE=~KjaP(H)&kThgr08dVJNFiGcBcUMlkQ{~J-o#5=P|Wzi^D(M zp1j_D6zkG=^zDXOG%r?bX5IA*$LqgKJY)HW&AVHRWzq;<@sM*|0aM5ziT?Pa<&&IA zSvva(&ugsbu`5}iMQFbiR9BSk+g1^TJ>9BTe^wCtR}~}erEUE0ac<7-UYgLoRnLQx z!1nm!AtFzrJiTQsecuE+m>%1oagfH&1ef97`4~tP%=)5G*^8cESkw57rEwxZfD%U< zBr&NCr12df^pSIh5>mW7<4JT&s-o5r*I?|cAxHkgGB#*mKmX=y8!n%^HFIr~X0yXZSKdsyx<=$piVgf@0BhmzOo z*xSflas2}~)TZ6y+tJa0oy}KX9bweQaatQwW6?;c&vIb!7Jjqpeuf(x4#mAVi+wMQ zUH-@x68YQbAE!M_DA>bx>a~318(7v$YO`JLfbON-`eSKQ=+J%OsYc*MS;ip9;tguO9(sBU9O}h*fztWl_L^b>4^ppF8E`W7-oNLfLQy|& zw(2^{js&|rUz333HCa8sFA%;4O)ljR_tP)KvzY;#AVFv7cS@GrV0i_{`vu!yXCKGR zV>MIH)aa4v(Vw>>hh&zBh<5Md;T~@ zt%G8%YMj+xnpjzyYM(>0e!3*qpS7&JscwrfKvibHj-rIKUfuaHxG8 zvIeh8X6@VyGZ9<-qv-dL81*~mF0T0(|2AIp_if8sb)U6&q>45&?0}#Ad1hH5Q;@~a zcYCoQ4)g8>P4uemgCUkJhi+JF%hJ658XEtN9Y-EV25r<##kOH9-+O#)_$%)0JC-}` zKz0z_vdwZY#FYtdJAeBK_Rd@pZ+a^#OE1)saQ7K4gjj5QJfJ|tr;R&)mG~aeUUfYm z@;SX)AtigW&4}&d3L9>x;-Ev-?;Yu! z(7!!@H^WaB%t$Pt&ArBtou|ETUoh(O{o#(=gII(j6}nQKkkH7#J=K|r?Yrf z|F3-jN6w!q=F`rBj-Bg2ow_>!Pm9ymL6rkA*>-@^B=x#Ag_M zqZT`VQXEE(m^NCSNs(n_A=&m@vO{yGcNy#2er$ih!I$Q!2t7DxFu+gE+i;Ra)UzXn z&`p~;)8#rK`WCBZ_(vXm>&0w(Q;W!Zr?K7lVji6j(Pv$XnXLH_9T@D?#VNbBn=xs7 z_aeu=P+8_zS2Zp265(lnQiq6+6b^7pysKvn$DytL9q+vf-bJ1DKaP( z7CKanqOzavBI2OklcSrpEql)rCV@% z(MF=dm+I3_Vtv&ZWGLu@%l>TPZ^h(s_RjWYo?V8rYbd@S%Dx#xdf?d+tB-^)apQ@7 z$)`RO`Q%WcZKzcXfhV1M?jFxsH;cg+2BL0{8pB&xC-(?@XUy#NZ*}dLz!A|W@;#kY zeIuMp-gN(!&`wC$WhYmA;So&yG*Hnz>nckhPTIo7PYb;XpIKX!QgBLio3=smI~aXo zddE|{2djtQ265;zW4_G1e3!^U=QL|(`%pb@~OJn6a!GjhR|K1{QqK`ty8MvSQd34axBR$t<9!CqYxevX|$D4&7H`0bCg*+B?dh&}SA4)KkfC%OaqsEu5m>a2%&vy)N6jzQ%vE zVwqq29LxyZzg4=n6BGF-o`{;7VVLmeBPY)vgOO#Otbio~r#9L++Um#X4MR%iBg=2v zaZssN(NN~js{MOm(CWv17GWIg`|8-QtPgeXepcz8R>5w&#{Bgw0@$#-oQsiD1B>7L z=qA7p!gQZ_T0xwZ+IC7$@fpZmKh}WolF`RmLwh= zJ8^y5XYoG7yNYh$qD{r|O^r>xn^mxLJp0D6qXce_;*xrU2{ zCw;ycX2fsIe5Y&&O1D3|$)VxkN8;^MKUbc%AavL9L;s$B~` z1F>`ahft?;-VnRayyc2L3CoVAB>J@sUReD1Ez7-kk zFk@&JvwX{^uT5e!i}SGitOl(*OyZ8uuC020eFB20`-L8Q`CGS~J4x^ozAEtE{nBPG zOEXSilXPRXE?fK5X(hD=YMm3x=LWd_LaDe_5}jj~fsc$bOod$>e9AKmKet}n{oR4k zJ@1FZe1}Dnu(*B9G|M0is>WhHr(;{8c6hm(>s2LEcpsY-z3WBF;@#r8-W@Q#IKdi! z*#U=H+Dbn-)5y}$P2CyO+YQA+k=0juPRi2g9&zIMx9<9@o-OY=PWT)YmR5|W(!R$5 zhu^7AxBf!sv&Sw!eZ_Hb=F5XeVstROL{o2kSQQ2;Xj=+D`~dQn=fUSPi8y#*-0}XC zZ~I6z9>r`&*58Hh&JV}0L^?qKhU`Zd-Ggy*@9f6_#&s~V(_!c>a{!h%d?DRkai9*F31}7aL6$xV9frg26qyR)as6 zhv{yhH6U`)i*eN%i7s-%efAGF-0(_TTvA%dsb zo}H4aRNqA;6-);Q9F8Z>ndWnMBl=rO^dcFr1lAGp#q^t)w>}Nd5E9BUx+OIU>a-+^ zJ}IO_8rVNSIvPh}j9~D7nWhT8Dm{IdzEk!3U`%bySA{+aO5aMp7rjCFhmh?Cp62F? zuDTzW!x?q7fz&?izjNSKUgt9^Ul4NG6b?ShZ-GvS$%?4oJE6oo@u7R`D84IdG5IBM z5|S#(ub=A<0{P)fI;+MVICs87(09xN`!qKf@aY1H#kS^+e$l^pu36B=@`*5ZBJD>&}t(1nlRl2t61vB1^L~z{PB@Ee?*Re>^kO0>$SeZspFe@&^a z-OhRWNuS0#o;6cz`Ka@iM3eU>HQt;B`|Y(oj_qfGb}bDp)MCgM-KDOXUrBX)f zC3&oOo?<-vvYtfa_W4+ycEGCpj#D!JvF_|P7%lNTC)IKc$opSkju9a8jE{G}vU~k{ zLF7Sm5t+}Q?1q_LC)*?CELQzqt?Yc;?@f)+?vPy>-V=^RkGbB_W#;|&=cxDRr|@v# zh~7N5>j{$Mg4l2&O^F?4nMgDz{WM(@2pyNMPs`(N@2UF(G`g7VfArA0cXoclVF%3}Ccj%(jZ44cz1cDzKNzxWxY!%u zj``eYly?|U;OM@61H+r`p!?@r%Y_ThFeyLn$Jjo<>ih9sXP)$Ai9t_Q-lNTtrBJ?q zgSUG(k(XwYU1HR22^S zoabD6t(kD{o~E2zkq3kdZ%7S~W`u&zPLVUeIk7+IhjPntLMJTitxqjW@ojAA^nBx@ zehj+Y8PtFGDnpY*>4|jBO6=}zw(Iejg^o~_Uo`Lmv;GL)C^--d9gJ>8wpX@8&#%-Y zPmYpt_I}o;KmYPP%A$BFiwqNl${gNdajhSRiXErK=?=o^aTfDJwlEmYdF7}|8;&zo zXWsO`A?kO{8ax|D%R8}J(!|7>Ee6`2K0UePwJk0>9QMdfy9M=0F78>aPFP&cDqlnJ z`v3bcX|1Hbd4kv)Tp3n;f3BRsIacX!th3op=vVcW3ed$b{s*tTuYdBy8Xu_Nkk)-6 zZFT)@!oT6>_I78O-J_TzbVd<-nR^b2@*83U`N77u^xLtsAZiv(3SjD1@nr*vX=p2X zmY4VO159T{xdiPe^a^M`4}aD;LG&H^_15lM$Y;cA!I-t{ga>d~SYo*DT)z{SEV^q*q8ki60B{MGsoFw%Hyr+tna zRPN~4I(g$L4BVR3uL6Px+*>Ao^y-Ei|2Mx%humy%Qusj7E;}J*nd%u{&c*x%r;kh&{WaNDF@T55mrUfOjWi>&Ah1QM?wT-x7RVY%~ zWGhRnZTRG2d?d7O+3{T7pcmV0Z7RgCx#HsI#lDLJ&tdW@qo&yoQ4-_D^u21K9WZuN z#$b)iZ_MYq-S0d7p19t%>$lD9ii8-Sw@b=h?nudho4QVs3x_U+YcHHqhGPBDo#y-2 zU}G2UDFKmZ(8(e(y-T0)iEDxMM?ZeBL%V%CsP6cVD+iC~Ts|ZN1tN^s-mQ2*g~pFg z=4vMxl3_QCT>BkIlP=z;5!1w8<)=ydNBCiEbbZnF*m}rSv%AC;@ES6FIu%%H;$iW{ zcD;Dg2P}WNzaXhj1}Ue;eT|A6Vf^N9`p-$g<& z3BauUv13598K+En4d&zNFiFuvxA?pS_R9(Chhyk&m zF#Nptr}HsXSo!;$WG8eZ?=tgj+8p=`dz;H>)0GLHKvEbf>xils=0vuBKG-S&MVZqg zCk6YkOM9euMU9A;yQj-$?!3JNb#r_Pv)0YHAimz%Iz0f$r-ClpzRZMSdQr7oCtI<{ zE~Zb#F9uqN?+CvLCU{8GS8TiFtZtBKd-F1d8JnQaH|e;M9O0uvk^b_5j;#dyKMnd^ z^0g)QEjez*V8c`lvHfUikqtO)6!%^1vk?q$TGHkqcnCOs|8u$T!&a!Qw=vuIsSPR) zPHGrGxQ9NwckO*7Er4B}^pP<}J<#|0MBAcQ3097oov*JW>gpAK<~((I+R$~oOT#Lg zz`evLh1hXw;5|A((>P)$qIA{Bp3V!hU&qHszN+i2}?4j&+NbOtnNG$r|{ zx>ryhZ}f|!P8i4NdVj}{Y7xID+9@{~!3^WiC_$cUJE8x0$1Sr3Q>k`Qn~VdB?qCDx#&k-EU~8GKnt|B~B95w%+8fAW0c2j^ zyeGVpm}#>q(^R7fC}OGt2K1XDuT#?e+dgY>FdkqpzZMMs7sA@p_I1L@7Q?i|OoZRb zpqeE6!)2m=6O+GCA;`@@tP7Kbhdb36aX4!Ib1u3Vm_4#(ibPArZ9oOprLkRMmiQ0;nmyB4*e9Z|Gc1&H zdp)BaJk|cKEq%TkC^yKDtiGMc@$GVN*76?$^1gbbYq9v+G$XV$8$rf3&_1CKLo!@Ga{U@K=XN4i-K8Pe!>u@ll?eD0fbQil+Dl09$9^s7e z-dc+{o1oVER1sX;2F80b=$N3 zC8s_>Mt1GN^IJ=C^mlM=>Es-ii;y>x`U$>VfiF2vvO57gGHb>89Y4r2NDIytPF=&Y z=AxR7=ZHAHNHaEUDV7#xYtyhP_m;K{qn+#j%g+yZekZu2F~1Rk4GG_9I1c^H1Gaz8%tb_DZ!_G_&* z$b;@8eHr&hqR_G6o}u_>4@|E)AJJ8H7G`^NrY8^PK-=uzXwRv4IJQ~0`N#K3tkzWz zcJFS5k%aJBw&oi+#@(VPmUR;51Ggq9G(E%`o8;E8KTokYU+lZp23=_4HkdeZya~|q z#me2=U!c%^&B~S`LdUI!>5iqzDJg6l?~F6iI}Ee09HN*MH<6fhY(8ieG)(Sj5kl_CUdOVeOs$G=)#ndMPI6nU z|4$!9Lci);%WJ#La~$aGHDZ&KgnC*lo!28Jn6zLk zrSi-TI&9b&1oigfyl{cetYsIL)D(I+9OZxz-Qgt0Wl@~%nA#zrau5eDs_tt9HJEsJ zO@_aa6HA}QurJbI!1`jT=kmF)aH!ycv+@cvbSvu)98nv`=3|}G<_XlgCyZ_%J}BB1 z1|{R^#q|3{A-iU|b@JoC`B!P#;dh$QnJp2ooH+Ja4aQGwoSgcy3zGL-(AKGEg0b)V zj%L@jF}FYSkHI9NhhDPl^Y(Z3s_1z2qYb^VCgk7wy|8)V5j1VNH&ObSgkvUqUo-rB zj$}QW%lm5#un4=7JmZxxe?a$(iXAb3I&biaCvx6`p>)0B5`{0=u66R9oc%>0^-KiI zeUtw0d4jX+jTU}67DC5?(7nPsoIoCq$PBSzhpDpxQO!Pvn7-vG9a9{k-_X!Q>v6V} zin9Z&{51S7OEN7!G`K-*{pO$^~TNPYjim6_FA6LXC7iPxD; zqw33G5zoj5MSf~t`fvYk*SYhh0Be5_*1!oOp6e#Dk)K-q!ig`8aW2ntaL{midY|!T z99P9F>N5`5d|1?$gNukG`gMY@k5mwP3V#xYRXI)*{MZ=XOK;YX^pP0Es|?aoQ?S0( zFk9f=OQ?zH{P1_!46+|ITw!Q5#n_>=!_!IDF!gH4Sp(TX|v1HCw$)4s%-`Bjl9QPo=0Qn9rctZ)w_`S>wDvySb7|pO_lH6 zDF%@t7Zk~M?Jyc_FLxm~f_To4<+I$LTVPPVR`s4QKY%=W)cE>=8jS0?m0cF4ge~8{ zY$}eW_LH!2tAW$DIa8>3YS|`zgjy#cG4p(-lT7f#yxobXI>!cZP~9b+G(^S8!FUa? zL5b4?s8|yu_EK+-#CTYr|0_!rG;BKbuC6N)TPPbZENpMYj?)FPd;I%ws=xYD+M$;~ zxz57-Q9v9<-?C)u+`UU;oe(IJb@jHBjtA8{_X)vVV|jaP;il^x|%|K{KHi+pmYhz8Ciy1h2$ zsDzdGPjfg53UE~Wb^WwxISk#GC#T)M0}HhE97kif;^N86)%%CnKneF3r*dv*=x5|B zkG;@_y*0;;*LIwTj=!Jktt>Xffa>-J_G023?z+F5DZeZkE4t60Pg%bg-h3Tk{jebh zh748;CZF>_>`l!jg_g7M>z0{&*xS=k*S{|BA=)Q|de_cqLGPf_kOYRJ+(^1U5XtALRcPp{FuT&){1Eh*ci zZY+(dtgo4B?pi{}CJmV%hlu#BOeUkv;|>dU&1v+%PTz!|WEzI9Z2JYBW20GKPELF;4-zf^NSAA7aXen+e1<<#Y}@&U(5;#KZ^u| zrMtr71Ycv!(e+i^_pKN!>T8U1g2OI%vlTK8-hl^27f zS$e8JwX30>5ll~sJ%dT+-^tBaiTpgdW?H?8O$Lip6wNwB2z}_TE+OU0br#T}{r3ac zZ;++Ey~Owzh&X-OeQbwuunmye*|YB(XaR|T(a|Gp4`9vTU+p3p*Rh@N`{|D7PmvS)QNp($oHANetaC1L@<7?8u|uR zPd`ca72m?Hb^Gro+FgLobw@iEt&-rEd+$(SI}Hxdrr#KOxedmY?P`5KsX+^j4tGB@ z!l}sicSi&KV6^j1(Y)Xr%s+6ifcud&%-RZH?(4q=%W}QF}A94Z)gt8Xx1_u z*7-Nj7XxkAuCbG)adVD)zveSk&Q=OsQu_ciPK(d$+tp$G{jyp?NRliKoytP^BnQl1 zd3T=dM~x$4J~E@5PO%+}cQV`lVLA)r9KV2H_8c--n1a85QfePVF|5o@y*RF` z4PCc->|Rm4VCs3x9{I zN#LnoHQocter2)y2^}Bu>z({%XHp=i_gMDU1ScZ@)4TA&?NSVk|E?J5l%VP-VDGNf zAHfQ%a2?d|bwtNcINa)l=7hl$)<=mtL)H4T*G(lHVc_WtSLduD?7c3}F`!D|{OZ5I z*k;}Q3;KJX&FP3&V{Y407#pV%^gjA~b!^iwT>1IJBS>QdP!z+?AKe*^6aBJ(<8sWg zFX8M!p!PMOTuGIb(W3xGdwnzuTMPAf-^IyQM3;QQMsiiIf=e{cS4ipKoR@%^6uNyf z#YDY={I)j5Jktz^v~G5Jeht7*z28^a`PM^U!qWCJQ-M|Y*|Iq+)nr=+iOw+KL(*v? zPV3nk>3QN9QAZy=@lko*LsuAm&Gl$q6QLip+!2*B9cTxwyPhumBz#?PwzmE6{C6)5 zRrWq`-_8c=PUiLM^%1(aUk@FVNR})B-?uS72QOQ~TtaDg6)&}aDNCDk&Mx0L8EZ~T z41PXC)qR7xKF7vw0Z(9*Ua@1##Cf2+^i_+Tx&#%ZIS<#*UeG4rQs{8T6MB~t`Z<@c zU}tc7JL}{(>>v z?G65SPIo&vm+pn9f>lBpzX-l+|Mo*wd(6Y3p=8$)lKKzGqA}IFyZ;?5*xYEfWU3#+t~*E67q$dDTB=54UX8hXzS!N--?|x*RNleBlH?GuIiraK3z$i=W!DQ zhbU?=Ws|rsh^!35KNYsFy=INQ=Y6Xm&8b0nDhC~xY%9)bc<;C-KMWPmu8huq&<5`% zrAJ8yY?eELEEM=70WouH`++ejf7_T4h444=)Gd zgfHLi6~=W~=y!x~ZUa?s0Y5ct>UswuaBaMM7=qx=z__>lj1e}Z2e zG%f2tQ1ugWbStgku^DL``gAko^OXVWzA>pOL|OND6J#je=zrDs7)C5uc`7{!9#q9- zSC0Fn7j}g8NSOx{`o5JL$Slh8CKz%CQ&&Xriw&G~#Vg6N|U zYAn8zy=4#WC6CU(daMMEU+zu2HW=Xl*%auH-^6&o8_6$sc|qrsb1U{onJ_;fb(t1U zLVp3{;=A9o*gMU)!7|wq`XsaTHq9Er^d8Q!i{CkbGDuVLzA*#+r@~JJ1^>e68vAu@ z9}}@6K#@c~o5J30niNrSLML-z+N6c$3mt}RVA#pDCE9^CR!CeF?EAL`znz70e6O=UTEseYBvTWunvD|Q;{ zPujL}_Y==!Vr$p$wXasN`F+USJ}+5F6qK=vS*FF=z}+drzX<)P?i4Ns*O!F;m~Yyj z+LG~9=(F5HGCI!>O*5oJXkdx!R$-M(7~Q>wdqdC-tMyshXKpC<2US% zByc=~oa1k%PcCBlq2Cwzea`|u9s21?{})#{%f8bKvOxa=>s?mX1mgZ{FAgMUe1aa* zpFo8vQLF(yxt;_9M?RBK@Av#h8ub5--g!uA1Unp+m4$b`+Lz>b5GQijz16#u$DJh0trKc;m)k-DlY0u2TnOp_58uzgo@c`>awEat~* z@E%NpzIKi?iDL~|k~K9knofA>3o5N1SP%hC$(!0bWy_auo|n4hq#>MOG; zq|r9!Xek!LMAPnDj9h}S6x;pR$Gr+hM-?fda_s;1(I1Rggza(ss)3>|9;zp}>cKn2t?t#vVDC~mmpABm0 zi2pBgzsqoiz}F7!%au^rLFH$|RG;yc9m4xyBKqn!tz8j_ldh(*>(j9&xQ@@1brtRq zfB*Wcp4?675KYh4Wh)r_67fKWeRxvWZCSeUz&mv^#C`WWi_)@#77VG1Y)z_t3FIaB zEyMnV4pY;Ev|qnxBr!Ak_!j$k258myKIq^yjTK_SKkU{Kx&|38r}lrR1jF=`mXA4p zdth4JVsl8UDs->EC>;7)8afsFA8n;4)?HuVuS0vRqoF)b-JHdE>#FD8&p7{n;A0@P z3K{6Ku{dML_q7vBM_2cu+kN!5gvXoW?53`&Z+AChYFW@xhnp|4r{DFHz!hVlY-W#j zS0?(vjV^_16AeA^AmcGzR|ip#rqFX+UMMhzmVHCnw$>GpEc&-zXe*J&Hua4C^mASz z(Q|f~omg1yFIADZw=u5h1dw!+0(L$Vf!>-1=Ig#f*nDl`>l$$#7`0OWtfagjTbbH? zUee!#O!e!Jd2bBL(n-1Bcr+RTod-Pq_ysKBmgq>5F@*-ll>1^7eg@-T2k(XGdxS1j z845K>kN$z4lqx@_m>`_mu1zEQAq=v_KIYv$lnCYH?AKi#!?Bifz->179p*>=P~Kfs z54ANbEj$~ql4$dfXr3k2B01-1(U`*%7=Oq0sDwujI^XOMIjm3kcGiX~={e2!K~K}{ z8q1U1Flti%IXZ7fmi8xa_9F}G{s(m{%-}Pt>?^tpd8P~uBu?Y?#&g@=A$hjw)PoR?2`^HF}ezqLIFvFW6@Z8_dZ9K zqbg3=ExF#?eFug&X;o}+GlkE$V$&BCiG2y_L-|!4C2*idSjBDa6`av|!l-O<7cs=F z&qjGO_ShE)-Lc02MskYoxkUK!He1dsos$>sQi4kWBb`b z%v#&J$R6_==CI!TaYYu6_;tP=jClkzJXXPDvLe{4drj~zh0xC&H1++;Eo1;>R;lL4 z0)*~T_$T{~eCL#47%v!&Xi#-IaMpEiMIoCzmM+EK{^(fv-{X%nyEV_8jw>h8Crh#h z^VQ>Mnj-z%QNpi1Uo`xoH$5keH>7wf1`_LaBy0DUN>?(@)ofaK=&e39J8ZXoQl17g zEp{K=GWoHgUM+X+>p)^($u`x=wZDZwM~WYBsJVv2GkRk>>f=zJxLyC}wpT!@(w8{F z-%p~o`*VKRlRy|A3l02}N!6{!Vnru~mMKP@2|V;zaDG2@Ys%jb^=ZSo8Ea@M5!-*|7`3)qN*HvX)&sniIf5}QDi3h@O$b?x` zutVy+1xKXqYaAs1Ne{EVliqcmy+^Qk@Z6`iFe3h%Su5BRqM(CwAK!Ss;9-I8tbv+I zzmb3Q;-K^q_X>=c-Jw(!P=ytZG}|0c=RpSFU)K2_X4ua4OEgI+0EgY%9*TQ@C(-Ae zJj}5!9%@D3pY&cj3Xg`e$|mJWP!iMlbz52{b}60{Z97HyTD5JFnEKdDgH?-WdVwn% zFyq>?asBo@D0gMh)#M`ZMTN)2NN2vP!!#>hrgSwCA5_;myXFNG{NshXFWXMtQiA?2 zd%^px1IXwd3my64Qx=X05{pu(*3&%Hu?T5^crY=G!JyCY zw~I_=F#Gn~j%UfMG(ez0mc5`IE{c zD$fz8->qSAJ(>z5*JA~967N8v@^jVrD(JxMU(sv3R{rYj1~XGLyk#fhcs$&>67iUBgn9 z`}=x54Y5D0@}l^5d93}=+5Yq>7ffHi_G}aIV$%7tsx=|EVc=)hi`mdlw0oN@;xb9X z&g+)pPfI__(k4>8Y~K9@%1-8H&H;89+xuv*h=(He{)U|%J#1<#h7M25riHl&I1s)5 z>jw2)S^85?8Ry6Uo!5W4L;9LIFvY4!^L>$k3<6z<1tC%lWOzC)Y59Z(GMCfkhgQlM<-Nk$taL`>l zYmu67b}ZlsO(_SZ3P2Nf2u#SynQoP@!A)j zcB2RKv1Xy%yP`Oxp4Mdj`zzLMxZ!*yfDNW}%6t;-#IX0c$Zs>GVBcaZv-e0YbRN|{ z;NeKk%dkc73&-|9a+tbh!rg~~=)bk@bsXt;xb@%jT$WD%@93V66WBNTB7nc;4m9

    heEI)@q1D>$W>r_pHAN?Pve`|H(cJ ze%B+5))ywjyIy@?vl9g0WSEmTTH9eAR!)BEa4AT``HKtV4=1-`=Juf5H@&!mJ6Zcc)R#dPx%O^4yfw#Pr5-E<(M&}&R#ix^RNGIzK<)EWh) z^0Bs2*(adfWAnNrbvZar!zT9RmH}x^_{Oe+f)yBiJ<_*rw+&3S8kKugMq&BQrSDGB zcVJ%l?#nD3gweGvOJw@15Pav^+j|3AK}I(0rpGM&o{p;#lF(a7W# zTDN1>KGo8vqni~cEz9WmC18AJ7?vBC!jFwUhv@_>O5u7!PqVi2$p*{IlQ=e;+re7W zf~}fjPu2Pquq2N;H}K6W{brnh?c+I?MfBlT)_rdjpWlVk+##cFwuHY?m3OfGIp106 zawT^+)be7d$jT{JmS)JvE_f|9!U2=dcaRKKR`GbC(WzJ5OR@(#uH88jD0vK{cGxy* z52-^}n?ZUO7s0#DO;*u5RZQ^J$@zT}lXHZ=T-OuTEytV*UU{`a(|gt`Vl-R&3^?@%81Bg#N0fSjust@6~s=pEJ{z8x~07 zH^|JZcwIP<@%wzTr9O05S$ndIWxz6VFD-*MUO}7 z43p^W6RNNJ5ICLBs@jY+Yc(N1x>DKc$YUUH-CTQRJso&bPQ5K%+YObfXZETajKWB3 z-1VZHN1(#b(;N5+99_}Hz22He1n#tv?U_;xr3omoV`W<^vEPunGgA(v>tP>z*QI;f z-&Q@3?z^|w-Y+72pR>&4EINO$BgW&sNY668I*u{&y{=9o{L4BQRQH*S7$W(XZ-jFv zH7~}#-3Qyhg{fkrp6ps3f2toU4zeEm?iTA#{O(x(rm5^3FqTeP5g#J_c$U9^2_BMR z!S0{7Q~wWZ{~b?d{Kt>O_DmXt%BX}WQPTE0WThe^tCR>04XZ&VB*`qLNGN6RJ@P*G z-h1!8&N&==+JK5*#6 z(VatR_{GIPwB5hasKxJZ#rYK8WK`sS-2aTO5emDMPIq@(I2sT8a(L^+8Z1@sJa@hk z^FI{CjTM>~3Ige5zhZ=`B#^bVcz?#`LNon-f%P{%u>AH7nM0@wT6Fk!w3bMtKmI$4 zAQ%smpAEVl+ODBiDe=v&BkG87wu4_*p%lAM#zLZB1>FKdlD&VN>|gz5#kT%Qz6&Jh z{;2n<#QQ+{aH6t4mJ;8N=6-kIP>DYUOY2V$1fWS6-O$}hT8pIoo~|$E z%JeZ3k#Cg#J&E5wpwB48M1A-M$?nqU^OUf3Jp>2z_nlBLb+-SKx#3 zQHuQl4Kn@VIN{uf@=tHJjP7kfb48)D$L1(JeK?wy*f%#kyJ1XOF`{ZGMIS?Ht7f~z zow58-vtjm34|Kv&My=u*^yB1e+mjPB4 zHa>5t{DEoC<;6V>;jk>f>#UJl4YdB|i~^Mw9F2GSaN#3$7lTE)l?fNoQnR<4zvvWn?%{Y*lNE*PhF-=kt)!tC-2A>` zO>-D~8xg#2ZH%2CU){~mKkI^m6#7bL#`q1Uq#X$nVvVewm|UMIq8 z0#cyJ(erb-tuk~;UH-jN_!<_f!Y+3o@x!yIH}RMZ<-r7$ah>>XjHWNP4z^uA3Bz)R z&$Ic;(8959wuC7o7~8zKAIa>8)h{a!nioW1g+FD(c%BR9tQSjm{=s;(ryn?p{(SiY zNS{QtZh4Vmwprg@N6P_6pZw%H?uQ+;}4ivtGME`_8OW$nH_qNjm1e3 zOT#*k9{L32MdM|g3su;;dX$@a--ny9cuQ3emru&rDx?h{%Jx2B+j!fsiKcU3}~p6sWo$kP50tgvl5W{xfE)NgchIQmSsm)`ZAdT5uY%|vxt&%M-gvscjx(P&?xJSfi#N!j ze0W+G9%JL$=kV%}_p==mAvi`>&qWQfHdy%VQnY{B5XZc&d_{iQ6vnI^b3WX(fYC9w z%j_3%s89URN!`wuI0l$Zy&l zfPB^Jxt8r^XqMNc6?e3Ro|<)dzlk29a{NUU);}aTgio* zIIc{V8yX=3aZvxXA%>$|5O5)X{csmYgX<2z(0c*LvSlav=Q5TbW~sqpZ`w>W5Toy2 zC^`ObzVq2DMZ4eQcLMoV-kWE-2?!sHCMrLQ!^*KSMb>~k=<`+Ib3xS#M>oqiB*nM~ z&Bm;~Z8PxrtH(SlPguSSD>9n38=601BwCm?tcnel|16TL>-`L~@#~*2J;B;DX}wNp zl5)Z7mDPO55!a7avbLSS{=OeaXQH5|6GV>^+AZsp%rNj; zT2HUR^6wLZf=qwUzw5WROIxf$dc@0QS29K4Ly1lAz8`PL{BA^L!l3B@B`z1u-8wtv z>GlHqov6W}aw8drzCLE=7eSPMPS@uRbu3W$3DK&Y(n;m9E~@!jc(1cue!mFXYTmb% z9Wz0*k==Plfiln^`ilA1;}#%hzxno!_cf5l=)_a+i(=|QFutI94a+lKsa|awd}SF3A3O;+-qfRd$bvbn{;ds@ znOB`wy0N%er{GwB+sm6^L5Fi-tce@QJeTFQpJIANi`Szpm!Fpq!C(BLxf|x!XnDUn zbVTAin%{k4*6(x_tccg#@X)~c6_%#VBXct_9L>ySkGdz;o1p1V%6!ujh1cS*oMDja zR8#}Q>PO;(ZR()ke_nnQ_bJp7(5R8)Pw^+gF!PPOU$iOq860g&*Y*K_3HWyFY?O@p zDQG&8oPM%18779p-c@K|JmalMFZbygOwSn0I~j4dU;!o~yy+zDC_E#u{!L)#r;}J7 z=Xp;($6p%sKvqd;kk!ZdHiul=<>Oyb_%vX`KxYrF3}rmg!uXjPw?v8`97hu;IDHLY z3Ios3nNtOrJnNYO@#c3M@eIeTQn=kI@f?t~*WFd|HijQqIH7f-!SVuB)7c5wjV`19 z&8G{_?kj-B(&#NAK?`V1YHyVMv6X+V8|ZMlmpU&O2eXHIyV&d~_yW{5e*cBmQ;ZKP zDg8>Z8r>E&R6V-2Y!e5xjq1Lr$fWFdJZ*`g?`}b6NI(1(Wc#u94X3PU?bBMMTCYt% zwzNd}UxJqV^K)lM1YzklZpbzVvmX)|T;B@#V&^bXXOSVo^J#txo38;mCEt&(Hq74j^)kl=6Wpb;)>mlGYKus*S;DXJZ_2QSt=z0xjg z1@rrcLs~r_0O{Qsrl_GcJR@1?a5&dvG#Z^GeNXlWIGoM>>d<==`szME8kL&E(e>@- zT&W*}aes!*rVTMbej=`6lEwg8n_@(^nB9kkU)(vd`?2`Tk)i9pbY4yix|3Zb^^cdrLU@=-)c7^++7^SOVfuK8kd_N1uLn_2&tme}hN0-X z-0v%}^kOXk!FwGvpE`Z~*vv5;qu++~gDo{^B-{MmYx%$HbKz-g?BE?(3gf0ZXoTTf zdQHs7K4wwk4$z1yEnPbgC7u_$wBBlF-^c9QwV!!o#f=PLDO95GB|8>>J52kYxN|fK zrlkIyj;f(SC4O=2oOO3$DxXQ}h1nYQ{?c_GA4`dch5}{13wQ29>1?&(Lc0m{`46dB zR8a8ncp7@)ZEW|lp}&5H%cJQw)-K6=g+2-=p*Qi&S%!Wd7<#g}-iLTl|DuZrUPYN| z{jc8n($#U(3ANO}T}bZThT1ttm}e>rfpl%>nI2zJ7}&?hxaV@!U( z`9-Wcnoql0mizt~^hI4yoO5KQ8lUxo8HrGHYJUD~(cVz{uQ4#d{nTBYa}SyhkDhtx zj^X(mPCFmsr@I0D9M{UrT@OQXaq+H;1a}-wu+3_Jmo4f`QMO)}(nrl}H3K3??!)Zp zzF!(K8p!ctfi&|a2`G3LEmEib6=uoN@om2QA-j={;mM;?)Tgw3IjqVE4OuW42=eK{ z{BJukCk2cLzDIr6F78yS{OG8n$bF>fojvLqdC#^Gun&r*jg~s;`%pzwxpO<)6snWS zAU6njz^sk>PRm|7=$R7pa?RQg6`KSW_Ewieu63`XV(7NR=g@-qZAS@e@dwb% z9pwb+B5d6A!yhs+6iA`05W?dHi+WBhok7-U_I;C+XmdPT?m2J| zCXD~PU;n@U#ckr6Z|%QlEY7B;U2n__Z!V?fZZh96XTSUl~F_?g;2|E&L?_03t0#@kp>^1Z-f zftpTVo$5c|PfKq)ye{JOzsg06l3Zd>^D(?UN#Eg&=G`5rD=~<*KZE`Mww+Y%C^>LF zl*Mfp6ASG(y(53WQz|O_npF`buMU$kWo_)A;mkO@UoPaF7smd|<~ zi+4%BLb~wA9Mj8&xH8{ZA2^&R=C0rh0!pU9v`~c@ ztEwT^*d#=FrP5(|fed(Ay|c$o)xvo0xT%r`rU&Qi@oZ8R4n^5vv?qrwl%W5LNb;%k zbkLV|fnnH+BJZPeB`M!Y6$<_eh)tmW5V9}#WF8`V| zTG}*wGvwVLSXbx5*_2}OLXCy^XIv8)&UNUGm7Mw?JyiNjThTi`7;2amamO8>p{iq? zax}8naWuPbauIADV1AsC|p46{1^&Q76xMze|Hq zEL41CT2c|Fx$;vEu^GTZciz>_FBqUssgJ#s>kE`>oKf9+B@B_i^VQq<1fvZGq-JV58H&}X*o6fpTIMy zg{4QwJhg#fr^2>v_V2y_R{vF96~-^z@@-0wFX{;lixFG(U(3Q&Up>vGvgM3-ouq?U)=_0n(dJxuMAGkWx{u$2{MPW4z$EFReTg+qXYubk%p< zg~jzd-|~gVQKekVmf3X`9F0Bq{8?Fw{QyV%@^oIq6y}#CbZpCwyZsrB-SAF78H)Kw zy9hjkMq!nxwYyi;I_DykjQBFk?LL4)vvTI3pcz z+%S{A$w5?p>*t*m zJpuKJp0cRV#pbHbz-5BG`q^+F@bNtiD-P0oI+h189B^qcXY! z6U^#ZKD4rLy4$(8-G?T)n0=mp5JnZ;HvblEf~1dO8RuOM(9GuN#F65=5H26`XpH46 zkc}b@k4*drFH8OF{I=>;9D{bs(e!Ir-pfhFDL#u2L1=;BXQIpaG$KA&O0N1?h!(R1 zBu78d!yr$SA_M6*B3QceE4Rg>h~!s=3_mqs@F>lpCyRV2T_!vplQjFXc_pQKy z{AOb(Ev#O`b<_N${EOgC`^8n|?|K-Py)u(|BnNHOC_CtBi=zBbq-&w^vXB;fHnfQ^ z3rByGjkDb(5c=O%uIMw9QL9JGv8nFs&@lV-Z1DmwO0WNJrZ0=-6(gwG>^&?KftuxU zuUeA_VLjXI5AQJwE*3SnURK+6$N*2@7t!7G zJz(NkhNKyftlYjG~VJ(`DJ!C4LZq(PE!oFp|?7yE=ekZ3*n-x5!!Vg2P%llVGjTBr5 zo^{F7A@GDYnrhJ8x^oYPkDs6XQ^Vp;g>QqVeSGC5iWqLYo`?QSQepw5h~*#A7wU&1 z7XuXKM8RD^ZRwDppt2bnc~~cN=$#i@@)72{@v;DtB6Z1&-dG;JSykpz^V{N3tkO`K zD<2P|FTQWN!r}7we5jyh*EGS?%wExEvnQiwg43+s$1pT$+isHo>kXdP>KvYAO3{z; zOtp49Brd&x2G6SkyH_5ewE^RXJ>GaApINsUh?|1`l`ki+BNEon^Vf8?Fk^Vog_ZlN z%J-#kv?ps$yycUDDVyO6(<%Y9>~Hsq1&+w$Y+Htm|ubErUb+t0f$9J+^wmD4PA zU9tGl0_pUmZyA*SqLn45TnYO;RC^)wF!Xu+-aON@Ay>a(1uQ*Qlt|L|Jo!%bH>Ss=<|+=p`j~Xnk99S-5O3^gU4s zb=s{C`&+KEr`|3tDp2kfeW6sEgYMeK=)>?nE|BpNzTBPb}hjFMEvuKZK zFhfmQC-Y;3vHhJNUn=tG^;?vFFKzkS9V0X=7M^Dk^$3nQIR$UY5!9H+%yTGZ6|#Ss#~W-Kf@ivhK@}_-8v%Azi@6r$-}b-uG-2R}nx{#U`n0g#;M?qr`eUM;>)^ z@t6wTq=oipyDNWrpMbhO`z4fx!%%m;eg}tOH71ujI311R$b(+(XqSwe6h3XVCR%&) zb<7piobvI~J2Zf2&BO^@SY(5iN#VZR^Skj3={qIgHQ$BN4deU$n;l_BL}d5r0~Egm znkwdvjoF6feHh&{eY!xp2NnIC9h?i8hQ3Z!&6JT!s`(gvTvD#8_7l&VPj6MAv4Q#& zbDcNLY0#=Fhx&(GfAwa?)*GX&PS9nhADZloN5pFf_6xjgh993PJr!$X2q^S4w?LNkP#Z^4FkxzDU0(+!1xZ`F3d2OKBKC7BDP{a1bD32ShpL-nt(sKZQp0L0dbN@&FD4Pz=##4K6ELo+; z`7>EzRAMYg|0*qN*+FLMiFgO|c~@>tCo%o^_)zLyj;v^F4+}^01erfk=Obad)6{*R zX&!2jJ>?mtt^~w`TeVVMOYn3K!@sm+%}{5=oXwe(yJ&jjRYhX#Ej-Jw9fpdBF}@?h zZS`6~B}%?+MC^)Epq)-e(`7R+2RSjEYWbI?FM&=0|7sVCIiw>O`704B-UV8PH)HmN z7@v)h!&p8^%JbFY-&IF@U*cI_=tL;RV15CT5A9r2zAhR+GF?($V+f>--PyCo7ooWL z&ASs{p5vMK;ANk@H-#EjzgxqKOaC0F6eY8t{=^Izd=;vl6g&Yvy!MA~alXYd>HqY< zytErN#F&KQ_cWo7?#6{X)>L`EQEoZyhcoheuvY&;>ZL<4On&(I@W#G7XtXl8hWCg) zA{HbC|r z-ZAp_1zI-ebo+)^#M4XqU*@WuL(KtP@6Ww9K|~LmM%L>6lyYfZr@zT9^dl0D{o~_y zRq);SwU~Z5C(M{lym4{s1j2U3Q?Fi1pmo~~=WRsHk2q!eeYKKI!NEgmxDKOeDV7)4 zce%~)RS}lIJ&UtK=GTQRwB|i$W6XaE^}f||;qAoolhtTlBtJ^mhdCCX%0Thc3@w-fJRMvP-LO`F0K z2d$s4|I#Lk!jcUir@ZiGXp_4Ap^<8ud4pQlo6 zo@_1}>R-61G;NN4TScY66n4S0G;;_hd!@tpPrHB)QOxf>-J|&>0(+J0)tE{!d;~+Hk@@^~`YyS6oz|lK0NVPWAV0MzO z#(oD!%-=uCcDQu#bs&%)hBijXM8IG|z-dE{9nhOp^^DADfTjgfYP;#Ndy05ig(+dj z4)D~VJsNwk7LsZjTKrZp`KVH@?2ux3CK?a(x%7}#5REsnIt_L8p$Sv3dmWrxvHQ-L zgURkL9yL4MeWfPDgGdj*E~}jFLmQ-G-rF}GqJ=GM>0?P0|2`U<{GQR2%5 z4d9t*>DceP>!Rh?Nj`pg`BeLNY577%i$5dkJK|$+W)Ovjvf6%@ShXX!KYAJa=~q!l zV&M|wwlwI7j}fK6>IHoY`wV3s=iykudn(T`5;{tWr`ECs(PCUm&B3F(u)c4!|4b1j z4hu&c7&Ky~`4bz*uFaD!;h3C0Q1n#fHEBDL1?{9SBw+cQYaLwf)+UdlM)lm>P)m&0 zg-|#>-jS*UguzX60k0~7gx4vnVk)INSK^oNAKjaN2vv7^++lh}$&-ra@A*D^YJU>X za&_#R?4=BS_8W4n7)|ta5LW) zQ)<8AC}%|U_G$w>ohCOI-8CwI02GNf^eAQVpta!5Oi=jzk~7Z zu87HQv=II1Ro8V&eqk6cIT2z~{s2w?ITW>h{V-b3l03B37LTKOT6UD~Rw58A>k@_L z1W;jP##7;wpJ7?>W4{u;9Q5VUT>7f_SFXw1lEL?S2*&7g)M$B0(EP#5&e-TR5c!8J zcE9h2UblGCIoVzmC2QI^o!y3Gs&ZbT{V5KEmjukOg|j2_vGth6*uL`E$_(`*n;v7IBoO~{FNuvSHCKB z*`hpsfh2)f6R@zSiJ_b=5jEWxlDCe>?vuu3Jw3;V?@{_`mQ(jaLx2#@BHbRh0mMVK z9f>4fl;2FZB`H}S2w{89?(BB}LiuKU-p^Q^M*nrrC2`aP{Z6=@C_5cbclNv05iLom zi}F8k;?y@Zx4IUY??YwpgVB%fT;~kA(A0c=#<Wc16?xBGIEt%sO<2##G<_drdR8vzkaYM3q}p4qlMEiLov71AkEJN zH1?-?Dp$%DjU{maOhB{0>)8fV^hz|hO~XC+kRkMmH3oMSd83th(t#1-LNu(yn_E4x z8Bgo2V(n5li-!2B-!<>R&cB8OUn5R(aG;zuq3ydQ_M!LFQG+UvQ&3+#w|o*AtY3DvRa;+Cuh`oU?^f7&Cka#u!M@bA2L%dTD`LX?#r#6Y*u1TZXrxgq; zdv8mAK+Tnf3@a;`f4J_+NPh8FHU%9# z%lOA~xw9Q;rBJkG;}?aG1y4)&>oaNp7u3{w+x5G}XBhD0Y}tP`2WG9g>n3WdP^XMh zP-5&uSTVPff2Xhmij{fAbhlyteUjZlE$ORLkQTDG^Q&0xUw!y`vDlzFEHUT?^e1-U zn3-D7hV=`*Z8h)dt$SU!uC1uB>Ddbe{{87>7%|4 zSP_fiD?Wwb(+)o>#{^>i~0A}p_Xs-IG#w68@!NhL4T7*>MWD~4lSakOtP zKRNFmf!b8xXuJ(q`Ky13mhH+_gb@az7Qy*qXzwKdy#MVejDMRitT(~-`ycW%f%<6| zJY$2L^~!C`j?}#Lki%!+MX0rKns{#T3zbUNE0jAwKud!0q^lEvqK~DCn`>ivCz%K5 zr-S7}@br8-hpZPUd~Yy6tYpWp^#xDQL327}FNK#94bezWA6tHpW&(cD-!RyOrn?8D z_8z&1XJHC{BR->ur;$G7`{GhPj+QYptxP{Q~YV_fA2W-i|uXYp&GAt|a!a zoS~apHdjZFnvy=9j^2sG(Y|{yt+T}*2#!V%D;dv0iL->!(<}`+F1r=@kd zZhZ#*vE>0XUhb$t`wI_mz#~i^6-||7TV(;#z}8Y3Aqs90jn3Rw+^~w{sJc$J&*+2e3S2zr| z8N>)TA~fGw(&Caefo4A!hA5dPqb0A90P|-I z8SPB*ChY^#7P+^$qo1Ky-&*c~8$NF#TeVg|^SV0}bYX ziA{WJk7uL*wfc&L`8V5k@5noO595;^Kh3r!wqO7f<#b052lJwp{U#Z?QqN(|MNcK@ z|9mgdE;@J*<1rg1@;>?`xEIenuhah;hw-HKpXFB`t-|!v_9$)jv%b@a;KQ;dq*@n- zPZ#eHD__CU#9cA@IFk+d&_wHt9lmIV4fpsM(Fy(-m}=Ej*&$M6Rn-tE}_sdRy4ZI^r39vC(0RJ-pLXi{?@Gyzqij_7r%BXNv-w*>==Wf%v1+d<6!xku@%T{mJxDFs zpUkzz5zmmw_5OA8f8bpkE*$t~ej5$!S}$*?|AscCrro}uv4Q1{FclG-oe;>(E+4o> zjjFy`8jVD+cnYr&EM2-XyG*wTlMw^wFYIhYqY7<*w59IDaO>izuxu!v*5%NSG!uT* zU`}rAbPj>Rq}g`GlR2o4->*(T;~*NZn(I%$i}B9qS-!n`rLz;;N7wRL6U8FY?3Mb? z{2i;RP#=2hf9rOoH0)ve;q2&MYDsCvwO+#nFEs>U>5UAhz z)KT|A32HSa=cczC;aQWJ^>%&v2nhr?S;Jp0(Ec#h#Pv2kRC9h7E3SV7jlAr)mbxf; z3ZS9q-6cMsd4%u#YL!M`0%W639~{>LS{QU2ec(wzh1@lauWT^+zIBn+X~9+uPa{Lp zTy3TJ!SD>Tl_^8U$}sJHo-b;+1W(V%y=j-=DjHF*KmTIc2WI@VBm$H*pg1K>aYW%f z@?)7;tnrP8@*Eolo}Rzz%0 zI4zpkg_>WcTK}kIM9mHPCfwfEu=J?meCuI~9*U|N2oo{)v~Y}J8VO_Pld<#0CqhX5 zxf|*cG2So;&IjU-iKkl&>8)<3@{QQl`bw^@CCi#rsj_I^G|yNXIh zWll@;VB^?cM{*sJ8AUN`rr&qB!qP`P7UbPkTkUBNy_^2`gHSHZ7 z&h&?E*h5mY=3sMzH*dQ4E^YyA~3Jt3jn>p+{GLQ}7!wXZbq& z(5@Y5^~jDr!t@k50Z)6dba_v35i|y?ZbWio`C&I?`o&Lgq3~J4jM+O0J;iNkO*eq| zb|c2)8shTx=RPJ|G{F1WKdxgIwZyu4cz@mwgFk4QCWP|QhV)IQJJk7iQQOG2?5(Mi zXz`oXmu`Do82zy@yn0*~Wj1`>{_JxE%2fKjquvzDZ$#AMm-`)F1Usjr z&rBjxt!+^=r>Ql8jv}wXxcV~}%X2zt?6|7qHaS3T-nol)3X4E0*HubkBtx6YCYjxn zT2$lI-BY-uVRB_582Jvx5ZM*on)_ijZIObR~Ut^|!D6IOy<-(Yg`MZ?bh->CXS@b~Uv zcGQDjd(t2N&N1jaKD=jd)OKi}c=qheTPvu2`r&feEFGRz(%B&2mtx02bAwl16t-jX zh%_wh zkoomGj`3#vVCBIrs7|x+wVQ?(RNAJfu~&%T>5b)M%q?mVakI*I2CZV$^NoCj=<^J9 zywdl37$}VT(kf;oO(IdNa8AY1ND2=VsvnaoUw*|3bv$B&uG=a86Z9$hct7u;7FvJs zaFbeeHZ*g1J=I&p@K1~OGS8I*84xkL#bSbW0UCEaJ#(X&GHy7!mL=KiM9j`v8}N{< zw2bkA_B1+QuCIFpt$sZUBCQXR*8y};x{82i6Ao(mlr%s-?{mX7-bpCwxT#WFdI_~x z*8kqSmkX^MU*TH+I!nDSs#~Oo1rI~D$KbkVH&y&3BIOo0?oqu7$y*fP9L;+UZDKc{ z$yMf|UTxVeDRB;HA?N0`L=ObD>|y&&>6CzDn@0>F4r%R}O`N`hHObHkAHl^FH@mM_HK=oNhwx3wMXT5@70PXb( zfGg@vT~9DQczlg$-N=OHZ=Kq!Du7?5L#r+<3^x=7(OTt`k6Q;9VXQ%xVEJnvEopjX z{vf2GUKMW9miQVxv!2bHwq+$WN2lgH?Jxr+8m(&nhmXUW53BLd5G+2XTe8twZj&@D zU3{o{o`sC3F|Q-+LzK9G9Ia8Y3Wwnx=$kxNo3GXfq}KywVJ!S;9fG$o6PsWm$MyKU zdOnQL{kWu9O^M4x)wi~xv(nf(Qhvy^Fxe0xVnkuY{=1ldHS&Elu4z3E`p5U5JWLw{ z6ZAN^Z`5* zN=LDIS$}FpBXRgKkT$NrlQyD5nZ>dv4;z=FzQMi1H;=wYNuL#t*~nnL6WzAww8Yws zpmWGhDPfvQ9>Vkob0bcf**qA!I{ct!tpeK1cOR$W#L6L$xE?5Y{Su~c4cRQxJHnUY zir+6R$52hj0p_r1kAIbmntrL}Zna~E0jb78g7FIUBvxw~&}3lq#cY1?@J@;xhGV#{ zICrHRi>vJCd(v3srv|CnIIWM23~2V;;+N4=M^IbJ6Ol(@Yj`^6Cx_&Du<@O?;MV+L zv50E6{bX}i!9kAHrn1+8f7c~0^NC4S5?Z7a8_Rn-f&y}Lt4}HHMcsX`y&ASkqmT@{ z^fmg|h+tab*+OJO#D#&USH9rTiVbo7`Xnu$j)Qmgt_T)CFf#N`A-cZqkYMh_+3x`Y%t&fQm0Zr_^1%Un0q_Wg@lmP{p{=3;2U$FlFxdEI18 z%n%Plw*>YWMPqq?TCcLI-4VNtMmf1ku?^r zXrJ82efbOeGr}rn ztNlj-3}pz|t{@+<}c-C|7yRPqTLy1q${~{mlM447+YfU4tyhBS@iymx>pt9>?_lm~h@1u0`2=B~b zJg!d7=P>NG+apj)3G$62zC;L|L33g*von)P(2@`&6n8)x{jRJ|TVl?o`mT0W^`+~I z0h%1Ue%no9Gm6dsF7}9~1+os{L|vYHp*ftz+Rm>%K=^7FbXXz`jh^3LGVdOTmdl!U z2I3}Q+@QK3-{&aIz5LWTFz1O{Yo3-I#Z%;Gl$)!`>c>zC^Gat42WL0%tg=y)7Oyg3 zIy!W2sEHTanA5_m_HBdrbMP9r7 z@Uad+UXb){%AP_kR}ZdPM(0uLou!kHw|VCO&+!~NoLf6p^a=e)yj9>G`W}^LnF>|4 zQN4#}I42emw(=da75mj^rShTtTJhKc*-$iFtsEekCV*#BdrLgwj!@P15l!wr7~c@V zmG8OHJ&ODW)0tmr`WnaZ^qS#caA|3%(XheMhVcmMe|h^e{+ct?F>e0x>gNgQl(*+B z&^QCNPS#F3WM4EtvLpZDiZ=9TpBz@ZS_zG`VJb5mzA&17IoOYY#f8o9u-g9W{AH?s z4(N)`2h*H`Ie9tuQQgOQhMbqFdqnHd04ZLFb<;4+j=PV<_}asQV7iBHc?0w$8}3nP zDMW@LSK$b+Ecg<-uVCQ@V$L@}2*4)yo5#<0=uj-a$$|UiO?q**5P(Li}P)i?u{0P-5 zUN%rWK1Dsx_{Bi=p))APyEKVqGa3HWe)MW-4uJ6w-eZ?~olyHKZFjiJPMDVqml|>7 zLW@mTdR!z$pf5|sY$x#vTHPNUaO|QmYJc~v;MzBBXmxewd)s#y$VnQH&u)0Y)JS7X z`g1G~L_z9rv)z*UXs%&3{Mz}y>tpX-&nf|aM3mgemLi}GqsbOg_9c|KUo_T!k>gOD zJ(_==`2=+Z!0@YtjvWD1c@8l-SRnV<_4O2JT6(3`TS|v|51g$YDf|JMw=$9iANJ!} zw|DPs8GC~|zYW!=zV3u|PxY@J4XyAwM^45@5uf+OS(WdD;pO-XZig`aDqmn{oo6p*PhId)Ss%5tr2L-|^x&we+Dm)Cy!DaEcjZZ_`jh>q&&vRjQKUldE1w1 zM^{loB6p!*&OuapCCo!6RvX8V_dGF+xe?OF9<`kdWgS}1GZ~l9n(2Aq}+9rnYsI+XRnLZ@B>c$bQ4H5Ne`20L{R{(2 ze*2y)z;POgaO#?lQav}Q83u)bhY|s~4YNIN^h>C`uYy^LfK}{?_WKi}9YPx>)*8wYfL^^ihq_#YE zJ`bcaM%=GRgnW(!bBQNBXzeq5frM)x4Ce_lec84Hb)GzFr>dVqwGO8Oj^%O6+5vH> z-(Ai79G>YN&o=wBb})Y;N6k5(8HP4YFKO>N1WSiA3M=yDQH!0*ruJQ?&=VyaU2Wt7 zc?Ho!;i(inK9HV{dQ?)R!V?5nx@=)b#*Yf zUFs9ZkIOJoBkOD(0&p1uul7@niY)^T{mqemQ+dKEw?5(w*B}cB>LivH!|hnJR&vU>X`-YWoeQ zAHI}4V{G;XHXLs(ii=U=Inl_3mPo11IUtzv)mFXwXPhUFHeJhi*CrR3Iz; z%Ja^U>HXkCg=>d~%t9N3={}S{C$zaX=sWb`KJIQO8ph?Xz)-gFeT5$<4k6x*!DsSxd`?MF2J5a;^>m46ksO-v=@#~Xo&{8#=hqjqz zb=5m=kjm$M^`MgsvN8R@Ze1XTXTkFqKXxjI4Vo)s$BtgY(=D}TK2Vu~3Ui4cuf_7w zV&BHik9YI_>MN(!_v%q_A23m?6Bun4MpYl-=SJQd8hKTvA2G5W0*u;JYh?=Cg>4s)_y4=;Libj^Gfk!v0^4eik#F;ed z%yJqn9jW*rHB-L7hPt||1*B(am~#Wl=}tM>8m#~3U5apbsko) zqp^19#H!}rR5}5J!;yyBJE`p8|6Et&&tj~*M#Rvf)Gcz$8A{$xgireX(|uPxjB0wG z{!Pa0YBQf>kDHvL@GQYX#D|(A1{iadRolApEizxaftZ(EDCJX)D zA9VCWWD>?xQ0&TZglXzDj^RS=`QurI(4zL#Fz5gaDtex?!<~rXVT&zSKWQEnfPtu{ zlU5Z^ASGk&4<8}o?|sHOFW49l3v?zrYB3pTNj41_k{z4*0N4L=v`>r%^~ua`UaerHDvx~m!RjR_g_%32xbDO& z{?Gb}bidE0+bY7w&9I#Yaucv>D>K3@OUd_zI&PMTvB$o_(QexOc;uxh%o&I4MXj#> zv%DoHk&`i_2Y{5Yk5^!Wf=h!%U(rSJ9_n*`S)J5#H1?l-45(9UWBzv68E7g$6>O~k z8s=04MSLQgaI|}z=nd4!Fi)Qv9z049-TQvryzz)CA1o^3#dF$MPNLdbXK+ibf{BK+ zPt)rY(DL)#%&fNNE6?ahn_*Qo4elRf{=6()re5C0-zK2udQqk1bgAd40r13PVI^`$5bYX$?(W|?Rn#ZC0 za91?vNiEdom8yR#k7BohiQm>9tg=Q_@3)s!Ox)5-08;hYk7+-(VY#&F{>z|`>td^r69gzvcjl|#QBMn@k> zzNNu%&Alcr-(KYiVCSCjl_Jg(3XTsEV#HMUe!zH>S1l^>mh85uT!BvGOXWErcQ_tA z_-q?A_j`5fSz&g#ZUxV@KZX0CE>xeoj-o77A1+)wlzxxc+O(m}iPXB=_d#O%&`)D-A>?ja-U-YkjKf!lj)*8pmQj_SU zP>Mn&tA`Z2u>2++Z0!;!DzG?;W~&2z&-A&`sCVqM3Um*Mbq@^rE)SrMxlm?Xd5j-y z>6A+--PkbnI*bInT2b~jns;^xm8!x2Yu=&O$Cbmi3**aE+Z?WDDixGMUr-pSMy2)@d$DdFjwnwRcAM$(QL$EdNAK!|wFGQWU=p zD)N)l6ziq<>k#>Fa`@L83a%J6obKh9b=Lc5{c9hk75OzD;h9Pm&VIJM4QV&`?K+KD z#?yY^&p9c7>i_*7o+jdj?Q{td^@%3D8pD;~Y0N}kFMbk7lj?g!s+|1rtnWVzv)A51 zL*qA8Dre?V?{QI@Kfj(qGu@5Mq#yqgm&U97natn zZ;Ouug*Oz&y{qLm3ErR@pZa^BTD)W_c1=9J*U;+fJ!L$d(lcD5VKAPlIrG6(@MBoo zG7Ila|06HYySu?8J&fOJl^DCXxt1QfZk6@2?!fpa)?X?=vJQ`c1;u-GR!LNTQB-wX zYedK9$-l;z>iycpo67P*d1$s>JpSJ9CRmpZpAUbc0s~2MtFhm|L$`~R+?He6(C7O5 zR1W<}j$%_xqgx!e9M2 zvy6R75)EqTib_2{1uX$c;S%97EZPNhu@!7kwUa0^Y~oZj07L9Z_o}Kc+K?u)KI4!< zCGMh~w8!b7(k;Lh9XtZn@^Lqd;exrtsZ7lWwb6pWC0%D&%>-(SMO+FjU z9p#sThP8<<8=ojVPN>a-etNqi#oq-3p@%<8uh;&woGL>a)i=vNf7_e&spth&oE4?q zpLRjp!kiIW4Zdt4!a~6*L8--7r-wn5JR67{Gb?46aSUq!=MNMybI#B!Z z!>=lwqF~tK;a9xwU6_+z-tFaz@ve5|NWawj_7ah9PO@akiu|iRINF@^^}#mC&S!k~@%jInSIT!Yx0;(@QcSA*Pmm!sg5g(0U7T9^aE1W7kg6YtZg%fbICu|hJWjj z{D0eStlxoER-XlF5Q$iMVRp0zC7s=*F8q$l?uPA`Y`5`y2W4TnDWzY}wh4Mf?Hk%1 zW1!%q0C~qkEcD$eW;-^26A1nMex@^6{=zj{#G6=j26elQi$In=3^Po)CZxUm@AC1i z_s%=MQ>BXgry7TdFMD*mc6nj@?wlemt;!3SQH$ZO-qHO3xc|+AbYCYACPU|oO0m@* zOwZ`mjJ1~;vVz)qD+k)p0W^K_Pka1$2y~y>^X=MS`IJB#t5*{A&+`R(uhS>Qd^dom zoLhGX-S@zJ3Ej#F6P7Qcic#DwDeRx)OleccKUz5KiZqwxIKm1 z?p|J`S9D~0HzI$~=iQ?g4kJ%v$tR*PoF@6ccf;hN@35ACJDi{UClJ2H^LCn)0r|cy zUF4G)SUE`dQFY59STxI!n`2{w+`@X#xYaFaOk*$I?Ztg)lFj*x8phv@r*r?xo_^yT zw9AcFM}=Yc>TrM0X}v-ns(Gimaw^d%AUoLQO56PWcHfRQ(v8Ty5`{xd_e9WBCnb7=QCJ zS@h26Nq1DgWo_}JFG6dKakE#y9)cBNwWc~PEbk}H`;xh51yuI>fA+H_x{2qx?VW%9 z4x~R%a3(MPVD0c|d?Sx0j4W6AJ(8b>*`p`u^*jzyl{c35N&fziTQH^U%lj?uH4HWF zu<@4^fGf1oTgeWE1fq_J*z>QxNl|_$Nd^QL2olhM+LxKmx+T>0W9zGQ+TDn7 z_ZXiD&mJ^VdMp0TS$b@|?q1k(`ynGF&PM(Y@z6rO@&>h6OSj<|wl3{Zi_t>uH|E0@ zPhtFgaZR^wIYg#GcW~G6a}5kPXKm;rc+dy8n=R%I)1S743a09Nj^y zyA$Q;I<}!%ZDT%`!4{Y@=6(5f!~&WR%QVuJj3SYuc~F#dHt9(#0}*%RbaHtn31k$0Tx-bdY|{6hvoAR zonMS>fyMP@p|x+bX!=QMz^?`Kzj}%oH*Xmu%qhK?i%lqm;Ta#*Q8s(DQ8wJCa3ABH z?|JjXXI85gb&m0$6I)Y*)Cgv0(orhjUFfXaUp1V(4>flU_#5pX#nZkyv~T8_4fJ_- z^gDG1Lz|Dfa3IMaO>%CwI%D|<#t)d;s9N>I!10X>b9=W#?WN4O9EmZA%yS}SDp45- zBQxLQeyt$Vfyb&%SM1TOap_TB)k{FQl)q23;s_$|xi28L9*4%eVsDe2pTnwNn&wB= zFf>CNZHjwh48sz6`aAq>U~W&(XVyAAj2Rj7zI*o-dh=*Ze(xlq{uke_^s^|U!H71xSDN@#2KL)f|iSVAK|>zdM|W zJZx!QGubXehXCtW+U&EinA~_l!MhMD%zDTF;4oe_^66vuhIMODQ|yz*gFFgobRl=( zd}}_AR)!ZoY|B9jO9qRhO(H=Z)naQ9o0F05pCpW zXw!cjM?*Q!J1eyt5rNw`m6I3(6G|yjy8Re0B;m#Dn<_0-{x{V3TT@n0`ysa8Kb;FH z8Y)7h;I>n2c^Kb)pSsi=!^XEj`l&r7_l6x6?tDX7IrEjOokTj`y>Aa-IMNk^zUzxa zf9I_x)`QGRk$(`8J~YTgfU2MWynn`geA9RLl_A2$oJgv$7tH-UZ7F>J0-AIxagq+A zL(^t|_IC(%!P>q3!3R?T)|v-w6R%=CWkl0oR?{_q;I;GYPNO$8Fu=~~?{0wc2rWr_ z#7RVBc!U;xp$+;ck!%yk4d{uLw3{}2h{3d9wc}Nt3 zNr=Q0L7Fh{L-RA@kN5vrg(XoUf9M_T{%LR8lXDGsl2T4T>-x?|w<4gHOQ*^7sVgkG z3z4|KVR->7?L64G2FAkxm-rXjml$s(^2?ZCeRvAhq%>Hor*22}9x8LJ*RXt1l`|22 z+h3=`w5QUh)=Yb}Dnmw zDUPL)W-KOU7feKUDSAZQ!p;?1FMm4=K(z&j?|5o_Lu8J@qT6ai$TPwsHS*C-XsDx` z%@2}AwXdhzEaVRUbACrvhn}|6OrjyCv>xS812En2x=_cJ;x|XL_;<(jgpFYENy9~x z9*W-+M<-(?FnC-95z=t&d+%RGJw)r0eWJ85HpQlU;yFc*K=Uv9_pv)+_fOaT>xI@y z?r4U8cjxoGH)uSaVb@M^4wOO&6)x}UMTDO(&Y$nZ@N{dxxlD|uBmc@P%CCMXI>jfA)9&9$L!b*Yzi2`f%m1=Wixb^U-jWBY*cDDtj%S>G|y~Jv3Jk z!K^*|h#K|z&|ls+u>Vibe?Nz%91c#6O3bg)`SvQRx{292*5ti6TlzCXbKw%a1*`O9iAa>g>N=jh+zRV%mnQZyQu(u?o@kMFT^f^r zr-SN(LYBEu^Wm(wg=7&N&1>@uF9Uv|k*1;t@ApvbWO#;iyH8&grt%+Al|Lz~ze{x_Yj{l((8a7(!ln?j_b+Q5+!Zxp<@pT8wz)x&{(i!>L z9}2BN$=?%}g+}jF*L#0lE0_JB>kH-sNz&85Xp-EOO<8RUr_^vd&Hm`yu;W$M7U?soaM1;RGHU8tCXa zDUnEk$q z?gni>)VhAa4_4iT^%Jevd{|teP`7qDdhscAJs+qtyN21xiR--dBg`28`#KYu?qn4f zuik!Xd(!c)G?>V9dKngXA9_t@d}D2~{JDY6^qni}xoAlAPd|(8E*Onn3N08`K;zDe zmjlAzqppr+PN_06G`My|z3?H1|HwU_bBw+g<7r$p+Yp#)lA!7r@j(rZ_@-SD`s)=Q z7yk?;`wm~s%QJv^u1ZL{sQOpFNK%aOrUAmsF{5Q(YslBTG$R&Vw*9n%=1DG9m{<`6MTGoQS+(YSLx`f z@(WRoQ|H&{7=0%je06ub!{^$h=n zmmkqg?}vofrWkKu{d|svbLkZ*JN{5fe+t8I|H_whdS8DQoYc%4*hC3v>0t@KdNEag z94KAfvv%{-UDPhdA75&nk5)dJS?DxZqk(rPGa`78P~IcSL@bpKyQAFDOYHKN6nh|= zzs|lYtF;eCl>|AGeYw&6pQsgWo>Z8N&fFE_a}T*c(*CgJ&>1v;YUiDAXRo8l9Rku# zm18)@0*(fs#%LVnj!KR^pFp#1Y}|(f3!v}j^OG0kIiPTV zq=Q!VpMQ?;DACP*sj(49zal4CytM)52oK&3pGt=HnG+dnWnBOLK91S^u2VW@zriu* z6bii?{0W1HbTXWDDfwt%EY4(XnQNJ9-3%5+`4{-S!p0}X^K#71b{q?1r=oin_y1fE zZOxzRbA4K1vG27-{)r|W?Yr$6X`W>0IKBTouVo9Y_!qI+36w&U%YhD$t_RR$y&mc< zc?f13&wdY*2*lQZGF7_?LG27D zvu{IVL0PqW`B@kTbrWsIUohWV`DmAi1dd7T$nzSeY^w35)!2{?>H44fZ(soLFxeaBGO#FrFVpJSP59`)@1@kfEZ7j$OZ{~m%Re~lX`}Vx$PDxkE*GW6 zQ21A2=6v@3lTv^6>gJWbMmktLBl)n%y7kusQ2R|>`EA`Bn700Ww|jpO#q1A-g}HUdD=~hoMF-Wy9TS*6ikS4G)g#giC8P&9%50|c zBT(L(<8)=)9ri!%qGf1b5NA-M3#_33dJSG1O3^HYXZl5HfK28Zqcb> zmq^UMnK)#j@GvqBh;PSUUgW^+7p0wn4CS|%Vd+P~xkl@EINEZuzUoKToGlTUOlF2Pqvu!`P;WuA?_iGE;nBnX4{)NMUGQ=b74YcC@(+19P9U$ zM+Lu%)nLwJMo*M(KP(J&_r#lu0{PHe_fy|4;+UT_u8gbY!-Fg7oAxIVP)oj0MSjf* zAZKdZhv*Mb&;Pd@Q+6U0{(k5YyD0Y2QQ?1|7nPrtu6gpK;ohU%ofFhN+r+O2ceo~D zaiOc#O6>K@J5V|I_qS{JLt(D60KaGJ8R)E{5y3}?z_5R{?IUXk>Utt=GhWVPaSG+K zAHydOlF{U(h18WNC(-bogwtz-)cDZVRFCeSJNy624=p>+c4t8=uV0>!5N5|1a+_#; zq>SOP+}{{SeRwkogzcPMV! ziL1~q9AQvzV-H#G$4|v+Tt*Wj%S^jR5Uec8Xr5PMg|QTkDLMvmn7SUupaEE%&up;Y zCZpO09DT;UTEWn8nB=h`uGT8R+=@hf#6~v|TXorgl~LgvvH8}O7wUS%jb@g@myISU z=Ph`7kDOFvp*NVTO?{t@9{Q>DKK~>2eHAP*&hT>`#{Rsr zd~o}3llK<L{MBh#9B->;ts9iQ9hYZX5d#UFpMAc5F`Af|Yq_OjMZbVX} za_J4!+^`-NQ=5aC>;swCOHZP{C*;SYsqg+-{?O`7<0HH4s84)wJ7Yd(mt7B$C}cO) zfT_;hU4S4z+2z*``eO3m-o~S>g)@KmX(D^+5XGMa6O7U2vERgCl7%aLrwtzRWW81{9@Ix8 z=0iIY9^hf#tawSrlO1J6N(rWVV{&vYx4qY_Glqk&v~;$^wNh|Tu&}AQheiAsknGye zYRY5n9^owhctw#9H4TuSouR?>pV_&1SB=5vXx#9pkFqz0_sqA=AV23jgK7zQ?$c(R z(C@OvG;yX3TTf#1;tl$kJ)IQXs)XO>3yYa%!OnK+s9oJwSox;UKgXZMw=|z@8Uju2 z_lzH8zQobgU9pTyGyT{5RO^;bCFj-APn&@h#Ge|)g7H)lIX&?qh8XYGXfE%`i=Qa> z1yt!zbMT69f?1^<$@>=mc|XAF_r3la5zDanw7%9zRUQTDk5(r?r^4~T==Av4d&_%Z zie=@%AWs7{h6Km&ztsnIAI#zxLqc(Mr`rS)hAH|3j`5xI0i!a^e%R~G5j}Qy8b`O^ zEI=>v0SwSQ?z1q&?D{M5e7yHxo`z-4)7h2LnrKb4do)o38wVN|f2M;68KHmSf^}K@ z7pS(ncwmP^1!~T&fB*dsB@Ph!cMB%6&mG6nxo$)b^pvCYQ2txdY!A@yH+_y}p+!)p zcQV%G(J5G6PbraLorR2Z+n2>@DZBt^wp}6J>MRo~^}5ztcyAKNFqcMK8;;>9rt+5$ zO|D*qg%|DD$8Z=?kIusE{;d8@dw zgV}%eBs%GZ^OXAyX4Dr=CJ()acFri}og8$ig{j?^T!`t@Bm3*4()EZqdQO&Y9fsRs zwDhO=OPo5YdZvD!@ccVkD*mq3bx#mx8N6CjrKHiMmYfOA)Gj<@jzb`W=58Qm^m8hb zC^#`R|1@kd%~KN3&{C1`<>EFNy_V;>^56{26_Z+sCqJS^&fntc2ZT`P7|r*UMGyG& z<%R#DKf)-i*TU<5AC~Wsxqs)AD$yhy4VzkH&BqU@jMO$Q@-ZBS_7Jr-kDi7=|LKCp z)qLo2c(Jc~HVqS2ZeP{bJ^Gkmb=n}Ui2GU9|Jrvj9<=db3s(S2 ziyC&>FM9-9Z+X%_bJ~FlHgM8*b01)|=Bd|}mv3Qd);ec@>tjd`$>iz^j6=ikx4k>$ zUk;NmlH$q_(ZQhInNE2-Y~K>Hvxa5&-9p2B=hw4eVfJ47M75p8dfiYXW5_4BAPbYP zA`~WnoInzk9vzO(6R&G<>Ego%YJ#^}KU>^w>cvJU-fB-=RKw!2!yG zde&6>15qDysGr;N9M-=hd#$A!Zo{mTy!?+547WSgAh#>Cl?_{O>%N`LmRzu2_Cf4S zOeV|}IQ7%oF+kvXu{RmPji@7IrGs-j<(x*tH-&brKBczD6CRm{J+{Jd=VsX^+eFR> z0-4CT`g8Uk#2QQSsDxpBq8(ukxxz&hyBdz|`NPs(dBHjsYU%QqGrzq@gVHXR%Ysd?jDNA& zswy4Eo(FoAc4K+KdRO+(l<(G}{NLHalGDA|4`qMkc1K2BMtxr;J7P+y?VO`RpK=q0 zsq!cNa~%(Oha53L%rG|H&slez9gU#)?S3XC`e=&r z1b5!xwfISY5&h^$)Y9QQ4C{?se~7VR_JHxqLa|~qHq>)pd-|+`Hjusrv3KT{P=0T5 zIvIBQoifx^-y87$=m_K52g(lO2Vj1;<{AEbX)wR=+j-T74Tk04-Dvb+N9EsMEz0P2 z0XdI1+2$ZcZo|{=pEP57XO5%gDvFr89Etipbi-2%-=pfX39B1Y)OlIwW*gS4q-9{? zp@TFdYakE<$MUc1til?bjrCM94XV2>R&M7*#RH6^J&jATc_@TlJ*?3#%*FQo=rc*r zX}=-VYU#dJN`xZ!XHuC2E(AQGvp>_rHPmUa?u`f8h z1=a^=mSfWGQ1Ogf*uH#i)b;Jo+I%n@Ea6`C=(|NiYt`qQ#~pT}q2W2(qLHWnbA1e{ zD=Z24#lVoDWK!Qa51z$mP+Zu_5Y?)E4JhxZ!m&gTSWYx`;pmcfi;a>auz4MkYAdvq3{&cq#B+t-`s+u zd2#=Ak{vZ}doKC*jE_nqOq*#`n19Fg=ZtIv&a|x*oGdIc_$an1V*GWj*AEqxDpUD| zq30Wa^xk_T(3h_tr)PZ(CDm`rD%GU;$Nrw9lY~oO%AmJW>FK#8YeZ6jbg7v1Ab2=Y@F|-hAT0bTxBWwLvZwg z!O7m1_h8_&1_#`F3cWwxK4DM3h$?MmiXX@Yp}EY5dz{B9{M3kWPeOQ~Bnu4jPbU2t zx{c~*d_~bEqMJRkQJd`Hp!stcx>MdRK34C zY!iuQz0zeGV>)2?d%3UMIv32nIdri{5woukZO{ss=DNXhVu2`5o+^(Nj!yWjXGG5j zs&z7)wk~u=mIEfzG?i`dr$b5F&$>a@5j?Zf6{+{9lwf*$>hH4(InZ7%bn3;}2lUHj zUvvgfIHle{D+O)q(=0H+{4S;74@C|_(~oYm?RFQ(v*cLk4Ya-g?|Shxn#wYHcQ89A zfu`^5h0~PyPpH1dTvA|39Y?vWIV#xq0kb1?Wq-55k-dPtZNS@27|SD4`+mRe=`Ie) z(DC`Rc7!sXct*>v>BAxzuR`YzHJbH{g)nqL(52kW2HK``japMEaSO1ZR-d!(L&b9h zi?Ljbp!$y`O4B zgh0Osndh&FCeG=GBFNy}a^FO$yLINA)vLz_!=D?6b>p*NuYYZQ(#cH5h<&sQKTx}@TZ_D39z z)iHsBNf%hNH+qyTfZ?ASHq+Pzo}}c9L~~yXqyv9UL&M?b&5}x4FxS6Wm3vMe6}b?5 zDhBRBU|LxM^DqUchvxg+to_q$(5%!m1N7?=Y`DjA-`J1E(@h?K%n%fD96AO49{58) z^e^jd4SeeW3wNGM3^Sgmn#cT4%l-Bbm=JMfEmBqd3rsRQudVoDev%eJRJ!{I7iyM1 zzimfkKcyWLwCkcx*gA)?2KtEd~!b>K;`ULF07^Gz% z^x7UsRo~&Fb!9{?2t= z)~bn@C?Y70em!yb2uxgbyD{smhALw1XgQ6s@oV~IJoro}0Ehu1V~pav&`9%TdG zKwG5gI*sR{Dd=*gUjxQ_+vUDo$sRur#4&zkV3i3=$4B37W|)Hxmft@Y8dqV4Kjqn1 zIxL?~-d*eImu`7zQKo^|HmZqA4cIno?!fYYj6??DFMTk9p^#UB!+FnP`E}Vf$4Ux* z7S(e{?yVKM0ULp5eFZYR?F;{No_=iCv{moqMG-YUGRgZhQ7>)0?g?MaUJ*qsJY(_wJ`BqITHH{NLiqHQ z>ajj17|Ad5dvK%@jc~fY9NPQ=b^Y#)8`U`j{rUvfC~Z9$U_4F-UKs9t+#_i zPL17yvDijVHJN4@x7oxUedsMLdZ-zNpS6dDLepoJYg%}g;U7U2sj*P45xGP5q$Tv) z#aeAd83Sqk7@htw1z&(BLuk$O^l_S z4XVV(zfk<1Ow%VS{Scae-}`gPFzNs4ex$l@S7`}X!d?H#*YVHqW>lRec1%#6>(E`T zqR4Tm8)n*PHMp(aP``XQofQYAeW=O}M|n&d;G%o`93`EL7!s^KW(bpG(bH|WK5&16i7%23 zin$M|zWbAXr|HLQIaJr~%@h

  • LGX^kO%sB2+DNPddw4zy#g9dlGX9&F*-}xg;5a zdROl?dA2j4(qLby(OEsT8fJ86KOc53Cneg_Z#%nya-ugjeSEh8KM#;XxpE%^Np5oZ zl8`B+eLi&g{pdgWl>dL8_Z88{mY>}y@lvSr&M!goag67^y++mGnjsZ`2<81oAHD;( z?pwluxy8$YN-i{dwdh9WdMRq#q7N)<*%)4u70?B9RjsWW$q6C#?j1>Esp(i zLzc2kF{p+EdWx=88^xzW8GGFhhI7SGO^CY<^6AUVo)XmO`p%eE4u=-XbxmfwqY!z=hR7{OO8#jy)V2A;^EG2czFi+?E^;5w!X~Lp zin@m8B#77JtKJEc>yd#vGXC zG`&SPh!Dxf>!N9*B>KcqdbZ8S6%kjmjP~?Jq1p-e-XjM7c$)b+uI)4&F!zd&#p9kl zkQ2XjofIpBzUv`wZNeDuOLEFd*OZD3=$6$K`W~+by%OQ~9;;ArsVK+s{7X`K3Thq> zdb)a}7j+8w2h%Gdlr(;cJQueNB+eiFH!G|0G^;)bWy2O=f?rr_oCnLpTQAvnV6r|6 zN8=giN`K)iDj#C9e0cE&O5MTv)B+Kx=9N72jF-W~0u60_Af1)!i5hlY-1BW*57qj( zhYpQrKwsST8^ie+e?xJThF4>lIn-T~v$EU01wx3or}pt5f`P6V>7+)0GO56Y7gjCk z7qiJ7wsc>VdddGlnR+qmd4)6es$D`|O|ut{(}|<;ie2v-M9QJM8JBQOA;YOvaUeEL;e?w27 zMcIKmcT{fq6W-jHfDS|XQ(jS}*!M&6 z`A#3%V3DJ)Bc+wv&fIHxSzMmlPXnauT!w^F>icWueVWaOs(-Gw3+{7kf4qMee}H?E~Z8;{xb2F<;wyl#gd*8_SO z>3Fnvll35UF3`w}&3K?$mwjDEdy}9{$DHI+L;_Ofo5)8O4kNO}$>Lq3f~dSSd&)6@ zACX@_=k)D31CxH9f;aT_(TL6M(XucP^r`K-#_omxY47~+zj&J6962p6R5+ynIgVtH z(W4iPCt$Sm?DyxF63}um?IqQDPG}vP>0~s@h7J3tq3Y@v(5jP_Z=cLt80sJSsqtG2 z4en@rc8L$eoe}xC`Wb#w&U-Xxp?Rt@w3})@)NSrlGA&Ml#_b8miw2qS3@ZWl2I3TZ zH=g!_`&cBC02JU}aHMR;_=ZSM2BAEf|IA|u!yD@#qt??=GqB7zt{g`Tms?&Fwh#B9ji|OX(LsM&JGV9gtpk_%^#yaDq4n zyN7Y6hEAyZzz})h#yuePJTRNMR0-r40~2d;{W$u}ydl5RvVV?qrBJ8LVV)YQ=Y#+U zbKM~dJ{ApFIlncFpxC|82#rgKmkt$g0nAp}o+(_y_@9d2e~_I&qK#+t-uM0HQ!GxR zYp;pQ#zr31_e6(t7uNczcyds0|NIWEV|$^=ua$Xkjs@zwB7kGIVnG{5)*p+suzX@& zWkbKqBA1|4wOMSa_X8?pudX2PW2} z3MGQ#c3Zhlpo$~kCmnM*Q24bkE-ZOCR6JR5kKZf?^)amNw!UEbSHHPc5n(n4c`rZk zD;e)aJ@;X5^RP1dl<&JCo01RRb4>4iONyZBnzv@gpExvG#3rxZl=Sz+Z{PmQbpchi zN1QdTID==9G#czD;1NOZo2-gY2mItyi;uJ5MjKZ{>=jZPk)Kq933B<3h*f-~vieQX z#ZO<*KOq7?c%MjBlCZeK=F%StKG-_Adl$Y-*dN-kNQut z`sJ5s?Y)Z+v3d~A_NR|6?6*fn>2c8&-tDlQe>T^vm=&5vN;&-M=}_HAS=XwYs;CyJ zwkGYag84lYPtS85LgX2-tVUl!^z+r}2iL>7(HILOSE%Y?Xg=1b^V2XM2&+%oO`rEc zaAUkK@^(bCGv^8!Dh}q}YQDC=VU7BP90r>w zo&!-Ma3j;`BXr&`vmJjrk7i%KjwZjj1EZ}xe@q>M@GK=^UpTo>0{Q0FcLEPT!1Th7 zrz8C&CrIMF|qgPM>MXCe{i#>5!&MSSp|GgKb=q>@E#kR2mdCBR zNmx;E{1TqVm8;{acm$9?9QH{5dLE4oMK4<3R|dk|)?dztsp8C_Fl^K2&1ZsO!0Jh< zmM50qiR83o_cEUh_1^{FH06DQHJK)zi?X&bNH@MDi?{i^UYP8)P4A%`y8WWlPk+OL zqjhe~NI9(T+`U=0Hv@_v=MS(n{)R*wW+lnDLTE}^tU`lN166blLFi<=TZWCp2i@}s51oxoywf$FYkFVw%K!TR>c1z4*n z5O&de`u86B<;1in8fNM1H=UjGM!nZ?QKRgbU3Z{3%<)&kF<5$9&ho?B5ArT(d7uW2 zzp??pzu)CXHW1%=o_%z(fognyibWTHd}0fexX{d`J%i9_@G;VT)4$`X7CpQ9hQt6qN0*{Dr64-_Lkq(<40dc=YTd-DPNL zdGuAo?gI4R54F5d5(?8z5^@4jZ}7ClrxOxoRB{oe+@%VOxx?#4Xj)fyoI4K7vsSOi z(h?=Q1y#1m_3I`6e^^J`v$xvp?1G7$u~nY`!voqEza+Kx8wL=*#wPWzTo``u$nPUT zD9Nkv3;h8qUZj7nzn)l;F|~Vxux286qkx795iXaBBm`4%)2LVanE4GFEPi)-*Ao@b zqrXvash6{A|3m0!RJzF_`5d+EU2c#J4}tvJqH{9}6;$;s1qN>#RKxsNA*Q;lb=ADQ7!0Ct`Z{*aHZ6DLDo zy2B1#ZwihSO^*LIy89s+mJD_~rSh`FNJ8t^9RF`vIrTaXIc~Z*x*aA3{aUtYanrYU zo)}E;C!MuF6LM7_M_;%R&l!pN4@p;!{M5L06&7A@9g8_d*>8{&{QA%3q#)|{Rz2BF zTSejDM2Y;l1|w4Qs4l7>w<$0WO+_-_zn9#Drj9mCZC<&HCclrZ9CEirE$SSfofy?% zL%Pbxa2Li`GT0PovUAfvdFr71qNaj=Ukx;Tt&?Q-H%FzRysw#Rw?OKvz~><%E(p(0 ze}ep19-S)v8Fo(352CobXpRPYpkcav>0iet(6Y3x<%@cZr=?Y@wC-EcY51*D9&(iq z!=G*_N4+_=AM=YZtR6gV^Ccb;vhow3XE&pg<|{3Y4jO3SVOB|Je>m#$|xWmM`58fX1 z*m`NTJO5PgnjJI~nRb6;#CUw``pd(G!rM@-X4aDQax@S^bm@7EzruQn$mttuc$i?$ z%e+K)3tAX2AD*L=hengFH2e+nQ0xC7Warp^GfX#bvG_Dkq{W8k~I%yr*tUhD@ zhQ)W=`hLe?xP{)z>x$8L9AWA4=h@^hY-nhuP^0k&J52m~+exnHfU-B^?-3n_w;aN! zosE;1MTtEk4#9j#1 z-^@s$JBsOviws#J_^%Ya1D@fAiP|5f&4|cw<(Psf4mBP;k+nbeHySh3`BHkt0$Puz z8!#Te0&UtcW52s-(767=rvl%!pxWT`2|eMvu(A@If3<=PJ>sJUI=*c%m1odhakd=W z$A?<^rcAb=u>k=`w*#z*a5O?C&3+dmR#bBi#qyxhMwYDLij#;aXxM)@GX*+B44BQ% zOrnv*u|GzppF1$pm*~{gqU3__gtmXP~+7F^YdX4zg>K% z+Ts!*_0~C#sd*SS3p`49WD5{N&+SIWMlg+!mv8?Xh5DagqSLle!_%40T31Yrqw%sZ zYZvPsXzrVvK$YYWw8c(1_lkW%WwzQ*H#@JxutvVbwTG1Z7WD=7r0%6%2eNEj)d`ji zi1v^E$s_O<&%DP{g!`2>id7*SFz3WkolnD7D+yY;e5ijpfrnis0OD`UnUF7Ha(O2g zUW8Nf8=CZdV#4hA30uecQhNe^xTCf@4d!cKA3zl|Da9hh6werWe8zl1291PnFePs= zp@dM z58A>qSMGL}=#Q}R=mv@7b_H~E%738^uS3gs-_mTaTZJ@j#eGtN_t4A=Q&4IPX2-7G zb|gr2^f-KyJ>E%<$U>FAs(*0rTVS2d-sXXD72sJ&-ZDZ}F#CMu3`IFk2znv{_gC?^dk+HHM&-{LOCuP^d*F_?Ccy}PYUlBc?`Y=c&ys{h zR+!!5s8u?J;lBBUtP(m?E=MqP_i-gh#}bs6)GRj3^B`rL?Ei{eMYvvgag92Wh5QTN~RScdQaIBru$C`Bn6NTq?K!a>ncX;2yM zBr6#yp+Oobkx?ljWo2)|aogE@@4d%;+xy$+kKcX09`F0xpZXP zJkR4ep2GtBW;cvPob7}T^TAId|HmJZpXK7mgMZt}RKvZ$3b69a#^NvQdPsCv)bzMF z?#Dmd5B)i|xC6($awK1opDHvpHy>8bM11<0B53~>eSG^9>inpDCbZ5_zSSzbO zMn4qA{+c{4Ro(MABGPeqXs0!ab*=iBJ}V0dG3{>M*-eHn+h;e(_X!@)jK!WaJLC7k zM84mj0(6 zGqrzec)n`Otf9d<{@?zKv&+q8p0nOC7V3;)Zy)2>?`vk^3uiF%z_n3}B}VAhxxuQi zekDx&Y2vl*rO}CilG{N&yt{N^WcFdn8l>f))m*opu;QfA_k`9b$z_S^R9_{m{8f}y zz|n$jD&h_gSPA@5$T@lM%9=TxOLfe8_$U?1>=eaT&;N$@P?cR#Z>YRh%+qz1_3I{l z*5`KkFB!b@gg?G4CE1^-aVD0v$$wdKTn4&hLRC}GQ0qj{X2f#Qwu;D;7cnoYI%y) zdlPDvKd^QQ?8lzdzBA)}ia5!6nd!v6dB_#Ky!Qs=VQ<5}>xG}c;GF1Wikv_nX{EAY z@Urlgf4yfUrVYo9PYZ6LK6eetamf1@)i(=U3m7 z%xz(=NyD)z_oppTg|SQkT(F7MPfQnXd4X z4qDbp-Z8l)1al|fecj$lTQAJZ@UK%i9SyZ<*}|euve2J@$2sj7p)XgTyXd3$rWpH= z4>;-F{tAPK(o0!V3Eb*ngY*{TW-8wkqocP*r(Je|DY*!W%BU>NirP6n%B4`Rm)*}f z#hE<@BXPSf{<``SYTJ3znhR+7PDFeZr84BJxEH5J-WDE}kcI*4KSwM+JOhfWtqjY1 zXPE0T-v5&m{&ha|g`IQ#!TfK1vAU#Xr;~$xnE%1#FaasW!`=sv^@WLe4EbN>i0uz(7%ymPs(kOowU3@9(U;4rnNoH^+ms4pOJ4~r z{YZ}EUBQcOd`|Aw;rWpIF?*QJLjftHytd)h##kNfZimfcP{Lt&^_2mSXM`(Y!t3Io{00(a0_(tDr3H@JKB?Q}ikYf2& z`!iE zm+r{VJ)O5gByhJ;*}Z2jil6W#l_y;>R@j;`va z-Y1i@Bf|0XO&F%D?n`9Yh!tZlH!cU|Lakwc;|Mc#e}Ju)X+`WURQ*#JSamhCe1V$( zMURxaTH*D(8K*PtQf#_>NCWYvH_ z>zBQRlATG+oY~JNPXt4h)2zu%`~)^07_jcNaD%qb`ZKc!T1a#u25xf#m#Ek8-CuUf zdS)lis=fTcu9S)GijB{PPQ{Vv&e{d%yBA_F+o~xe#sbXa`FKpvJP3Y0iMjat(pSt# zy){$wC=2RT# z?L@q|P_&N#Hn^9l^>2R+^BXr0GQ3s*vcjfQaF4n#R-`kIf8|QYgA+nIM<0&OKxI*Q zQlkP5kB0Vs>*B|qv{JCIcy?vg7n;03iM3z%aLLJR?2_i=k=*|T8xoGH1{o4Q!;2p? z6M=;mA5(qWbmrI;jjrTB+mY1KSbzLuBK&z5`%*f!6sx?_>Czqc;JEK3Zm))5;{gTy1CXfmNa}g47=p~y(%mp>#Fx{ zkuId_sYAcgRqugI=@3JvODD%X$H{Je?H*JDN-OtSx$Xue>saZes}gmV+PNA3E7$Gu zwP|f*_*o9fWZ!;Wpy)Pas;9&h$hJVxc~QS|>2p|>nkKmQ(P8N4G0eD{_8l5Cmeua; zn8v)k&VkiwU0B9sP$Toh2-Ccy)I^_lA=&VFDc|5ZXq#BS_4nvb@N^Q}dH&=g3`^WD zZlG6yTI*xY7Zwbl??lDM13ukA+2WJyw_Faj=xkTL+Y{PX>`Fh-auB<9c)hgD#&GQ0 zhIG2OSFyoZC^RnF4k*$+FE+gSyU&=|tJSoMgmDLyWk0;|!OA5qJ~e*>EPm~CiaUNS zOzd>_+tlj>k^BdDSZG^7R$5Q~d0h!C5d5{c~*`~mw<}~XWwmuVGcpr#ZXdZEFU&7PZB!C{4*1BBTDsKW+8{VCkKd=p!EI*8li_&3-ll8HypRPlX zH04pD-nPHzZ_Mbl|Bjv9dEWz>f5K?@{!ITp>@X(HyzaAa6STxdm5~2@gT_y5JvH`= z!oWxiFSGLrm_7PDVKa%E&%xnu_Uxt%8u(|wPz&o#0$(=vnVo@to1`M0koCiuY{K`b z=DLnbjt^Cj0eh8qemb%D5mxx%y5m+KF|M?uQB0ER4~5ff9~(U@P9ct;|GnzM&9}J7 zr*uN%%0}$s(3PD_kB9NSt2?4zm*IfZw$}%FrZLz1g2cw&|B7G7QbvQa3*%&%eJ^z= zg!P|#e7Imf9n9xI)N#9p_U^dAVFANY9R9BdXni~{&Dld@4?Afm)zd;%(ChNZ(=&?ac-NLCopLAjA9ZSo?`DLPi>3<8kni2Xa<6nbo z<{oLm=ngsNpYjzjtM|B#Yuxnj_3p9rRfxa@Ys&BA!p4}l`m$5_r2?#t=qd9TK8*cL zH+W|aYG6=D=f1!qm2UxU=Dy++y5~rYDY_R{-kJGt{YYZC5EqneG_7 z{*DDjN>2*benZh z{Xgm!J;!^`1k5$UiW$ z<;HP|V-nDrzH`02p(-?XX_?i}{eqIHVdbGVQS9ime~@Jo3iYgahldZ!LU;J3<3_ghYs99a-aJC$t*VP z2&}wO9!ZUZNpuapvyY}Zp!^8)jTg-MI-_uU1Q zilZO(Og{sqxY2ZDP#cbYSL8SKeh7n|K3i>M`LWgFK*Zv8YCZ->;k$}^ln1tsU&@;j z8^joC&4x-gqMkXn#4quy*AU1oKU(dMuEq+L#Z$Jm;m{cQ1~l3UeBi)~T*rh9laOM5 z=yttUIdo}H@Hc7=!L)DY)@N(J;8?ap_NVPM`7>ztcE8X4BLyjMUt5a&)WkW>r8Spp zsJbvPF|_LryVPZ*gv31auyBPsFWFK1)!T3;$5eNOegraR1zL|JKZR*d8HMPrqa=37 zoX_K@)^~Bj(v{v+F$+hHKVNc9*1^F`H7>zv%Os}biK?Nkf%xlDO#Pu=R}#aGqDwjN z1)*+&e|u}YG1jc=J6+^O0#}p!xq?ip&|~L*?@=>NT#Ql9A>q84udvIsVZ}YE2_*ly zq1@7x15F(8p~mnq&Y7KTEEN#@uk%55+%%ihl|MAUx22yA+_yUh!;Z-fiX^aLMgMTD z37KZUfxoQ}oeDl~k3;*eh}?fJgcL`Ss9nX^|`&dS!+dDt@v@% zx=siF*i%X9*tY)M*s*Z53A!0SEq*%r1W7Rh$Gk&=p|}3V2{mUf=z1Nq(C2%Em``1s zOl*EqVC;6qIvCnr9e7vP~8@diZC)RI<>o3lQI<6z$Pk-`k zLPZ9Y@!EPMUL|mgjbEn~jy?K?GlwP9r}HxX{5Vo;)iPiHAy(((5_CpmJV~~8zQw>Byd1HWQbI=6JjemdIwKa z<6-FRo|AkaT@6z)pYKm={9RvP_qu1My?|+n2V*K~PZjBY&nqhXS;2%U*iW14L-F;F z_KoMwA->_J=sRw1oKlzAd%~mv+Uf(@Q@he&-tTLR=?Q|rKNGv*zNGDTs4Z=^91Jvp zp>>LU@rJ@s5TR`4ev@7dl28@53(G1gl4}(8K^&&+aU?STf zino>6uTRtn2aC{KgWsMR#-Bn&T|G+_gx;)nhd745F2AQ2V8Z5R^9n0R2=ErZ#NV<4 z2I*hay!=e~AXgjSObj_m#XT!BXR2(}X9$7CwE2vEmMe(k_>@{Z84CD+`tYs;aB z&Pj<4KRyRSZ=r&Rx*&Cbf%73ex?iqwL5n8CzD=S!K=HY4IV9E5DA$u{c{(eKzZ^ewf zZj4Jvo3QPD3{g3CzKw&fFjjHnUcnTV2Z7ZEDN=g$%9wZTnQ#mHzjgcSBtz^YO^R`H zcy7zU><#R>RP(@TAF)1^boPj{%{0LH>PK)|q75gr)_#z?9HdAuf7wP1i8@<%_(Weu z)NA5BCvAPEVP6d63L!r?Y+VmCQHptH-v8vaQQJM0eQkSq0+E**Si{8FymbeV(hh1L zy0HOf-W!9D=w&Q^bzEUpT?R@qUt4jYb^G6W);p=GoeQfjPM_WqZUPj44o$snmGDP? z`1A@^nmh&M8MY)eI%+}JiMr7FmI|8x?{3{s@v9WTrnd%yD?bu=q4~!v?PrVWU?TkL ziL;%Xpee;{yz?N{KL$xFotMb-SO42{(%f%u?S&KZsohx5XmI*WD-pjneOjH9U-TR( z>*BI~ldGV!%J#`BPIldKoi zWc>Z^Z%UsRq4^FK8D_?$uM`}GHcO>I^UC!vU$2 zq3hzbXr6y;J887PU^g~&@(tOVuY+k1rP8xpKVV=(k!eMXE6!cEPqI3Y3PS@@3}GHi zFtmB)^&9K@6&V<8W3I86!Q_@I6`lz~4`Onyv+#o7fB3uJF5mHO|5m8qy8TBw7xi~s z9Ax&b=Qs$I55017&V*jY_(!ylczYbmnX=dJ6|TWT8S%Mp3q7Dj&U;(DTL-O?3+!jA z#DUyWQ?G-%KoNPyee6y&R3DtKnAaMGnK}Vc%U(iXyZ7F=b;FuoFrHO(+T#bor)%7O z&&qt*MUm-Q?0k=kBaTXZ68W|5B#gZX3H5K>s>pQzjLvc@b)H4?B4a)my&P818xGz( z^mjfF-_BcsN?1^5x4%o75y%f@b#Md0FKuKzz1ptQAI6tIZ9gE7Q29o>hozc|6NIUF zwZx|y|IE8V+~v)bO(Z8p2J!H5i)3S%<*PMV82m+Z+_1%(v`S+M=#6=rXZxvB3+ z7I)@9>eE)Q6&z!Sprw0*wRK(z_LEpH?SC~7t)tNPW<(7J6^8fCO$T5#Q+k$GnmZ1d z&-5O;)r89oZ{+y7vtj1^&OQd&1>(J3$_P7~Mb$$f(YtIu!pPMOGfmF(%yQMly0p$? z|LztRn)X&!-)odGdXHrVu`L%j=R*ar{QCVog3u+sdwHU|1E$r520pUP!UT8vLUG|q zAmxrEGrg+98db5`BgWh?k-M6o?>3DeAoO@ofX?$_)K};z={;0Oyx-qB{DYgjU^)3& zY*Hjm{RHdY>GMopBJgQbq|~8HJ7`Ss?0X-jy=5W~!g$8R;``DqFw7|0HswRBuQZ-Ycai&68WuXug&QuH zL7t242a!7x(E3KJ^mG6Xj~*wq(hDSmtZ;FK^+Qh;8vHnknV+|)zQj?HZegQUTm3Jn zi?Pm6N`bMBc1O zs^UJbCj3!5o|H4|6xJxx7iQQ0v?+#O%?!Ot32_*5z0n zrQr0#1J>KFJ%-nc|x5C2y$!AZtt;QscwX1p8G5xcjDQmZWs(M7-mq1tK28qOn z=dpOE!1zwR74~nQ$z*IG=9dwHqt_g~%3*xfI|uQ7)H)?LZ48>KVl$z+e#&j#9bZrE z!IW^ur3ioOcw?4HjHYsl1rCC1`m(evcF$0Ja;LVzwB6~xAPXB9Fy!Ri>p`=R09TC~ z3B6AQAF<|HZ`GvfQJV4Wyj5A&dViK^t<$;MI<226|rTO~(J)1!6Ii=xmIZNOs z8#s^2+aBPAF<-yVf%DPOC{X=c_{A*@98b06I!EX&kAD?Qolu*?0fk$-d-DjNxQzQH zOKt*8NNMsp|DHtb2g-amHSbLeg{i7nBg%&HFunR0IiZ32BV0y7q!(;?TH>ZVcDfp*v0L^_`KE57$=D~>LvVBrD zyQw}jFxVjadr5%62M@~EG3h4xV58C|k@TQppbT+alO6xYM{4Tb=aMcXr7(tGk9BK5H`=4`Q*Kxta>22jAxm zYG^~JsJCrhegL|%bcnb=v&Gtx&;|WnDlp#7e24Y*Y3Pf!6}Ow^#fqam)|vy60J@Q<)X?$0n)Y6Ev}| z&D<(L^S)wPh#$R$<{XLcw*uXZRKgc;ghOe@HI=~G_B4xjzuQ5){}UWJtoh%mynd*T z7T&w}CgIziUMRDEd~GJwt#%s{XK*Fa|E$RHA8dkw7ov^V1CGPo+;@)~t~5ILIQ%|r z<=LYvXx=Nz3%L{pCsPuOG^=YfC!zB+!X0PNtGpTOG7RjbQYg@Rzn-nkA^_9sU-4x` za${Gpvd&HZ8tme6yEh*rtw@*TI}}^&4zt{Pj~|xO>hI6q9Sd>LC-4^Yuad64X6M8n z;d7s_k}d<8b^b%q(0`31AS2Vj@`*Z;za5pnSL?&^2*?SK*6y|>_#;*8uOIqd5KQ#r zj}5_xfBuK>b@L;EY9lq6R^_tK-9gk3I$fJd`kQv*`2Cu4kM~bu%viqb3vVY5H|#s1 zXQhavfnjEmuY93jZpUERqnEVf?(w5|&OiXFB8GYlxBZP9<)+RUtnK@6&qbnp_-4jq>2B3_%gaB-ub`@(h`7>jRfUeEXdI^$%Mx+thf#-&y7y$j3>27)*vxEAb zp^g`Fo)nlpvDf6jpkzByPij06ro5;@ z#Gke6KQV=W&cLt*`&T96Utsc<-sW3%8W_P=@qLL^7>4zCz4#_}o|vDNZhD=blz^_& zN6hHSJ7MO&1(SpQYntbr$#M5!IDPw{?;UxF>z&shB2P8OVRQKG4qAU!l1f%cTCzNj zKX=k7RF0;OFQfT788IP8m^9|{zpz#qmh@{5*c~8rgR5ok%lCU!!;Idhe(9Yf(EEB$ z?n`qKG$uX$*cdqtbGkbF+SMDN-(j&#sB{{J4aH5So=CwY`<0p_Rga*J>ERQ8Yudn@m>j(%pEpAJjbDe{%(vEYcN(i>-7bd zY=Rkf3U`}7f#Y8K&|<#-t{E)f!+N$dCPdq48F-RIsKOpiJb1{C|_pkkeZa14~ z%RSyO%20VJ+TI8j4X1MsY}<$=x13k(oOiK$;K#~28AA7O+$v+%M1C*Kr|Rq6>zjgF z#z1ekS$g98u1k0#qV0@x$v>p+W>PVA&rxo(9}^_{H=PY8O}A*;HKgRDGCLJY7SKV~#fa3cYVpk`y2^A*t{$K?9^7OKZYV}x{uE;w-t))ADtiXqkn?KmV8%I`0}A6P%eMw zeJss=#^qvhPcoH{gWU&)RvMQO_{IK&O3!Wk2%gdM<3%s|JHa@1DB_J|APt{hk>Ql* z-c^*+JlXfSa$Xo>jZ0~D2S9HQ;;j5f5-y6_Qr!&kO!2n~!3armm!2U&FGB#Fs+YBtk&8roipItI*@;G{-o-|8AFE~~LfFCI z*Hpy_UFrcYsiXjpqJMs`(_)@CZyBG!xpneunJsMi!N$gE(JsWHnD4BdoQaB-@qPd+wB(BlhzN8XA7%3b2XrQ_psEl zaH?N16xEYNxMQMVDqC>Dk39$a1VTn7-N6Gj-5`W}JKf z&LO!O1U~uOQ`YhWnlczMJm9leo7g|Jl<%Rl`A+!$lsWxsW%bwwlr4deex&_`SxLtc zZ^5nDP`^4sELaK?Y}SdjRmS1ebgy}!a3Hknik~&mbcbfAtg&Pvf~Q?-z#JT9#E+e3 zVGi!sf^b+Op7s5yDiU3gh1TQFQkW7|*f#%a0m!-6%^4DDeWL~r`j_s{O~O3U#zwU> z06(*CeANpofpP2cVPAStm=*A%~ozMQ5Ietc*;aw26on|#bdm^7a>+iQanTIwXMu|4go%W)x zSBea`mlLl$&p^L*y}t4T>N*1xr$&WtZLY_pikI7#f&{VYE5jwpz*jJKTSfF8hbVS# zaxn@1KpbDIW%1ol{U?;gx(?X2i9_y#x4A4+L>x~RELQ8!`atYY#4n1jA%DY(m=3S~ zXQ|%>QglMX(l(|-N7f(JK_46Fv}_2@BSk{hmi6J2&(2~~46~g3^%kfmd#d;_DZs+5 zd-q;7_hOA%uZwrL<-htJ8!g8ZHKd9C4EeD={nH!x&vEMdR?(~=M)<_cOl**Hzrd?V zH`0H=$BBq*$DT57OsyTE9cQ{jGTpVSX?~wocvz>TTnJ+_mW@qSF);ghb6WYqNeE1H zXWRLI>c=JSRkrpo73rCtdPn5OLw~1G(XH-2pg2w^I~+`>UZ?H%_vjl%MBXOIUgm4+ z6p|-=8uqa?dR9&3R?f%vB)pi6?RU*n04l9UbZ-BK6dkK z@%REPL?)y?|Mh^Fe>r{?s)=;q#Ko-msx>|kcJZg*Qs{a`rqLgsE}?|3f3x)NIsU^` z9W&_9mUez9CWe(i%RU)S`NCr9jfX{M_AofVA?#AQ88+FJG>`gFc@v8C>~?+*a}^Nt zRAt~d=N6bL%u(AqtPG3EO*bfV>WWMY>Fj2kKSHaX@mq=0`=O{nbCq%j1!g1G(w*J2 z4LW%$j`v^C!`PWG%u%83(EnlUgV*FU&`GcNA4cBKNPmjJfx2t+5X?g6Q)BlyoO#Jhq1b8 zc)Ls!SJNC{HD6XNr9+LcF!r{^(Ve9fTsYpy}c^*FMldg^93aOQ=FkyMqluc@l0j9(=3TW z0I-kJR_w4D~0S*g$y z5*$g5ukhpZr;d@)g#US?MAvKK8&qFM65V!=JWT}}-2)sO;W9a7M2nl4(KUALOXUah zy`=lOQCpzv$=#;Yc2wUB9AP}?Gw<7o6|ec;-ekH8RaWm!cdV5Iiqz3ybu$_r8R#BA zxTq^Q3j>a3A)XyUSoz+)#OsF|lFu<4P8^Me{#EQLIo)wES+@El&we(X(Um*5lE#5|pr#CHK}Hm`7$ zwW7y9-`+PQ!6!I9B6QzV)ezHcf*qX(sOvc{9X~cdctwUpSH(Wq+HsV$;$Ujdt@2!) zwT+BV*jPehTt0H9N{!IDXcD>Q@ac^pi9K#?(|Wcn7`Rb%aFe_$^?IWoo-aji@4_+L z$@lLgcR`!PL2Cy8Hfa0mq4xT61^$`eB@oR-jR!HygJJroRwlYsMjqI6tPiJ$uWbD> zP57~7lvPZbJ4}=49Ol@L96t<$>>C2Ad=EgId}Q@5*4t1K<7zWHQBR`J+H~D%;2?>$ z?frm@J~7UHPwlUby?Q`1o{gHqEn5$lVYz}JpHj^zWSrXl(|r2T|Mq>r62&$0)+bm< zbeY#TRB9F@Wo`p`g&GmhbmYWsUD0t3XO0(1JiGFWdfxO0^X2eGqE7ua>bpzwG|hSo ztsW{e6;Tq{tnmG(OJEh2-LQA__^J1=_K+BLN}DC$A0{z}zBJKz%!)$}d&eH768h>b znPh2iJ0kC0nK;7sSewePhpwAbV)qu^ko-mKxWBqOkiKOEyGzc(=pDXf6NTqEFe}0L z)?+Vbz17_zLZ;SPvE$}^k5?Bx4m~efWpGOcC^>=5Un;2Y0}P+ql%0KnI-g=|boCHV zF&|D}6bftIu153RxUc6{azY%mzMPqQ$Wwsh+~?VHU&ml-_~=|A7aimo#@wfKr}9Iv zm;K4K?0@)GOaxh}^9^_+Md(vmjewdW*#d?i07qbns8{lUBcix!Pf@@`=M@8{c-GhYtjk@%h^=6k4Rw==P}oQ zh^f;0_AY7zFcz*Pc=*~1?2-`Obhq*uiD|nyM~Bc?9P^d-36eL2%6$*m#Y@j9(#>+6 zW_?Yewy%ETTT;f!c4!w593Jm|37u9Zkq5_)Lr`mt&R8=#9VjsQYQ{y^~rj=R~zT_93BMj6U2|xG%?3%Hzg`u4LkpP(-O%z>2qHMj`ht47c2V_)N)t9v)mL2r^tXrcoziEYj< zTZJ!~L~p2Lj=AH72?b`o{f!2ris8U8LXLV0pSt@bz0 zPJE)fxBC;+kj^!os`?v;>hkIDlDmgpFN3e#)+&YCtr-kyqYrTU(~TKRFB+eFoQqE_ zU@jY_`F*S?X-H6&(CNEUeDSX2XFu$S_d2=%-dBt*PB%Z4LAyWdTPDleRwn`$WJ{7h zZ-UrY!f^r@BS`e$1)@?D&f(a&)%3A8T`2R-xys4rf;|`SyD)V;hVng$_l;ZMfVb+- zs$U6JGrRSwi;nb6x(vxJGy7~Wl9WMF(hcWrVS)Ae(mJkjk{NTpiC^EVi zX#EbyMrPxu-$mjem(kP^`!ZHki+;20CHklH>U>RCg)z2Dt2!pceZ!tX>+G|$sxc!zm z{^jak(+fEL>RpxbFC(n`#V@;^N1McWCg!wU3q1ahds)Bri6y40A`;yHd}${5C%}{kyMaVLy<31H0m@ z1T4I&phND$IwK8LXb}I@`AX{-REiGeC2;}_um;>)cXcz2IXK?%3@XKmgA3Eg?Z05# z(eF$(k`Wl8Mo(d2(1x_=Ek}!vn)UQzUw|hx?Ys1YurPvQ6;kG3ujcVgCD9AX zu2ju&fMiPXNx+?@S1?XL=4d?@L37|ZYT zo&MQ>2qV5)q#m??4^1tldnds{?K{_GIoF0^xiIrL+bgalm1s=a4awKS7|WzIcbV^@X+|2LgS~sa*+!E z)J?&mir%+@-=9FO@(R0;5%eSmzgIR^!%8Hk%`e^aPG}(|Up-P_G#@8-r`j;2o8zQ} z(?~~_6OIKPdCMTJh2y7mZwEPS!t$Ie`K>F&a5{7yDZx$~dta}s4mqoYL%%&=wd7Ox zt2md*;gaz@1*cf|JrBgWIj|j5zvy|8AoUYF?GZwBtj^RZnF^;Wx(^t|-!s$FT6Q$gTY~ z5M@4nZt(!&-<~@$-R`!9@QWQfbN;E~sac%Z&!2G5tr{iG*a}&vsrYab-O)&w6)psh zY>0ozxcZYARxar=TV1V%euWZ2=WX0H_aCr1aX&`C1}EotUTO}Lg>D;l)`~OKd>!^U z$J~5!n-BXOcH5F#b1{o2;ZllR0+Kh{w6y(==c_+traj>0fa#0jUz=JUlIYmhgeqUq z#mcqx8(EITVHY#~@QDI15<^o8;wjU;EbKqUXxM zBdJt>85}+&ob>YJZJfKv4u^z<_3UVYwqgNkmsmdVS3fN4o^u&`*7{}L{vr&+I=N?KU)w?5iJq?P zrv=#aZd|yXa)-oPs@)@9P3Wvl6&?P4x$h5DO}@D2U=c~8uSn-(-qi#huOzMNo{k|U z>MWS%&SAlHVX}|=-*q`5_pWNUHc}olYp)l(j!_y8kAJP@C(+#&uge+c#>O=qg8PQ| z;pD(RCcoVgnCo`dlKmT@H~VL>w!c7A6AI^>^?b5qAfv$Ea$}VaOg5_dYZ~o=G}YPX z4~{EgRr3?hH~JpfKncFE@uxggY+HL%A#?;goa6Bk3l08^6@Q_9>Ef@j>%T|TojoJRM7+Wv}$eM$Mzr7-uWb=8Z7 zTyWQ4wF!4`p=n2liFvj8=2bBEHk!U!Vj1cSA6cqiZ^gyX57U{^R6H=G$K-7LbC3lV zt`uxNuc8UR&nq#Uy>|~&JC00ylI<~e_~a2EcY?1%vHcjVsJRx}pYe!y^ySjDZ&}^t z-j`d1p6bM^fYqqM0QGj?jA}e1V6@<_+Lts2MY?qH;X$bniVP@v>dh$4KAebqHLm*E zIT_%r5KFw$(@qk-{A1?!tQF9H{lN7)p-k$17Y=er>((yda2eO(f%H2tw)NiCgVof$ z08Z`WG&`h3^^JzgaGCbmDQ}o@IF!2QkO_``y<{LGNaS7HjF%LW{AWqb55!Fr_tLCS zicBWI=2HiSh<<6RKDs86n$N=Kj=--m;ag!^>r&OMO&JdL9jU*)dOy_Bwb{@K68>;i zH*HuSjkdth6W)iH+x;PPQZLKygfbL~XHs0U*W#?}m}afB8Ct!x6H$K80Ym9K>VElG zV$Qj`d#`1WL!Q;;1u7!nO|A+<^WNGy9T_Hnrpb>TYQV$Pni*U)BNUZl0^D%3W0EG7#)!TPwI8{EHd0Oe2h zUURx=s68gze*VfM=$50H7U^GrYUkG@LHw(cg0Xu9tcW~Gd$;m_bJ;9teVkx=;)yq= z)PK8sqP_w$PbS8GT-5;?npeDQd<1axW?d`CYbNNo(Gu9-CJVhWduA^(y}-^WsiAYH zE1~FNkKw%)M7;E8_i*}&G;01Frlk%|H^&|2FhlTz=BVl0t-UAj;E(q$MIUvOP@`Y}!~_TArl z><%{C**uUtG=gQLrq3FuU7#~z&Oq4g1>%mW63#6>&|5d5qnY7?6Wi3{yKkL>v9XdR z%@g~fce~CF>9^~!DxgD!Pnbq`5ZlG>e$Q9$!oI71bOO7%us~&2m#;SarCeI&L>UYNJ zwRUf!7V2^0k$zNb@JQe)%wBX>4nqy!x2;vo-6VRZ2kv!| zJXq6njxUJ29D6st@`B=NXp-IRV#Cb`6^1`FSS+V-Y`=3yTbVUh#uoIriLAlN!B{2D z$ZbSi8X&)ZvAYcVe=v{39jY!gOeil6cJI201*|_ZpKG?_w6e$A#ucnsc(B$aV4JLzRcGib7>KI#`ekD zfQ$16&~fg9p^;26kku0A#MbElooB2jCQoodCi%(s%UY&5Om~Pcou8`LkA)|9RI8}| z(+?9!ENjcRu09PFCfiG#Bxrnv|GDl>CT+4}&7$T9u;!uPrJo;Y*QLp>2+NbJqCqlJ zp#G5mbDTRC7Cp$G4K2QjDN@*i?HfcxgKs)olCuC|MR*eHrq8{PaO83 zX=mw6X`afuF&IDoW?Ngb2#HNWaF^{Hn))~9#A64~2jBnsyCPFf>d!TwDzWDwcZ4Rh z4945smXwvJ(dC1haOX7V5gTm3?R9W8t)ImB-v6oQ-?)$ESvqf!Eur7JYDKc>p6yup?Sm={vmH*|f>oI=dq}L3etOj>i=eT-()(#;3Dg>&RX<5DfWxgL zmy>rdVdw0YRgvH9a9)fl>6#%Gmqua;Ik-5+K=1*^cPbq2xi5(Qf-EAB$NflbVO}=5 zjZ!rIHs7IdQ2T9b6vZzJ}N zAzw=OGFpDX@k=_J^50*9?(?@+-RY>qguIW6F%oh(zxu_SZD)dL+EKvle)&5iHUCCp zeXm!lGeYPVCrxSO)E@puq8AQsJAPsqhJCt66Q!B|=XGeV+h%Lc`*@nF_X5L1K_4pn zsQmyu*_y44dIxh; z*LvTT`@#5Osq;35cX2|f_(kpEb&#zv&CWTkh+|(?n3?`D#st1k7hX%6Lgn7AACu+W zU?7Y8LU5W0j`y{0sIoZ-laJXBPG#SO-q)OtA^Qnkm;s)xriay8plqCHr|lUkju6M> zeT~=G7eZ~sy*)=Re}?g(+~#wc6qs45b$*59G01c2+OQk< z2TGE;xaIl<=(|4lM4;p+R>gH6tx^nxmZTPLw)z@qy70v-<oK%bEHzP50KZ*#4^CbGjE9`Hd+=2Y*MCPuCYhcXL!s5gRKWKVrE*bO2276vR zK3Hn1fzp`Zh|fwrKuNx80uIACzO(Q4H;zD@-2dW<&|RV~F>}}SP`FbljO-O=u+UP% zVHRmy6x>2$rsGpNwLSz{hmP`P74@20Z@Y2{mq;2x>J?cx5!0CvnImispu-E5U{-Z5M)_P!8_HD_vu`Hm8a zA+!BW;XO|n4@i6zz_&!A=Xoo5j%yuGuZUXtJ7tQrA~a}pI`uP&EyUB1l)Rd>LiGK$ zpqB*SdO_}M4VM!xX)fakiRB`DwMncY ziS-Xl)b-lG_uE`i*z3kdV&|7Q6S;|q!^rWE1T^;1=)P0$(>c($&+KOlv>jgqOU@MV zcD>Ye>mwEah;`Fz6kK*$VW&m<;?)EK?|0Lx=>GPq=QwYa+}`wx;1&1Zi3^ZX7lOH{ zqR@8EG7_`wneMu`0o3-SNE==_QJ{t;Blm8%B+=l%sMm>^ls~zUMZ=H4ay?Dm!)-); zp6s(~4`(bD?*rwzj>E~fH^P*k)cIP?792aov!L4TN}@|lK6lDV2BtVfFRYf7fsqj9 zHhC!yY*;qGd6vNgC>QS?eeWWRbL~YvOL|2}@qN%(yobOGma_{QoQn=rWPyvK?^TXK zw`@t6x{C{S{ijoXr@C#g{XgHQq0e=P{Bz!7k4$e$PU!#IuNgZ>k6iL7B>13hmM7O1 z<$Z@yt_;uN94cP|C+*37wYLT_RkKcuotMDDHPK}`^$W)<(ibTt(GxfW>Up`m9yf=S zs5~r~Rd903Rt|$XTb=TV4l17%{&adCtl{av+8i}CNe2QaNOA8-iM_-I<8hYW(bCp1 z5_c?2YCqxII$2|r(qnW1x_2!ute;bXA^Df`zwe!bWgd6-w>n1nE>lwYa5ozaSnW7y zD<%UKJ6(6Mt#teGZ0s(j^Qw*B~XGO+ffLHExB0 zjlWZi)YGB4Bw2-T@g}XTOhU?1Vl!k8NB(X!bEAW`D7|Gn-N&t>fBfwN;`ddiLO%J=}M(H}BUuk1hX4 zZ=;0kn64MVtLnFEm2}YCb2`aRza1K0+C9+SrV9=4)w#=fgmElVCrEFkAH!b{xb+M) zVf&dIDLgO=d3avda{X$I1lb#VYKi*E7{?;ZG5=H~2e>ToPv*h&p)t1ZGqEHFes;yp z)3X>Yl}pw*bsc*fe4RQ;G&(f6BzjO%(zFFTB5v&!)P4_z-bt&vj!xj%_~EZ*mp8+( z<%ZioR#9;+Sm*!H=?147WXbCOaS5Q}-ZABnUz+eeX`Hx!Y4n-{1xg1FE-1=8#2(9~ zH#hqWNQ|dhp5%=SLG{(dqxwEBIJ(3C;qj-o(ABj6bf5?Yt76I+RUc|$rBUCt!|FsG zEwxP5V558j%&8C8=KpvO!`vLHZ0r85SNzB8mV(*BGZMW>y{z4STA$1bH|Z@8l$`MA z(OaUc`9+YtU^~GdMDQYfC_364%+H~WO=OKOnH^KF#r}M}9E)ZCe`Zb(l;H0y)|P%# zN$9aL-Y(Hv13fz$GJ4jwLH`tg2-g`3#P1FKY;c(dcR;=WfazP;#a@fQ<08G4tYi%N zBE?z4--O|Z{8Cp-ua(%~_h5RxT)^2B{wKlyIm9g9xNMKEY+`!M1)56txIe%X*a04?`rtd12j zP_LVJ`=*rjST3eb^f|FNg<-|fQ++ZYy&*?iS@9r0mA4C{R>cel1LmPI*Z3i8B~3gH zONQ>s3iCaRbe9(DFDg@U#~2p1SG1_JiCCusL*sZauSK$D;&Ruf1DzjkjVV2=hPW3m6t2**V9cAJdsax6LC4+TD!S{-IHEt*O4j}gU8fD-K5g9% z6cZN-8Oa+^z8v!FWMd4Jt^Xu^bj}0)A_hL~({aOkGV{TeVuZd_wTzBTlgVeCyxJ?5 zIZJ_B(p;hIDtGAHOddS+Z4?Hr0)#*Fu7_fqP<7sC37FTBz0Ko0BUaH_Jk`G~3ROA| z>)*sJz+lXY%}b6qphb?AFI_GL%C1zZZ?ctzZpF5|Z3XPuu|6y|yf^~~f4eH#Pwd1` zkGTfU3nxIc)&b81$-nVzUe+EVNFpORWcEcYWifmAUp1->+(r z%YP0VWImDKwADe3-%3AivJ(_RchwiF^CMjtQ;=Y3d0ZnT=N-VgF^a|o$2v4 zFgV|8ced~;ENrk4K3yCN#elb>25lx-kPFvpMi?=#+urT zal0j{xG;Ek_fx{f5nB=+OG<^AcmlNgzoKLQPV{r5+wYF=GW&60IO!UT)peMRNDnC$ zdIy6%rqf3U<`wA|6o&cB3}A90amIP51$v{&U-kwjV32v)7xz<^FfjR?r&n7LirgF* ze=v3Yeb0)LlW*OC!t1sEGK1{!!|CL9RYfTJ7f#$>NMr#jz_E{ls&7l$*U5By`!Bh3t%c4zoc`u;82BVt?bh&G90i0@Prb zD_?HOTogvO|J*5;9f_IjYM1tFi(n!9pU#lApJCKgo@ejG0n9f_=?GkL0^KTF(2|at z4}#jq!Ho}?>!E$_R+Z3+CYT|43^uOTgauO`K8AjMB+>s&{_MB`;@QRHi#m!Sk5zNW zW%&rGkkDLnC3#knWzC*+2mWr%Q&+jszQqD-AJnaC3hV$<;ADyAlqd|{7B}KKIt8Qs zLUjw}w`k`1I(mQOg7f$^17JEU>Y zMg3{Pqm|HWO;$B-aKmU9+q+RBUO4G0TGlKs4nNd{JgV!0pvgaKOn$v2Qa&Au8xNqt zX9K13!jtW%wIEX>*EZm}1f*$wm;4y}5bKtw%Sv9SLCcG&IL@>?7$==1$i(ms$5?!y z2uR$+_D$tCNE6LC(pFhK7fs}e7I;?XSAXP%2E}zIZD9-e=WCD^UvmodiTcRrue%5o z@yq&;jc!7R(_!Zc%^HlfJE8TaLKZqIvqpFCv?S5F?UvCu{(>c+Ysv&-`*B*ya1_67 z#*D|q{S8K2V92*pQ+v&ITvTkq+9T(nyw>E@F(*14Jv}d%_V6i;{T{I!vbzRjEdnOi z1LtujX*7UQkjQJb@65h4-F5~`#9KTz=Th^3kXt8s`RYreu3m9|*eiJ0 zd(qq@KlPSE#N~NPF|s8WuT!km|0zLDKo=3~wz3@H=!u=KIiXtb4RjX3~&`^8>rH z9=<&L_kB>@S9V7NJMYe2FXi8Zoge=$F|_$T2xxx$Y41h>5=)!k)1p!uzUqH{&cTzZ zyPAB7@4j7xMd-}#PN>)NW%rFTf*zwju_LDlJ)T_QjH3`b0mGX&hbNqK1WN3KN92Xu zQ1@sNboRw#4bMjGYgz*%SJQ*8+CGNaJDTPqvGe6*!uuAkr9smH{so5zFQNC>szny^p8wnahJnE| ztKt$bLHDb=GUFeFp9Hz)qfagel~)2wKV?d}0!Vo(}DPo>K7m*qgURJ~Ywx z(`gpEGDr!T8M7DIjiZNVrL&q{VLD7LF!5A6wj_N$DK|&uSHpO~rLSj|i9A8eRzWY` zuD8_dm#Vz)=2%?7ALo%|5Yq>ZKF>F}eIW8WU=6RG^wrb&nDxH z!Ju!OMG$7$q`W&UBZEUO1Gw_R0qkx39he&*0F-BA>zEFiLiQZ{#7ZlIH{0%MYq&VG z30p@$6iV%p1s|Ep?sau>&=4`4# zsKF_L)+2WGW00>ic1@$jk;Ig^+U7}M70fxRR3#=B!kjPv)YWQ2kF!^Dhepm|4Rkqu zm#>UJ4AH`$yb4-5a5U0(eOWo72NfRU#LZ=`iY>ns$f9dwvF+_In?T1SfA?p4`ObNN zFh+X;`QqY?EF4H?Fv8b|dP~NlZPprO`%i6Aapq)JtwB z_{d#R*355x25_e5(8x=tf8r!DZ)-1~?Llh26B7S=}vm555m>-GNlQ79b(d-ltfsYtc+kCKoMLvo7Kt1E0 zWFnt5M0#}2cCH@#Vx9|SPf_(eNNjg}blzB9#fh_VbD2j7y@_!-1J#wGS8#O2p(DF9 zX?&KkLHuUq;5RB?1ByGnPS-Kf_#pmW|LXKJ-#6pbE_EIEWk>8fa7*f(JB<$(E{f$H z-jdITQ;xOuuA~tX{fPP_%JCemIR?7PE#B>H!m&enM|3#SAz)_@iM3G_2L@bHxW5qkIQc!g&8J^5 zVSC2jgNqRj1L+hc~+@asOR0AaCwz{^dZe&yeVX zgIAj6zJxO8D$#O3KIrCt%#x8}f|LM}jiNx^SCLk9m2Q!a)%zPaNNm3T)EnoS8$?=U z`{B3h_EUGsRR1}cEL#*xxFHDD=VbEj6~4f*L3i(Gt`ktb*qzW~dJAH|qp(D=8qSJ# zJUSXX2^Db;86PD2u-5*=(d`mc?av;yTe3?_k^;QV=gNbhS|G4*i|-To7oAuyPiOzH@xzIyL1LaR!;hw z4k^N*a#gEVULj6IxE*+^`3xz$bLYA5JK*@D68)wX%uu^)qCd=v4rdPTayWlcn6#pB z^BC%kKo(1VUWrN#&UYL>-*%D-yGLBNu>`+|s?v`lC`<4>+Z!ZyOZrq{LvOA2HgF&@ zU-%*=vc(k2nHIVo7CWKsh+s?3El(Khlx{h;{s9jAvxj95Ucs68Q+#R)uSjfr{?PL@ zzQU&3;Sy$%RoK_Z?eXi%slVUfGo^60-57potomcc9xS|5_U>-f{>hcdFTHFtH2GpE2i-q+ z`OwMH;(^PQPub=Cp1_4$VVBQ3ej@6?4fPf#^gJt3jdrx3&l)0 zBTVg(l2G-#g#ImzCp_#BxP1gGB`j<%sm_t;l}K4iZVWKBEkEF$0ioC4u535#<8F%m zWn7W=Q-m*VN846I#g0;(TgNZ8LqH9Ou3Wg$;oF4W;kTbv4_(E%D*8mnm24zBD^0d@ zog-M8_TFir>nqM_vOM;2sfDVF?yBl$4H6@7*7UZ*aAN=XIOa}54+gDX zn&WnjnHW@6*2iu87zJ(9YX{vX37y}tPbC&NVv9oWY8$dI8Z; zQK(eg=r!ITj3EZ}w}zrKkrYSbjoZeL@99TmPO@oWUw7{(j<8LT@#1mXQT8y%-si@` zddvYvUmaOePJE41&t=wJSfdCXH5yOLB=5m51}onT?`bG0bGVm!VG8D4WRAr}I$=Tj zx0o$Lgsx^Cggsd?WCB_JsiS!t7-9Ub=9$EX5eR%)a`tr7U7#HPAn$u(J><&^I0sfg z!=B@;>Fcik0=KYO+ev*})4ZRpgI0XoHdFCU z&~y04=&7)8fA4?4fPMO(_+CiN3U4`TPWU+d-1@Qj-VP7=H5rg8=t$L5#msXb4_;^z zgSNP?Ikk7om~ZaN*~CJ^=0UmHn=ST`s4OQk#C{GMUC;6hrLbUIba3YQ-C>+xBXGZxZ^r zqg!$_y&(~@HPbv3G|M5HE+~boQV|EXmJhiwm_yl%&2BqbjWI*j@T6cT;g_0!fd5qy z%EIKGp^*N#Nf?M8NwnTQ3o*5N-dow|VEpp_wAE@~p)1d4ReDPrG&Tg99(tjMAx8JF zJ!kHP{`D(7^*>MIVumT-?15<@uL_WLu{c3u=&$7aF6Iwyc-3>X{U&xhf8h=sBMVHG^ViSy9{ua)% z)&Z-1heO&8(y{JSoMO;wcZd$ue^MAk^^u1v*ZMiZ9C_%GpVRtwj|Gb#dFgx^npdQ& zkK#P*lZ@nV^srd=4SSh;70P@4AX}AA>y$7j=51QA`IROMqno$hTsXW25<606-#|3B z$C#PS+Yxw^Eee3FC2_Zxp=PN9kwNX75!)tfYsZRCf0j(K~w+w6zBER z&?#(nH){qB6Jrw)g^tgk2R7Qr6L zYj?`r1)+o2HS${iUKq*_V1C~hjGf(G)xVwNVMLs^fS7MaR-8O9u@U=Ur7HitOS4agvfjMq%!JcOe(TKpQ+NdW zubAy{17{c=4&C^b)d|XVqb?U@v|yw0VE*jW_0V~Q{R6ac&3mOvYii`Ywg~I-30@oz1tcfyewyrx7#_0 zDsWL&xwH|xO(J&qVKy|~yDI;9?HV9kZEU)V<|O7P3}*6izo9|>=KlA$2|dFxCasa0 zATOkt>#TXY7z2|WA4hcTQlb6EgUR9(<+!|GcWtT}lotS@n9_6q-3jwl#`Mjy}7*y|;V-_=O28yOupH0jI zTvXqAd*bvT=(*J^cZZ3J2O!Z4<@?+7(CR7mu^RH&R^P(*PZuU@RkuU0uBjEnHmY6? zWJNU1?2-5zKMKFf<>4drC1=lD|8clW`0h@$eBCO(`Z>;Mh(4KS_<^o>IPFqKYj84W zCLvjx@N;Va@cok0^$KYBW!BhGP>qytZ`n_nr9c&D0^^*#FEnH`S^rd1C$adfTOWC2 z8~Eqj2&j!Z6YIjovTyXv1z0mRwsuT~IuGDj$#a&fehN;d^xA)_qw0z%G8#zt#IDqV zDfb}PhfU^CH+yHt$)i6>^w!T~$)XF;aPey~NQsab1;Yh3r5OKb|4^^r;u-DmxP&HO zhvc8`=~hD2-*Ntk`$}off8vd){Ti1Zf2qc93I+wto(Oi+;81~bqUh+&1;VFvsoZA8 zxE&WxfB3pFQ9}*Kn*?@#B(-7}*Tw!3r%ljNRul1&laIs@VlS~RXaRdCN)IO!GdOXO^X!Yn1#B`palGQU6;iz2<+A!^ zu~pXNjoam8IG^d6M{k%*qHlcn^|pXCjuegvPFiokNvV_k^^23xOHUX5f~f)L>Q20| zSi^z!{wEGz+V%_k)+?$QU6R8=M`f0C3`WrCepk@+=^pGf5xAJrdmhW0Qg&P3xC>p& z%SZWin{m`v&b_jk(7Omtl;C6k#f3GXKbjS4Hr<`hh4^(){O2i z!of@C;vd=W;e_Ma<7*-S$KRmCX_6FnUjHJfaP05%YVQU+PFw`hmBO? z#17*3d67Ti6GVQtJ>?+fo#rsM`ZMtTS|WVZdqyP|)4#s41UB%xAkmzc(F6gRp zh}+VYjMD+vURnH5#3{?5EL_?JJ?u*!Fm8a@wkmz!jJH=K%&2hZm7nVYkNFC6l+11(8IQGMFKb-ar=sl@i6pndUi@y z$pp%r=w1)hQu7YbscBv1#T*U;bssaj^qxRZfu%r}5GfJkFBE_9Pgg##k+hv+RrQh;LN~HLQ zJ1f;N$)YX)rk_2OX{9v=E~`WPDW7ctdW0TNuZI4{qXnBFb^H1K+5+0xTOyCw#w(%r zPWKYlGePfATIOfN-_RMXEpePIh4BWOp_jgD!sG(WmBbbL*tRZykAV@PF7!u zf=5n1+wEBLLmfYzTdfe2MCFHJ*@MC_1HMX6+M}sHCTF=M(Mmwi4b&&TS-v?Ff{-U*{&uWQno8Ii7 zazBp4insd8Tc)6;{%c=AVGaBdmp`Y+14vF;Q@-v1k)Lhx8dq&yrUO#1{SC$DGth8D zZSLqU56Cnqf8q9x1Rfvf-J~86@lXSQYMke_Qf#*SXl7JwiS0h3>?G#**!ywI1lwI| z-Vl2(7U`?|#>4RCOqZcxOGtkG_>J)neaK(m^*Skz$j7&a|2lOpcmtL+MDHNceZu;% z0dZ-6DxMO@MDAKhdo5wt!07rJTM5YX`^>g*D38RZUw83HN-ws#8)QVQvXanx$WS*hgt?geWR@qKSK*KF$p_{bU zV7>HX=+991Z~yx}Hn^GR`)gW5$^*eRxBT72{IDz<7Cd8srEtNc=u9yV1}AzH=$yvU z!cOLhyep8ZvNGk09{>OEJn(q$&=Mw^fPPksYjiKi>Z&4x(Cgd%h7#2XeVX2P z-!$$9%ISWyzCM5Gu*$A4(Z3FTw`v|LY*)qNcc#DG&ivh9hJ8}#%@V?@V~IUC{v3fB zo3S&MN`(JXt5_3@-ug+XADpapO*jaH1FK6GTZ*7xkWMTk{1R4h=sjHFR0%z6FLH;w zy@RC$7uWp;gx^R#&nv&*m49H&cgOrdx*XOig`4)Ce~Rt;i|#9XzhPrr!I{Q63e>Pp zT+iLxixcicEZ+hNzb1bXwH4tP=`izANbV4g^~|K*9Ex<W^!|P@Q==}U%QttVfRfzmKsJn@Gh`!VvEkj z&bPDunv66)vBW-H!-2VY!!nLFRpq#u#Q-ILyJX_}tp9l&bZ^>u!BX`##Aq|_c2W+9 zNe%JX<40w&uST+KpWZp>yZmj=_4ZpJpXX#13GIgQfRzPvDTJR!TS}JS(V4&N!=W}A zPTTD`r&Qwbr2ae%b5-Q(d<%o=U0p`1Cy(H0;OOR{cV*bv_wo`?(r!qrWLVi2SO62U z^>v9is61FC%@}MC`p$~|*MF6&y3a$}$Nr(2AJSMyH|}_&hY{+(TXAkV+K0mnt9HJ5 z91k6gMyebPTL5c59&vNkhsv?6tGiCDh9Mrm_v48ofDfEaR>X8+->IjUOn8(rRwvrL zvEv4GNX7d*vR%i-Yh&r>7Y(3;Ta^8P(Q9am(GmP@O86BQLwnbGj)xG@v13Qux}TV@ zc3FMIGzA8YVm|a<(k87;w?3~?KLjb3N7gxIE5qz7$6}tJLonq2tt{b@8BS$vk@gvj z18Q4%TxcXz|#UFv2f&p`*U2_H>`6jL>}hVnNGbQ z=Y;N#+}b*m4=~=Tkj$u{jpb(ZcCT|2v9^qJ)_p&LKbW$Ad$0N8c1Tl_zx3=4KT?ie zZeY&)j@6&q`JY))d3zXPs`@9un5u6MLz*IGhxX26Z;F}D6+33A*?hd$OfUe;PrVDD z-(Lu2YOpQbNf*6xj(=?Oq~=wSl9J9PU0MeP=N<<4JS1?7lvlDLmhn-4_n|9HtM+}y z{{GGH(y|CW$;$FuFNMC^VSA0b3y0}i?7S>1psl8e{m!cns(GJB3Zs;|w^tC>l=3r4 z28t8U$LKiS&ril~n=?r<{$4oQeO7P6YbOlrbLOuSN`TtmY+>dBG1xquEA=g;1Am__ z<;^~0g7c)fjRR?MP-$0D5$Q3Bqi@Q;?xSCVyhDa64&`5wvh>WKU$+OUKb6Qu+;K

    ~TX0D@`j_T8VGv2YBse3J54|dFLF%cEH zhNEzJv>eMp!7D7-cjfi# z=cY3bGEvSqP<)9EMor#XOZal$-jl)ziJaK)PC@Q7l^ zp88PN&}p39+hzEE9ig}1D*fXT)cIo3#;e)K0uDff${F8_;gvA{;b_aD=^`jMs{hp{ zn2!Tn)JwgVL!g(R@xXTWx43K`^UCL%3ykuUpM5NM#dbY^KeyFCsrQ}c%TXE%=Y{!s z-@g&j1a@w8dSz#yiqpv)F{5AA6%k-yuZ1Jx13u+;9uRxYFzsr<-*_pd@}|L+ zIUM4AdbLrJ4sx|Ougjbz@Rf5B=WE_9sz8~`caf2ywJ`GHsu9O%Qh9 zG26x)@=8YF(n0}*bv}mJ`~H2-wr`NU7+&!%i z*F87i^yTk&cV+iY<2TUxG}n)fJ^t^!O25zkF`*OE!`l+%b-)eVJV?90dPm^cLzk_C zXT7oJ$0xrUyURFi_nG88cN#{`w1291S7L8#n|spRZXl_MkFst3gjpON)%HUBpzw+5 zSx@G4h-Gj4vag(2zsH6ihNK^)#`)l~$!<5df-MdlxM<)|R}X`;6zlu$A8>-_`SPjh z5TH~GZA&KKfsuikedp4IAoeF4(}FjV|0)iM3aU;d_+C|aqNe(qU7+~$K^2dc>Cjb| z!p8lS0?pb*lHUz1Fn>(sYfz8^G&_tgqw zD8rm4+ghNqcVb-pRS^_Cp7yEYTLX=1;?MRlyu*gsn;$0bQgL-y8+`rKl>7}W>rJou z@R-nVXx86m$MyUh^!>g%s~%7S^}gi?=mYz)DWEv*=!#CL(`=Q>o}uwACi=@*$30!N z4#vM#=JIy_$M?gCbBrDf8Z^SxX!?&+r5`3T` zZMyGmJB#r5yYoLrga}?rh1KktMp|A4_5Y)Et_WxDVSwSLeSZ?v@8ZNqErseER3Ahv z*LwH%sb~mPFZ-^$yQvw+Z0x_+uYLeE4~#6tMBd@$lz><9(u&v^_B(P;g9A$?)z7`s z&xF1WNsc|giTJD|;?K;p=+79g93jI!%7Omd0(1_yuST-0iIUD9V!df4HQn@9H^4=b z3+uJLKe4^;_wGopIpR6fSWh}{`G~{33PX(C9+2I{ysxE<@Y85o)s%aBh`{kB&7a#9 z-`WA$89)EHBZ2?Pbi3KQgVzZPl!yMzJ?@0w7ww%PSARkchtZwd5jyC&!T7XCiSXl{ zS^GAqc!&vnSD z+Yv8NB0*xhB9=eR)r;d34{g_H1i*+)+PgE~4X`!Gb7M2hDM;(hdvdXN4R*>7ej&$> zL0^oA`j2TsZ?n@e=6?257xa19f-hK{aD-!>R;skG(vxhjEAG z>%3i%`L)tOb$27?U3s`8xgncGzuC@SLJj}M6E*HXdJifxFa=9-?E?xV+G#KYm^uR3DCGp%6r!uHW|h>zVL*dajF4aU|tE=Co2BBgdiHF=5} zAHvvFpzWCfLeH`6=!!>v+r=1 zoKU+*^+`d^>i9`dc|Tm9(OIWh`y85|B$ig(+X?l%4&AU}jKt27&!ww_X?@ieHG3*Z zy~iL+!;`-JwF6Z6F}|8sslhfCpUbUc`p|QDmYqp(4b6Euh764l?$Z4Jzx@{)`=SFj zyi0XPJvA01dUO}BltDe2298A=yd))LKgP}P^ z@3}_=9_mWb;5#TbS7V&bA?F09x;-B&EszhAvg6 z`g>OQ|Jgt7HCkM^GwU2G`DAZ4-ZsNHhO`>r+zKhyN9#X&@M{6&5i(dJK6Bo4|7k9BibhlUAd)AH6 z@y)n}Ic+V}*e{*Qk$DH?9WNbGnE_fNBeJPJu+PVsP$*o+SDo z_jPz!K#RAnvu^SWsQiDqZo`(;T-7P)9=~dquj35~HHDnPr#zte;<}C7)+<07qkyu3 zH9roE{Mt0YB!%hmL5b%Y+n~jA`+d`VU+g2VzRqdVj=4>Sj`Z91Vb^o9C%hWL;619h zebY!6iSeE1sH)x;tV*mL?RrD-r23;1@2`1y7)un6`c1Sqq3Z*Idk*Tk7z%2K??1}H znJY&c@*ZtKiln&m{MzSG*_1saT}AAFTQ2CPWof;}No~y;0}){uw%&icz+?_OgnFFJ z^(0A5209xi=HHSSl(YX8KrE0j-aC%ApP3d_hH!=fN2F^$ZQJ z0B(PG(pnc|1Z}0M+V6~eX*Fc98HgDScibN;Iar8h5RTl(vCp^CIJ@O1% zovuTeS<7i-D@^nyYc)#OfV~KJP~I-saHr-WDW!J%yHjp%ko8s8Cved>d93 zH!nSSdl$35KYMN#xu3+&xOdM9V__JWXn%C!+&DCa#=I(El*dm0Lk!LPEurhKP*daD z+JAnx{S`A|JbN`U^TT%&p7ePVz0;zKn(G?qGEfR15-0RmK_7Fz|> z_TG`H4m=2>r=)nk8b%`N?04^htbl*sC%!lD;$ayls8T)6OLqDXTuYZy&nf8|%s1Rp z@#YExcF}zk>$qtM^-rsQnO|0co>wPI=w=e*h#A6^`hR{(V)?6s<{JJB!7YV{ZSNNo4pFjej?g-hN! zrDD=v=*@G-T!N9Pvo~=ig;~&P)7-Di_}cH^p{+Q`VX9HZOoKOtYSom{)12o?jLV6U z&j*S6+t|(Ld*2P$LhFd1_Pe)F(EXFv!8$HF=*%Onn{0^0%(xS_Ou}9`yn5w49h)kW zZ>IR#t>VD2d?Vn$B8aIURJ#ox%0YvRrl5OxJ9a$rixE364vid}=Qr7KW1YUmg&L(K zNRmw1_H#esQ#&&L(75xDb2hIfkyTjb(NP}7j)G7AcW^oz->Eh1OIDG*KTy1i^Kc*V`UcMK7 zkV@lsf_=7O+qa8%<6xziSLMzPFcOt=Z-KQGx}+rpe*M^rouj4aQVl&Z@rTLFB%TB) z(M_K%368-Dxi?pS)o;N@m*?Y#wp>uQVIuOX?R9KBqG;vL`5DKBpR7>Qs=|q9>0*hc z%Mfa;G}3Pz0DrceUpKyED<<`RvNvQ`1M++^k>do+gS;S!+}5iaw4$e3d{WW389{k`W5R749D?NWFSZS?EeeHWjj%aB&5{xnq| z2<$qdR@T`Q`LBE_^_doacWkP*?qJQOL$ZN}l{-IG2LpOSGZrjt7;#KY>)^{ab?m4; zd;O08LFm~$*m^m76uN(&{=C3P_&n85WuBBeZ2_HDLCKH5>AUO&LnfyZ1RYNl5IVQ zo7lwxEpuifFN|UL)|QBsE1$uzY|xL>uUR0Xq%^Mf2`|m}-=sUKE?}VwU2M{piV4(s z9A}SWkGZ7=wtGE&ZyfdmbFcR~z1>5rA2n~y{Z+#95Dxw3&Sx@C#@Y}+;T2vrFu^mT zsFk}1S}A9x#DcQ1M?pO7`UQfoG&mNxL0g#zYugtqbPEiSG9b6-yh<=s&m_1Z(*TUs z+_BDyS4EF9ruG%gD`7EtCHC};=z2Sb#D%AP}G_EL(DtPMz>Ad$g&%3KZ)fJzL zpaeoMg1q{~z)r(n=ny`}QXFi7`Ay-oIgeDR{~!CFy+dF8IYy3Mr?b_N#kn}srknTJ zU{q#Fe>iI+cJZc;TFqXd)A2s?=Pe+>T51j_wB-pJB_~mo?&C?_ll{eV0J}bsQIc zlDUa@0Z{1R0-Gc4JYCQjyd^bf>Tg`P+oN?F z_ywV3LdN;n%PU{lkwxS^+9~0cmN#zUz=NBMGW=El^ZNf>9~UlGUv}=ZC*mB^voF_g zCqW)}XeaOR5Z2djZh4z&4TG0151euP8{b}d96y$AfXgNZYXzl_LEFow$A(EmSR9c$ zH$|5N4KK7-_04f$uism@$nSfxZIhmFDC1LX?(*qa6wJqxu7#K7bFDbp73?9Sx_}k> z&c?kfT_9#th?IUrI_7*@-7-C2G5 z{s9~_-e>iZ5`+aWzV(!j(9`sH+sCY`J!76YG0|vrc5MO9fBo6L)lZVTm^R!$3+)Hpo~6@;l(sX#Fg? z;j^g~c9|a&d30wDR7yFjXcTP2p8FoWzE>_FslC^zUyKPB>#kjpw9JB<6K_7oebs|P zC)>%uA$y$l5ZY_*=#0}3F2Y-0R%|k-_Y(b0qeN4IeyIuBoi#htR!Q8(CVsfYkDRcNpJl7;<#U`}&{`i`qp#`E37$ zopr(1uSKPh65QwS7%z*H3jW%alLX&oOzKBYpWg^{+UnnaIZxEj+spfy*WLAn=3RbL z-J5B?f0!-~y3;U8;H`5-yO@NZa^pmcm*^udS14XZWtf=SBQmLc z9i&KK$nw4Q1A80kEBEapWB-Mz^}^?B@E2XeMH#n!Qu5iQz<-^G^Rd@<4qtA; zSkCd99mTYIMXxxThWOqdg&wZ-<0rpAf-VQvep1agn)^3VL|C(9MKMor$Ijhjv_6y* z{^{CJ%DFKiou#};&lLJUu5WAKz838cO2mapeu7&2jZGg~BC)&kP5jukRG7T+NvUzt z2WQ_q%PENwIx$nF)ohA8bCI%i!Br*mJCG(;Z^%mM#gQg&k1IlC91DFd=iyGBPoPle z-HOPY1TOUFZpHw%Gg9bS!}y>&$n@|29`3!CISf5lUU~02bOUl@S-hS)+aaYcUWcXe z3XW;!jx*5{x}LS%2Yv*#7@||BAI^O0G7oMEmdLrs_tiH;f4*N3z5lTf z`&L^V*n3cqGHMEWW=~q0=u4n&O@MNQAB{f~bsb`jTuI90w8pt|_ngiBVK6pal+(B? z0{e!Pu9pg)0P#aPGGFTKRZPPnSw=;;g`~8D@I#BTq1V#+_%c522^jwPsXuSWg9O=*tbh8j4`! z_@`f!s#@^lebo8EI(BS(6J5kR(F}Ht-^7EG-XbNGT`i&@2%XcD4|3?p;^3KX6x~w* zJ#P-{CbkjrNk^>+*X^$>piEm~ncrg?$29^+n-lNA+@anr8XP;ZKabr~l*bOca;!21 z^BQTMyQZ9N{QItd`WjNlu||({9X;=3Xh#cNcDfu02*a zR?GhrFH7u~%;&F3uNV4v{~lvGaoIkHcK(QB{-w%r0V$(@4$sv7f#eA3phFrSK;Eyb zfA+H$_4-9x^lK(n7%+2>$eFLLPXD}5w^snYoAGUIs*g_jeDDv{q$xit5MqZo%kG2? zY4*^lw^}t*?;yk`7pgEnw1BG3hEZGR5krrPuGwupMRUD&*X(cf+X>zly|5&cyfclj z06L!QI6lVo70P4ovuzt`fEmN~1cv*%P`}wUn72Qh#H<`!+Az5dDCa7k357O6L6uQ& z<3b>G6cwM}*`5sZv1d(E`HW#+<%rXg5j%EEu}3}Zq}d-sWukLpw09epAD@3$pl6Hy zEtUJlj=rb1LpLBjS;NzhwqMo@y3{k}19=0b<@M1)82+XFG}o9LhHB`yZc9+YsFES! zV0l$Vy5kJTnPe3Zo(hC7Ba$21GNMN& zXrJX%=-j)ZYTL#uFwP-0ax9nNC)e8ZDt3y>(45!tB=We^M?%*nw>Vh4sfosy0TM}> zFNGKnVcV}yZo_Q1uv&TV$wQqISS_+I=}!G9Oe=7*xhV2NiQ*r-z<&~XacDEt7b zx%NHWy?B`BIVh`#*Y#IgVWnhXYy77{5*@enu7@07Fs9{d&=nR!=W=MVKhI0u1jnK+ zu0$9OL&XjGH}#iFu_OQ7*hG>ePUN{C+4_~%SH3ep>X=YAm3N2(ZQPm#8S>cis6%T- zZUZjfu~spPqv;Qt^^i{dYa4&njem~EbkFyiKl`h(W0TSPwUA6=*ecFyQBUEq zZ;U5)!~B(qkcH<>*u4L3Ql~b|B)Yh-#P6FgWp$kH{O|iCk+V04$EooKv{sVu>9{&$=P$V|{y-XiROqOZ zEJ8c~>pcEn*NsZvcXcHX|9+49ZOm4yfz#2{y58S?Thdpj!Zpz1PN|Q)r~}1jg}znc zuK!=#hm{{|Z=@A-(6qb9cK7O1Wg4A(9CLIwH*){=zrKEBrmz1NPXeD?xp#cif{PF| zSRZlDWRb+mS2suVmxy@c_f4k<>2eo{d3c{++BPeyJ{gu3D>>Y>*o|F5K~CH9{=I)+ zEV#QAYwQgDS*GlaNf)pz?MSLx+Z!nJo=$ypl8WmfF_r$QnL9_#pF!)2%T*=2%%JJz zKr%V-I1Kr?vl_Gz^Jt;?X@wbr_x!)^hm89ickWHn`dCd5-eoMTCwwNF49{F-`+sQr z^KdHT?_B^k&!VU#WlCvKlu9LQrvZhE2Bnb(L?lfbL`tb7R6>Qw6qzF9+U9wl=Xsv? z-nMyqJAb_U`Ci|Bb^XqH&w2mq%6^8=^BLE=*S+o)BnTrJamJkb34hn+x@Y>bOnH`! z>j&f{l=@d~fzkHba=B9{F!yoLW!7O+GKY1l-Ibe!?nu*h>PVKd;D6swVi8|BB=?NK z+dDD?%|0eIli9mAhbbjk!X%~Vw0va^H0OwAy>uY>bmRKDXLaV7{8r|3?R>}!{YBmX z+x^rzE%Pk;!T)GKli33wv@Gt5CH`-a)yonx0Ta%W@dZn(nfi$$=p4Sp@ffteG)ssW zaUgMxYA4nXzJ#8;yNuQ~IAU(aYmZRxOZYiFdj6x+AdH1H?>^magHx5%Jhu|2eGco5 zrHZv(=|cyWU7Nm24VY547*cH#f{yuRdnKYLt_fNc8d*q2;w zn0`fG#WuPL`b?WGPk0_hvi743mCt%H{^Su+W6nY-^JP2U?B4=O5s#}r3A~4hg}ow@ zR1?hdIBntaV?EGbxCUEm#bb|S=4WruE}VU%7WDGU8yr1(bA8z!37jnd-SX)Sq2u0? z@-41aY?5hQkTuwKnP~*JV{J!yrx{E=IpR4qx`!B_JRG2wtP+^A*?eV$CbBPg4^@LeOo9@{Rb_)Q3AAhk+b@y3&9FsMw` zE;FD)!{^S6>u%aXs?dhcbBnJ*LjJA$I|uqPN{BS`*sd9R@*8Hlgtf4SB%$NvF^_-x zy9Hv4OrTXyr~RM+2acERwq9e$iqmf%S1a!fLTag9%;`%2el; z5rxa&ZYPSt@bYaFmUk&wT9YH49U=|6Unr$QdUIGG-ut>)g6aH*X!DpLRbd{c@t-+% zKjh6v~CPU$!bJ zk8w(kj0@keU?ian_ z57pi`KkgB3#3B8wY|+-&VA|7BAvDJc`}%%Uy$U;Fc>N~d3*A0AvM*U~oqaV9euR}* z>~t}G?HMhn9Z5{z)uJ={?pz0<&o@{wEs(u65;{Hy+|f5qg>JdQ8*RQs{qfueleVyK zRj4O(3Op^p2Zhmx%(9Qagubb%b_vlnK!2R0nsVp)|Eqlx+u4*juf1gOvdAnI(nI5g zd?ar3+t#K0JxteE99eSRa-s}of4rVM_J;%=Ari~mk|eQwa7bv(P=o3E@r3JkqD|Up zcCTXBiKRrngmbb|2m2~ovHN>{wxw$tG(ofq^>;j@9jvD``{v&{f&bM{b}UW|y9ef$ zzJ1V09)+H|&(pUXJ+bb?oY+SeA*gSdm)Rrm0{fm)1NAqVVeqBJhMNYbpjJrXk_Vyw}tD`}ysqK3`OoEaD&9-!(bx^bNX#Vs@qE1i$-gSH{b=Hzu_+_XbDWksR$0>~Gv>LE!eGR&dgW?&_zuS~E&6~>gTRkRQPU3)dj>&ejKA0Ii z!kKzS8XBv9+@kJf!krlX#I7hPZtQHQC{j(vyZo(iH}Po<+9B(>1zoJ-{G3?|&1ab5oF zDXk;r^8fsuK7C^6IlVzB^H@=GCT$VX->*-S-S#o+hCyfHsJ(-PANJ)uXI-?LIUbT< z@I6_+d<^H#N}rW0v&BMR=g~8(oPhpiyKQX8N~q6jH+uA&;LYBXyv}B;)((vgF&~@n zSX1UL*9^urlUx z-wRhQ{ipv5ebxN?4%7X(Qk81drCIBsaPN%{vOfhI zf|j>&7ck+{|D!$4JvK%gm1W`ylQ=AYR(fYL)+veQl+yLgmvaA$&uh8m5y8uBs{Qiu zULa9lrFCHWlAkA-{BVde`A~WJ+h?FSFM0deow?4ySjoh-cSAfll#=!IrQ>PnKfAka zhhjlxz8CGV7ZpvMm>&`9iLZi)(yb?)Sc&KOo%#-`KcZ$8!9#7k9f18^2&Uy zTggYcvQ7J+<3!o}B_$zq@c*>GfRdjyx-?@W4%evLWu9cJ56pDEe}#)kg0>FxdHyHm zyihoS=Z%?(8%11jNj?Tua>ntX~JNSt@94Oby%)+qdT@jf$2G2bA8?xdv5;gd$4lnrP4h6|84&P zvNhGuv1~bo?u(D)9Gzvf2kRv|QtNwPVv=oYjL>#2h}%Vr`)5@pntpv7cn< zIaw744RI+W>)&M~?a6OX7Uj1%d2aFf1?nAeNGg|kB3+8T&Y8AR0w<7i z`1XS`n?j_^Pv$N=unflq{U4ZozmF90cNe?m2XN?=))n6UO#Ghz-_4$Q4g@58U zFzlx~#bc)ewHZ6NOo=h0{fW%M2#D7mOc2*OhC^K3O}bM=o_WLf=fz^|H=$v}&TjcAO(<7p_=vjR7)+8<{t?*b@M)26iKa(Cm_lbf4E$$YtFAO{ehib(N-Y61y zaJlXqD-*|+&~Z&ygO>4>@%L`c(J|MqL>M2GPq$Ci#_{MOz5X^YOh4tHo2#-1ng?8) zr1moS0x)Wu@9tA&gV76uyecIOy9#?Xa(=pqrQwGJddbm~5!{y;MaSBb|3OUHT%7U7M$e(&Y&K#;UNRu7|JO0qQkc? z%L;6JB(UU}#!Ku?YB@N_&E$78&684b1qqh(iP)eNry_ef3yR%;%ZFxL!#j6HRk7C~t0mU-!@N==L}lpSq2a z7lY$(Ir=xgj0cLU$MfPdOney_wq!{+m)HA0?ynvmUKv{8di0a&Gc4Mj|Nm>h|6k98 zS!>SN7>j-X*UuT}H>J5a=fRkzLPD9a2vC#GTy1)N4@xdczP;8j4grEvDkTc%p}O4c zUf9Ix|Ly1IS#Q5%eD})#bR7h}Uwhi+$3OiY=)29e<-kK$tjw_3+_cz+={*i}f5ei4 z7=8+jZtrO6e|H%}N(xN8#w?-t&Pi@|juni3mxViiyVl18kke3mn{quAdraOv{oec* za}HkrBA~qrMA5lXl4Lvtawp9!ju3p>Iqzb3QKFeGVG3gm1|3s$kC@@{64^axkJ{QM05o z7?FpXP*+&XoCuE$8T9c7Ji+YNm$wtW7<)t!kIXLgvmD%9%AL>{{Bwr;Vi^S(sAS-Ed{^OZ@)8x#UGX346bb^wV!##MVDS z>Yr}*!+>f@nCUiYNU$TdeO;WK6!Q9#~S}S?uHD5e-1Rq@|d*JUm-{ek~(S40=Lz|!H zI5cCA%%X<|%2U{*dGaUSK@dY-@#RGkbv+Ih!@dOYmUncW>WNFDLS( zbFLMJnhOtM=L?^=w^O7@9P;bhootUYJTrzO$8_#&qaNr2*seveh z=SkbeIx3RF;MHPgaMYO-n*yNubk(AB2I`p6{f#g7UM}Fd{U7@TiF%DKou@6=hSG6p z=i-{FWyY}ZYmwjndZKPQ{Tf^RALSM(ld*hje<2^HQ_Nhanx6i5J7J2g{>)orA`hQ( z_-s(j6yfJo=CnKAG35d32P>!HYaZC4rQ&7be+ee3t(}#;hp@vPuJ=zd;yPp&u1~c1 zt-9ECJYVPVJ6r_L3W{MO#Czm^Hp7ueTS)zCDK^#ba$UG zg`sZpq4~2aB-XogEIc$J=vb*$eKpk|aB!ZK)qQyhR6n6wAOHLoDTV<&Zx3EV%IjzQFaIE8 z)w8wi<0;3m{(u)8xzC5?Uq0IS(#s+La8VoQL&kmw@hczy5$<_TVwut0YDQDJ*N7G!xS%5F`=GR?|ARJXvF zKSiB=^xl_|--Pi>-z>Lq2E+XD37cE}o!ECG@4M1O0Mg7R#;;frdK$yG#JUftwL@vE z@6a1wUHI*%RdiG<6q`tAck-?F!h)#}ez@#Ec+$?l@5ii;LK=GR{E+VgE{8U@gPKn-k6@?tvD_aXeK=>A@@knM6Yq<}s&nK(q!TMn z$ew8)-0Tf)JN0UZ>Ly`S{OQb(l~knN`D39QY=VV*c10X^D8jfrZ*{|pSz=tjvGPAu zONNfMVW*29i6PxDY@g@R^H{$8yS3SE8K(J^bSW|Stjs)4i;Sd%EMHIJ3~Zd-n%jre zJcqRJI|QNX(<#4$x683()$NH*eqW)7@;j5&lCjRAb9CzBk&RQ>bYh3q{WENgc621& z7~8ys4&(ZJ{eXqQ#S058%k+hfad@i7qdL+AYfm=yI-C?Ead}yde-*k9lZ$vNx35@4 zVtrlw^T}GmuW0plUPQ0{^55+z_WRo78-~mJVe-PE(`%FtG4AuJ+r4g$axDh<6nIe% z7XIsg;+(uYChsP645qxhY%LFaL7QQ$q_`Xt-UBr5BzfB}O!byv++_JOKhs2@Z+f8` z680FU60>DE>aJpu>(Zm{VinLW{z%7d-Kqbv9luTqN0l67=zk+s#82i}oez{Pj_eW= zCG=_HKB_6&@clE-1qNk1{AGDUNIV-Y-L6Fu^=%r}%a^o0IRZ&id)F2<86joJUU2Q+ z9Hd$8n3KBxjp;cp_A}4!I*LKB`=orcH)Gy{{Xp-QvUQQr8KUBI;66hK0Vbvo{d(cW z=pPL83=_#D{|A0}u;|wlc`{bNtB;#%VDOnBd!ys2qk7wMu=UHg#cN_9{@45;o0Uv; zL5TTR@Ohscn-x~7@tNK@+=v<5gB$vmyvNL-g5q0TjB^(UJvgs+_{QOoeqranJ52Sg z8P`>Xjgk4xw4mGS!a)2+0i>k`{dv2w92(;@I4-;0V!XGmwNbN*oq=y*@4GjrjGuBt z>6JAeQ0iD^I7m?K&8|$3DNBoCQQFYU$ zH(FxRWEQRYaJhH-*f3yy;I8XA91K^{4h18u?sT~sHDyj_tMF45U2RR`pas;6jC}iR zH@tT4;CO^1kG90>f9?SKA`fR%xh42QM|#Y2#1)&pShIEWHeq|MV}Wo+9MpC+sA=>t ztqY{Ksy9x8p-YyBR}ZN-FC+w}jAQ%xWootc-?3X@_E;-QFnw2$Tm7>86~P$F zijQ*lwqw3V>X`$z?AVlZ_jLbmBEPLD-rM|w>I0_x#|t0z++Fhm+Yg@3%+9UGkX*;B zE4mnZJW#C?vgOdKU0DCk#HX*Hu-EH`m+?)vE0WkYJ1Rb2vlm-~{n-u@aY~%7G`jiV z6%mgg%{aYrM@0HxJ7n6}UW>sqfbOb zL0?SrN7^1 zJ2CNTAZ>;C%Fs`4#Q6NQ+PQI-k;jf*bG-L;o}7ZI=NS$s_6$O=fv?Jx0V8h#=LO!? zuB(;=dTq^|k2r5f2!M-K(D* z97D_lmFs;@>o39J=6u7D6uIIrbVwNb_MriuM6KR?ek3W6{y~+Znv~C z8H>ZWFB=^sbd}-$v^)2MjTP(iQ+v&==Rh0tH=0G6v6bb7y6lH#bmZXn;YUvVK{Ij%;nDc-7tUL zfAts}8L~bHJWrQksw0W@Zc3W%$tBp~iSeiY<9siwOz%IoauxKPk@kG$dH{aKf9t(5 zy9vvx&Yo$$AB(+fw{DhM%cxh5Z6oe>;p|2vmbzy(-Ba7y3%t)2Em)9A9S=ZH+~LvebMt?f`l&1^Z}VL;rm1#t7_PL`y=%QjNEgY z(|Mtby$12)Z&Xh~YtJs_t`P=~gB`-U>8)xRP@lo=e&A{@@m<>*E6%h5Ms^633z)Z=Wd3LGAPsh22*Pd~JS<>X?WX!_FqLynJ`p^X(jA z*OXfrr@vLlKdV2qz5Ypo(bT{V%4)UP{M&_p=i^i;SiLRQaO)({e1Av8WE1uzd7ggE z#f!?g23`R{gYFz3rpDhw?c| zJY15ghcj*f#egidZQvTv%1ct;eq!XiLTy)=kH88qpp2%tWWSDt@*96fCQUBDz`=)O zKCX;B7#Nki=s_A)A+x?_*{6~u35~Y$Kc-Krz|6g?ed+dYWERD1W72;pI4N>mmEW3C ze-uYU40?a9_lB`|JC%R3DC4g}@5dvb1#ol?fAO9tC$Xp3?dH4A+gSQs#60nV4Vi0E zWcrW7V(e(W=rHn5hs+kCG<5TkCNysmIAeU9pQTBN9s=h5*egTw*rm+ z7RU6{imdmxL&q_d07*c$zv*`IIv8;(bDK=1!-Cnl>&20qU?|UrTljnhwAWlM|N43} zjGA?cZ9N)<6wZkEpXC|22#H;1(8ct}Gx#3J=cMIO2;*19qqOfXf$7fXi(W=QvHSIs zPnYvn!PJ)+F&XFESh?MQIjON8TFUfx%Zi#5eCtkVkFyndZFwJ) zTDD7A6iQ-;j#X7xs~C<|8weB?08szvEq^2UO*Q^{ ztedAlIK?XuL#$lZ8AfbKX@ALIdO!?1f*m(0o=w8YZ->i;SkaL-p?&<4A4}2oVeK zz5GWOCMpIwG~VCD39f0@@MG)XcY26liwoF(j;<~k-WkW0$ zzRM(9NkB^vw;mL0|j8zy!Em@|2g(Fd4D6}H% z>+py?PR%dhwQGP$j|iypkvzLq{>dvqN>^jalmk(}qJQdX%9&Wte_aoaiE&ojGMMUe z!AS0!ZbyH0oFH{Y*q1YOR&hc_$Dyz48#D=Mv+n9wC%*5f@KLuV*3hJ^cKXM@k3jwP z?SlL86QJ^MEmFFf57Xna_bd~{v4idM-L$8K9tb^#=b1lPW8+A_$;h|A_R-nR4|hpB zA+1<#Lt*v}s9jpSso!B4aSm6AE1RzrBC{%%RFv;X#BODudPCI(G8@P1$LvKF&=+xc zk>>4hP-T8&Dk3%uo5v2e@2IsR>D^y92UIAoYy$;!{X0^9T697dqVe-F%l*5 zZ2I-IOT`kqyxmQ_6e3Bi^B)$u1*Bnn$i=C6t2!)i(tc>O_7P0p^37hd&lj4*uFynw zF!cUl;-}2#cD`mDK5G88{|*<^d;6bA@N3mG=Y3QJ{zz2iC+rF8j^{@$&Td8OrI)fX z=1g^K$ZVeYn0%7@&-v10Ar|$=;>^E(C$l#8rRcn4BeVW4U!Ip14a4=VyF0JBz{2wz zZMSq8`Cue&mPNAC)$JsnRwZvozG0l6EScfCVnpKEso8MEd>o1n#~pmOeg-@4BxL;g zEkN|=N3IYzV;?Abx_H};pA4K0+C9`ms+QY9_Z#PQwLeZ|7R{N6LtW=dJl5Z{e1!6$ z+;{Ms{uN@rPK4OasFV_U6Qefr1z3A4;JWL9QO`$a-fH@k}WWn z!Fz%3+g*cx4Z~2%UNIaePJyw{Ll%P9w_z8*H;u+h;ML8U)5=yZ!7y8LvX(Tk1jbyo zf)mO)AnnWEwAt=-tdaVZZa^aRh-x<9b%?XO0`s=dM*f6}Vz%fr%}2?Xv8V8Bt<l5aNgcTn zr&*ZTyz*Mr;rkf-E4u!L10Rh4y!ce^O)d7gLRzAGXwSmA44izBLeU=tYz$?6buEF&tDgGJB3et*goz>YWbb|=UPaw!vZnbJ z!HcA%9LPUa?@nUf81An5Di69`^2eE{!eJIJ%(=nX2TAO^ zJ2&^ReZy+r$JQSM3BSGe!FJKug`3cMR`vb^VQviJuNO=Pkv)Mjz`%dm zp~z&$_jKR8pE{LM2&JW?Ll1<{0_}vdF-<=8zrPzA70Ly-{gNWa^;T-;cC&u8T4Y!% z8p4$4MB;EtK9TW@&=IZk@v^G7BXF_Ni-HSMamm=z;k(y3X$S}Jzg_0t9f4>z>$xy* z20856KWc1Dpswq7!m(i=q*vH_LteaiQgvtYMQVYI8F(tkgfaeqib&2N=W=1}Xuw&yPUFvM6N zUUHjD9ET!#qI8vC;Co?dT@5utkL#}=MWLjdG45|Bvn%~Pl?&}OUyH~K+vV6p;tZKtr-=>mc{Y~<0r{VuXjBs+M85?fvp| zh%-qC2A?ciWRYA(V&7#?eNnOlCh&YGYr#Wkd&btch3l{1Ur(FW{&p2s;mD6iIudnB!W8(Cb)8+Ei`{y7vGHc`g&lX2GTzPq?N6VC;j#MXR! z*umgy;zZ%W`CbzyzBER87siY4A^K-}oL+OfcB=u40FRe{P-u*pmEJ=>45PyNEN#n z992s2amK2SZ`>2oi%sV4@w-f~wIGsPVe!Ch{ zCzfw)e#cnnNTF}OIaIX?znBHaR|hibr(@oYnysutrZ}iA%Cr6kH_+3}L^Z6|!$^UP z+nQ{q`iVGpc9Pdz^E8R)obj1JF&pf;k?N>l6o8FdUt(In67@4XMJ%lMIhw$LRPfa= zPG&eDb$r|ART|KWC=Rqy%z zPq9Ybfv>HD@CWlpw5=6HH{s}aNrS68$8mP+&+Og5M{rcz_0Ziy8<;%AB1>au;Cnb9 zJr_F5$E0h4Q&GzXj?*vWs6d{4^G`GA@z-)*W8(s~)2w^h8tbtvN8nsQn;z4=ALKvv zTe>0+#~Um&4w=p4K(9`dlz}o#n5U|q_A|tRFY6c61V4~jE>*0}_{hKqada7f8L8|U zcs|j3dkf73uYwFuLWv zXHVKC>`+X}dp&Rmy6a?8G;caX#1)$Pf$32wAN+NG#)DCZ9sQr{r(8Cl#P$vL-fpaS zU_5`=nsROlww>r(m-V^7?Pn9B^xr>#O|q_up9dKG zG>PlX>aFSA3Xo`WR7dI?;Wv$}SuH{9C*o-3`nhc-H_LI<<6OPxEF&IC;>kV6^Lx<} z=K1rGmiN2DfW#uQe5P#`WBtQKWc%H+Czp_RFH^!jj7e{T@&1L~C;V8}xnjot*Teo7 zSjpV07wN>i7~^27YvNNkR}u?{VrFNo8mgn?n#j$-jzOelE%Q`U}b;-YR;S#~K z@9eXg*1q~5byXf^%{^>otUu`R6BIZnLGT&7mzk`;m%c!>=k6Vi+mahd?AjM6If6ro zd1rgu&Tz*!XqC0N?bXGUcY*^(u;W6-ICSrQ_^^+!8am&$-r)9PDWe>&!V2ZL1O>; z#wh0J9_+JO=YFQz1*^nE|8QjxdJ@xfYkS|(PvJ=N&i-8&e?hCT`v&7}yJ2LXZS6!X z2}Vt}Nh~{i0J^m-HeRk;gSBGI>d2djc{J@iX_a^*9wyz5-gR4sL-U`a>j~c(acCU8 zG2_)*RQPw_$U49MXehD1N(=cLwwGbsubJ>uMGX4|dw-TjzS+A58@D;KQr|PKhsF;) zWKS0d=6z7HY=68o5r-UK*}3DlmM~6e9x`Mb?E#07ySt{A3PbN&TKJO!Ka$`>FS5uY zhCK^3;hAMWBz>TqRf(;^Q_KS^!*+wT)=1~)O+L+T%Sb9P%;4wU(@=zfa! zz`1MHHMCon#QYhl>Wln7f$dLfJfkywv8!qnD3cYC_f%_}lIj17cW#JPe#i=?@iE{1 z^IL)X&|%~3c`2mHn)t11SqeY4SUSoCi($^bk^@Ua^>KLjbZ@X9Q$9D22TNxLl%->) zsq`$gCPM9098XIOe?5uwB0-bHJg4$UWXzKB)Yz*UbefpG_vj%nlI_0-mUNd`^< zL)}FOtP_ZNG?Lr>@re>#(E zD+`oEJ7ndu44{)zwte6_LpKM0hk4W|d}pqon&vLG;>LdF`q3?kwn67!GwEv){SxGO zcas(KysJp8?NCl`gdgOQj@9A!aU^l8fN)qCG`!#C=|Cgu&`1a2sj2%Xp9vOHCoXo1F!~c|X%cta zJEF1gl*zK5VGig&tZ#jMCeOwtd=j?e9HKz^Z zmvZ?@3jem##?u_D&y4s&u~^oYc^3wM4XArQy>7|d1r=X+EjpKc11MW>7%pYm2Gj>} z^aUde?AX~sI=&|oM#<8Kr*34C*d~_R?AF)^vPZNO?ztCGOH9UVa(a+U+_qui}euSC_953(75q4*{m0ZC( z&oH3oX5o^GzwN8l-)CDTL}HJ)*u!5QK+G%SQP;{&#{PzsENAqvJq*?Qdmh+4VB+zS zS;91qKUni`ozSW2<6+B28h03 z@O`1>mP}Ts!64456{N5!wvae-rm$k9pJOuysq5tIU_XShP=6 zHeQzwto8w_hC)PEkbP>~h zno(LZIM1&Ky_`{3#B2s3=Rt(6sgDVk99aI=rSX41udglOV~^EzA{4z*>|LbY1>H1{ z^YTmIGy3hfxrn}3{W0h-_@wB+oWQ?^^EL&kw$DM{%{^*$Vjdw&Y$S88b5{FuMih+BTo}1>^#;DOE(5f zwBU%?YYVm_M!i*NU3+(`y>~I@vWJJwy|eSbQxj7Uf%s-LoQ_fBAp95)I?&Ny5n`^KpJHK*xXmTAW32^NUZ;zcO3^$ zq^!)9H-XkoqxYn)6S!yB(|pz)?a7$N*}+<*V~li_73{9FVlbut7Vo{ahQ`#zi|>sR z{FGsO>+VflAvmD$YK6kz=S}a(P+Zl+;FbNg7YwWnj^Bjr7L#)vE1C8FrVAT|DtyIp z^o=nkEl3|m;q{N*KZh|>?ULZTXGfr8>AZTAzXO)rIpWEuf9)(a-o9jGL#TFdUaIRQ z0PWF5lYv&vI0!r*4xb0Hq3TA?lKhv@cFZ^Hx5#@a_t?=@Gr`c^g{(9U+Qfl7Fx0|w zN;#R(&u`1TLlzLE{q;Mye7J0!3myHuQH9&xp#8yznJ6!H9Gkj@mfgHKEiXGQf0_=1 zV@2|&(R*Q9jJ$`9CJ?>+GPblD#7eh5 z#kRiY#hatkp;xtKKIOzPME#0i{C%SgnWgQtebSdo?EB{bGfA4@zsCeU6Pph$f=Ik} z$$XhImYd!*n0rQFH+NUanIaN_WEW3AgOi#|wd~ z7It;R(WRL8rcT10!;bOXye+RE?7S=j-Ske5#crS@XP-Rhg=T4~P{3U6S&%j%;@9SZ^dR5}Qo_M{+&GhR{Y;#H! zn~i6j!z31`EUl);(Kzs>+BWs$JoJAovbv*{1yoc$BIvsxB6&0P2a=m{nC;~w;e)l% z*-N!$eb|T;pBj11&q~l{=z2kM(*jTo=WG>@J^<&V$EqYfDM-r}wE9{75yB&{J)>Df z!f?%F*Zmnpe79Tnuc6CHhO$;sy>--g*rr>sf@bsx7To9S8teYexW4MxU3X@ zE?s~9$VVQ`I#xZz-pQ!r0gaosSslJ@3-slw!p=qvT$#+iy(I_U-Nr^44z(M{?qbfd z;^(?U+o4j4WnMOUEq1zJ@N2^&GV6|?^q0fCkk-0W^~+bLcsz87-OYXY^#_()wpS!8 zG-K8mu}60&JfZJbjMc&o!Y?Z=7y8+k!UYSD_&L*VGxREPFkHUC(6I?KyE*PYTz4F5 zlzQqu?>ED-d&lD<^1X@u>5%oM!_rTop4B8ZmS390qxN+1Y8G3_&1l)OM}SG^4W@Wx zEDvwp3>|jwwKn&DhVsSzl9d;U`oQ!z&%X&?V&H#3zS9(6f9yUsbe>X|dPWAyr!CP# zR|&q=lmxA_ONt-sWxpLz&1)y@ZPQuHlLId?d8K`=vYQ?BmVW-a`_K&I_sXb0yXJnZ zCbKtbKYPN)z@M@7>+P^pfrXes8`;Nh%;1{=CEaux-HH(pg$13J8Q=X0yO#Q4^`}1p z41HA+YhXyaf2#=2FK^nq0pG#go^@%9;|ienYP!`m0V|lgR-_ z;O5VgMK^h|qffOm-l-7^b350*etMk5vc0Bcl{>+gTNsOab?;LZ*16~gc5Cd$VPJWm zUy=;H&iHi$}SEac*GSO6_acCy4Vk)6Zg{M1{a_T6RVov8 zR~yDUK3dE#zV!ivMlfcKKHqR7muKv5Py?)s1P#4NHf5x|V(e`Y%C9za?KmRPt zz~x9RhN5QteP^-kJ(n;)i_t&rwNCE<=jjwe_ikXad;hE=AC69?A33s*zy(^LM>hC2 z61<1D&V_G=w{vi^F}u&*`!R{_(`ZuGRUJrL9FcoDck6+id$G!hfDzb*M&s(+3jEy{_H0k?hc>A-Jo$>I!tOD_9+NOdTk0 ztBw!;BEfKZ*q`g3mBc)JQgFG3y@AXooMb{tUIclY4ov8>6FBJ1s0pwvKa72R_S-k! z`tZ+w7(SjBWfpS<=2u23ZutBa241sRYV>W#q1B*rTx=Se{Xh2BO<95CinTv{3wyD{ z;ik@n8-dSs*I!(F%(xuLh4((wrJ}HCE+w(~%N2-Lm^%LOsWT37t`6MA_8X?hFLo*x z^&^F4V&c@S6BeAk)Ea-Q3);hVhh$ubux@m`>i7{a!k>xr+^6dF7`kl^C9bo|!b&Rt z{g4g*P%GJ${B%PqMlCWpvAe$kx>kN3a(pO7Vi%uK=-#pvDC17AIp=G^`Gw7IO^ZwL zD`{V4z^D{fe!cJ`(c(U|leV8$3HQc=9lUostpSR9b^Kdj9)z*yiY&)$cp-W1Qpzz} zKD3T$c!;qP^}ZVp(~Z?C|F(b6s+t%_qFz(a-lYX$$Cts}4SPG@*Z>&QDka&uCgWzNcC`K;?3?UT!@kY0 z(+*CpKzh5_>4<0X6x@v%vZf59g-hKaZqGo z^=p0u%u=WrB&ia4j1~2toF%=?$!xFU>M1F^u`j#X=Hh-WXj|AVeb3ke29&I0tW^kH zF+S-~Z+c}IOiHuqI|vTr=nHeR`JrKGExPnV_ka$x@8Rte8(jh;AFr-mO3s71bh}NO z7f)fPV7zZD%O{M}GYj$wIES6f3mqz)HUGQ6NGxa1yXEv~0NwMK#G7aOm^oKl@W4JE z%eHZ{n&pqf*vv$XwErSl*up5vtlqd1CP~5Z&#tP$ zZqGsD|1Td{56lT+{)pJ^mmykY7Fqx7K4qqQOxXNJ@`P_cjq%(;Qeoi8W<99Q zJeC}P;v7&|4eivY5~13lll3{PD^!aMHQf{;_}}^CE0?{xx))2ERNrn-%mn1V60mS| zH*{>ca%b>25$77b_fx>kzyyXR>b*M682LijdHGMbW!-)lebl(~ntvmI?wG}yy#>lbROP!I79FcjEKtJ z+V`Gu{*rm?l}el+?ZF{_)ynr#k4PL|TaxVKxiR|yhb`aGZ6GgZcL>=9#QL_7Gk+Vz zuv4M-MdN{}0!yS`DDbNeX41!nfx)Wx;XMbT+34A-RdG!7jB#Ddfwlbi&WfROniZVp3|*|S$@<36JCyk^xu7pJsG;kIGK59 zneuDGzNM^)zD0`F2a05ug6k0y&In)4)xYPBZJ`1DF1I5gv0ll2to9m=XzeB|z2_e9RCA-rD@kUK06A-7l!mR$E7i6|FgWx zz$=)szc~F=mjjf#%5KRpv4)wveXAYy44C?vZ2eR^B(PA#&@LggJjl%sVsI+At}LJ^Ge~;>jjJQO&krPX7o4`+eL; z4=Um84X+~i_y5%Mf&No^4hN1B`cji`>RHrESb);KT`@5AK2}NZ&`zzCb!` zU1V?WXUhWh$tr6Ldn<8Bll1L|CXp`|7d6_PQvrlA;x3Wi=Ubl$#U{fDZJdDaqA zl~s&+isMNt)WaG{KotmQ4_@sB<+l2R&dTjL@GRimv_K%_d{14d&<%%9A@%KdQ-v7! zQJ(DM*jMg^ajxyza=j~YWb=metCfSXub6+l|1NW0+Kg-EsMZxbY~9e%(xb2oVlMeQ zJMFs(1%n0Vyc4oWzx_>U>Srn7BHok16AbRgDzPJlt-oaX0SD};GN7qgnqiaE4ZHlOjQtnsMNi*Ig(qW^9L0}IwjZkZtNJx- z5pj!JEp|S44mJq9=%M_12T{MNzsRPZ-%}OxPKWmxhO=T@d-?9|=DH-7F9jiqa2n|o zQ;YDH1Bq+lgW>hg5~N$cZYYW0g?URtk7wH~fyo!J(|E&L?D%w;^Ptu}oQw5aa6d}u ze$2H8F4bMM38}R-2bl^&&!Ru<(F%{SWzZrWd1=*(!&t6qT#|1j1sR@Lr!##V`r^zh z7Js<~U&eN^dW~;}uF&FyKddK^d`!oEZQ~4->hn#`iprz1IW5WAlV4zoLUK-Gnv8mUYEL)Sq26xB-XZ~vSxWfy62 z27j$9MWwbJP*Hk$-r!9j3?C(p@m4*@;ry6%?LuOon(C9Nsx0@$=1|i^C)clmqGzu> z(`E>qGoIhyV)fEWq!`tf{CboFLuWcF6>f0AG>hxbs(qiKb#t_(yjuXY3CF!RJ0k&I z)7?TdpUbdHlU8n4%^Y88FKRrZH97-R%J%tnUOq^lopIP&J_w_Vc4Z$P?}L;HYs08r zTag;AE?Jh(j)SEOK?0o0(Dg8xN`vj_GF0*1XuA~D@!L%A&>{H1wAB&szLscXgKtN% z#p1(AZCNAtb>TUdnf2|IQSHW2dA9M9idEQ{df}J95)V!sjViTM{)HVJVRbTJ4luQ| zwXbn>*ZVT;EE{AqpaqdwoGFdZ{T8s?;%-ktgg?f*=3f;rFhc64!(mdHR2Wk23cs`= z9Q#bSpZTWK2=RtR;X$R6(7WV!fMpLK_Qr3a_ff*3#iO&Lf{CSt37 zot?*C_xF8n2c4Mu!Ie!%sLGmYUq>oON9Nff59seVKXGFI3Jg3bSRR?)0+lX}H+eoE z!VYn*k=F)^NX0za+Ap(MG~j15Wnv5C@y^cM;_3K1gZEB|vKKZK?1@#$D#kAF(V$fZ ztDx$;^77O?e=JE>{r&2-Hw3!x{dHOY5)4)F4GiwRi^F~T7tDBhp!HGvjnC0tIFQ!= zR`2UYm>Jr4DYM%eClhaD2x%0kMpm)cYvXWmjaBh!I)krIW(zO8er6{lK8l^v{X_BR zK0_P-kQPcZ^90M4_CL3GA$XvX2X>t3``wF!6fa?+oMpuRBUcnMJr@SmZjs(vcV9qS z7hlqnK`rRpM;^aFaUXwVC7LLX9stS`)8}*F2wmO;VcS40-a#A`TAg7Kq|S8iP`>oL z#F{e2O|fIw3LWtii*T$oBca>DJ^ zu5J>$uSdYA(UTZ9|EyDQx)7&q_8fbc@f%xiHi7jnMt?!qM|=L7+BTp|el;&R_y#5; zH#-KsC3sa7o3nEMs$AF|m9%O9Mkf9m<2hALH!qTIo@DuQ2qeNh}A`+Y_hG_Y&p_rKbKeRBen@fOVbuN6)naV|FVkez$-*}>BtFqp1pzhk={ znPoKW&U*1`;{CPU+Qqsbp#3`Sz>&$VSbMc$O+yu9ePCNd&>pSn6dX(Xo^?_&1!<)- z@%iz$fnH>OK;N|r1|-6T@7ueOxC%X2?VMQ*z1!cNFy+d`K&OG=q^y_!>_4i<hG0i1L7JEOCV;4hBlJqq6P+Za1v zKFS&VbRLI9esLXLW{F)7?|tEQpC_(k3ynT;n!%$Yv&o&yThr%@$;L00H}KJz+ST^6 zggaxb4aYAC78*{aV6Bm{MyUf+-4N*BmwuP7T?wP6TaT}GIFHoA*B1AK7u&3xcUIU;`y@Jd=_ZzYt`m6JERF#&OT2EqDTVG~m2Qni8aSjZ zJ||E^=rauES!P&>SitzrT(0G7Ho!0siz|J~4Cv9;lWNrlIO3Eza`K5O(6ySs=&%uf zP*<^)rsRSLnZ<)_cg2*@6&SkqV*W_ZGp6q;NKv5t&SKJ8#I|1tmn`mMhq1=h=O&LB zL#^ZPGJ$6V?}^%5_C5aRKl9*lzC=!LE1kK%SraL~>%${9C~-A3Nti5yR!i%%_b+pi z*seZnwdN;uvBv%+@$1^Zf`*k>oPT-{c}ltD?c40Uk3$$sO5x*gjPC==0bk=uX9gZi zVl5nUtbF>{Z#?t$Lo5*cTc=2uLa$XVPKX)oGTgz8ld!!em2^Mc`>*dJu?Rc3=2igNU?zmb2B}@D5XEA8o6&1?Pizd8{dr{dIjr~Y zKFhAw4wbb*k>4hRVT>g}FK?MRv{(9(?F8RI*1X_{kKNf2bGlAG;tY`&+G#H(BX6Yx z9aImhkjNC&T`{(G|4!g=Q=5{G4(vS+HCFS%dY{H{&?J!6>{taFJkKx*6K27hjLAcd zjeIc8v&pOa@jPVs7TlkVPKTzIcd~hptAnqJy|RR5Ee=`~ivDmY!F=nT0cQ49Sm&{X z`fPD0EL=b1`=CRa7_aer-doQEV=brk7$=J`j%BDT`R$B>vD41U?ENRO|H?Zbm5&in zu{zXx_}Ve(eN_MT#_>1Mr6@bQQ{5is)0#ziL$+d;r?Fstz;ej3yFDuMSqtkoY`xpC zBNZEy70H8VUt+(kM@hJM%HRFzeVnmN3+4y7UOIK>H*`r_#MJry6F+D4+st#F-y4qv zLib)>5?`HxaO4sYUER6L}^16=+I6zO3 z*_-c+nD{Q(t9E+FHQihi$Ixl|rklTr`JuV$g#~9hcAwTvxm$h@M<1mZ7-ulmcSl-s z+41H(k0EWHwB4o$f9;cp$xoMGkie=vXGyAll~DM2ZQ0i#B0suaif4V1oE6h^+BZ8^ zA1-Xi2IJxG&cW8F`!;Yh(M9#*9>y)LM?l;LXSl!C@18*T;O*24;(*`0>>|JpwsM!q&1 z4&vMvVYX9ix4`sS@kyhb1P>-pWAf>-q7gEmlm5`?vbR{LciEBuT`NqupY}-p<_LYC zs?n;#l!!Ana}S3Fq1(=u5K*eYcMO^~$NsBjA_P#QLFjA8dMEG$qta z)D3JptMGX1{ZbrBGJMPV@*NiR*362CJ;U)6Pp68Py~3Fj&wG;BUk4gp<&lR%A$Gj> zXu7kq2u9SS#AclS`T_Fd%fp9=@t|zFefsj)U7(#edH z$)>Ny30`~SDUCPI&&|l}D!cRU-r0^5u`2XNj?XZ^@w=bPS;qW^ZoPmFORgS==I`D{ z;oUEBYWaZ**}>a?{i$6G8D@WQBr(OmuR|ZkNz#^~_s>G7xyj?05l$H0bf$WoLg->= z{1iDEs&*Jgeuq}dd*4Kw=Wh1<@{GJm$i4Uaipdf!_@2poWx$Y;pN@YLB~J;?_rg$& z=a8Hvv+n#Dxn9>3_hJX@MyX)8d8BBj_wQfFq%#K{di6i^!u63dD`QuGwHEr=w#3DL zRE7LslNaB<^n+ewOy5^b=z*nrrCh2Z|A2|-86F+2W=J)R*}6F168p3K4O9-L5igO2CZWe zIIz}VhJAv_n{Em#PBKwr+p?zQT^=PYuo1me5?4xxeuv7YOaIVft_}nIApcw2sbzuu5kBrP#w*R@|`L$57>im-s2WEa$tHr@ob?eQL zHe5F6wDmQXemU*-yFsZT1s1J+0%9h~~Z!qCIO zxzu9$3Z(?>dN;ITCW62#KCTe()#}jwJ5Qf1vHC{rf7DMi`O1zY?6*{7eajz=^d%!V zZa>Ju{_iJLhhLlFgS-OPsCpJ0eX>DVVI~pjw(|2HI~m`NogXPX!*(+A=CCa6>f*!0 zPvDPsz1|TohJB1hDTn>dd`=?$bbZsO9baMMtl&0dejA)AUixw@YXZ~j+Kb7vnm`p2 zFB!Oe3Tw7Wv1K_L;LzoDGl5nAu7}&6;`d91{VUS^WY#)xgaJjm{B6|7DrgdDrE!_h zLf3?D)1SM-Bu>3Pp^Zgg<5yL>~x z_F%pqHGbolBRF{Hi{mK&PROvHr1#|bLVAUR)Fbd4&8r^oyvNw~!5r&J_o7%~mV3+#rHrqzVy1@E#R%Zfl z5qyin^?vzhXP3iJgqx+uwJMz6W)UMWN`j7s&t>zPYA}U|eW}K(ShualN$6c1%-ViC z^Y+~(4E5ItE06ie)Q;ZE+AeY5MzLm8_u;1TXr#zfS19%AL-pHESpj*b`ol0XzIMI5 zcp{0-W47<5$205}h4n%!WpK3blvMcs3(z4?d+Odg2OWKnWobN{pw%C?mwQ*iV4tBV zi#}l&QujHm_xz5sfS5 z_C$R;C=DnurMI{+wU=%d+^~V2z<);K#KV{ECG7Y5Pb*9&qEewnC^fk^U=}6}W|mxD zVhiQ-qT97ImOz7za*khaJ&s&+Tzuy9b4=AAIbglb4Cz7F^gjP2;!IVw0e7z##^SJi zkgvH;H5A+GWf!NZ1Eu`y9k&7k-)LL6^9Ga&KM=VdU4f_hm7ln%zxwx)iLHDiB;fi9E$>hCsF542(t1B09D`Wa(tl)_LMzSm(ya{ ztI*&1oV0d(2GW-DJLlYKh3=h&A?vSLLH!rBGwQ8DZ)-5txiN6Izx!BH z^ZWo)ol$6dS#(m-IvIQTmTWm_Px#RfMmO0Ns!5SpxrP&d27Q7_k5T%^{LN&RFGqHM zcIwBGTJNZ>JNcmSmP(Mzg(hO%hEh3y?;`jQbUyx1JQB>f`S_Z)Hv5}hSQ>cx!N(BB zxdrWyFZG`DWZ;=Fd-kwi=Bsw3Kx=TkK_reInUw?(Vcw*WqY|=ZnxqPn!PM6gP z-MxJr%U_K;s4hB0;`i^(I42l|Gx~+Az2Dt~;c=ef?zhKD9F^o5(*qt%{W{(KgO(AN zhy9!nCvqP5L6`Pxsf?-@xG?9S%>7FpsksHKvpUJH?M}Mxz&NXZ{8y+7fam=)@ zcUNWT>_XMM*L$B<5OJ!alWIdZyEKUJ^b(YjIXj26Ot$PJ_v|rE=zOMG+h?ZsH19lO zqF_((12Yv)B)s^!5jwoPhg@L+=H3ab+MJvP+8Yr`uTg@JSh|*%?LfUQHcT}>s;Xl6 z^Z2=Z#&-2n9x}%-)_^;5uYmR}M(^7y!r!FdrS%ut|HAu499%4tggx4uE3!x-pK%?D zOEGYMw>VKxY>a<-(gw{)5?ha!#*vML&foY7_iYwO-(b_WkSLyf1surY{J>yw+q#|l%j$s= zIc54lih+v*Cd_Ytd4*9Y0@{@QZ~I#DlKB1hou5(~V(QPzTOWE<6MV3nHyU=gGyEE? z_;&8baH$WZJ$4A&ob?+#CQ}6-93uAFEG68Z=p2nhO=~Y7I5oj|UtL+)k2%x3(0j4_ zDZ3M)BUy#t7d=ZzhkFmu9*GzcRUIvZS(ted`$kM*1CAF@ zJ1B-6BC~0n-ITeP;KP=htZ}~B!@wOee`=KVDoT=B?3Evu3Y)<2`Hcq;eyD(c>-ze_ z&g+cp1}M7MA~&XB8p+W#-*+2+{;*H9#D;>{`>zh!;bP!J}_&J2s%lkJ#r7j_L;E ze74kU*6;H;`Djh}%d9t0_WbL%w+{%Nkvj2XKGO54*y1NuKH*mmWk1JdHa%ov{7$P3 zIkR5m5>!*_4{GYf!rXlk!T7`rBu>`g<8wAY$Q*B2w_k6&k4^67YYKfn!a(|(MS;5w zv2K)f?fS1qXs;jGx_{jy_^y9wz2>|&rf9~9@2e+tB`HoTT$I%OpwCD4!D_p2P`_y7 zP#GI9(rdkURX$39k-6L&9?ie|vuU+QA!RG1Jie8`R$LwHhC|+b?DWB^b9$Fw{}{xv zS05~vujIn+S(Ya@d}A=V{ql0fkw6@hx{zx8fbb^^@3L+Sl#c=`N98BwkF_viw{f-A z^^e%TTQ6yaWG{BV(#tC6reVkD(sSK`3?2jwjFl;@giQEw^z_x2&j_8F%YG1;j$>LKmy_Q)VOO=MoAERL7Dt%3J-A312Jxk&soXq9am^n1;)i=fk1M{P3fDNL^$9vYVO#I}Xb!f&^T{fO45qRjT;EmM1o zGL~NFRQYG!Axya+Wc<93fs13m5oP9N^HvhC-{S&bSE3#={k?kOKqZk+($})pfUHg6 zNOav^&n%^X=h;_;%1S-lPn=V(JzRb@fdeslcp#Gn_ z*HCqce&9kDp<6=Di;`KYH3&_sHs@yjI)G()u0K}I%R%#QXVrau1n;yZqf(-+im=;r z4+_q9icH~zCa%x3WU{xQ^YG%%3Ds>(-_uZ*QyyN!#P4O?*O=YKap~`Q)2cCd?@Bde zT%fw^VIA~T!@`_e(XNNKB$kKT6V&TUNLjJyp+0%>-+kxxjD3$ebn#2P-9uY|{Xv1^I4biJkt%(&TGq(mOZYNu}Zk6f#;ne%yvfTcC#JIi;- zjOa8l+K-hllP{|$4MUk*VcXr}aja<3aaHpmtS{M!S~<%;3RPNkl~5!v0YawgxPU3Wd_1a~mjOlemgNmK)BQ%c?D^($Z| zwz0lG=^#cOkjWa|D?#E_2~9e;t^&p%%B#j`@4!s*))VBdiBRL58-B=zp?iVD{#H8+ zPyI(6MA$tt;O#q{cqnXs=Le$>3ed;b-Z-zf0kb1dxsetI0p!LtyS*!|vDa^@0PQ9h6tJGo zOMZ6@=8o|yf8AdQ&Z~?fA*~e$JU<=gsdx^Qn8kOda@BC?H#_GEnGFz_AYCK6Ss3EF zBUUbceiy%p?pgm}xd54yHmzB`w-tKYyxoMp5cxqrwn{{2bE{y#nbzCFRYo{^B}IgP zHz$cj_VkAfEp|(RmU5z1DsOo}IFd53E z*Ny+W!iUw5i*{WNRzoUJ%<}d>%`j#t5as!n@m><^wr`rVA@8BRCc>4&?>M%w@Etux z==1l2CK;t^4eM3|Ri?cX<$}K`*YQ2@vx+#maJ#~P#@Sv&qrc@o=i&9f;ed;waRqrU!SqTWWI!IL|3pMT-_HMi51=0~Bg z%*=d|*+ncd=)NcORstF&NY?q+8F&rU>&%`i^l!zskt^lu2j*dtv~tytkX1O;cWNtXW62aB33PjK?sqi4VKf*^ZU`i8LYf%xH#Z^{~uzji`w zsiIRG;inY(n5=Es(Sbvq!iUfMpT>C~oiLN%2~g-@cf99_5K>139yOYr#yz=xnHl=>?6*Rt~;^IJ^cGIFVaODKBvXH)YUzQWE z40L%ftVwfz&bTkiqI5X;${8~I)8`So`*k3?LEwVA4iR@K2>jX=OkIs#1*9aQHxW3_ z;aHS6M#LXz?nVMEk1iq2|pPcO2ixp2lD9CgYdJp3Q ztOvx*8MrZ&HuXI?uV;XjoA%!3_9yhOM}2XTHe?O0S?y%|Bt+vFz+sPezesdF3 zeFLODxw6vy>UW@synj5-v4E7rWh@ddhoJkq%l#ocV&9uE_{eLV%A^YgGaG;0-m+~a z)87ZIA1lB7!r)y%`>&e)O6o>9u;PJC(9Ql)Z+>ic5l+jZ3EZ3OPrbJJ@{>gF}j zZJz4r^;`(al`fZG*AV$R)RDj|)8Sk=V@|b;Xg-H62iLG4@75!6XS6F1+V8-T7xl;6 z^GQ%S+WAFSb(+LRy`t`SFCQnWb{4p0!}RTA#aH{#g1nymwq2~0kTAzZY(QPJ_(Gnj2A?-pQVns(5=I1-j;va5Ep@W6ojaUY|w zec0=8=j_;5J`DO0HE;8J7E_$k_76>y$!wPoI3~{DBC|Lt-CD;+2Fzh4Ef_s0jXGQ-k&I~i{ZD^@8~E~+b|!MI8J5Ak=Kp*rcR@vQ3=VxA=4=iNWV z$R~!X>CSWJOVfdJG$TlK#|B7GIg+|RVn0*+CPaR{G9CT9ubf!Cc105%hH^S+*UuBW zMH!zS=c~1y$2J9SeYxIg=-_brYS}prvG!qSJ{A&j!#=B8Kh29qFchRC#wpCi!yvJ* z&SbZ!jm8P}-yb`EK7v*$oA=hjEZ}ZtsSx z1vnobV5}L)mOgZdNPmAW4Y zUP;>5tIdg4S~#SMyjQ<;!}t&GvEk_i?DUJxIrr2X%lKHm)DlEt{DIWhFZs4m@KvGw zz*~mS5R`Ym`B|4xgN>`NU*#_0flA8<32`xgNc$qVN79zyN%w@WGZo!P#ButHluz#} zjAZ(r3sZwag4HbW;V94jZ&if-)w)l23CJx%`u2gKdD`dYsBE2BBfEC zbKFB6rulV5ryR1d_@L1(o)^nuwAk$`e}NOW*(MgO+BJY>8KcbtW%qGm$$`%bLi*5) z_AM9H$XM59Uf;2xinQjaLGdvn?|FFb){iwljJN|7xg4L6x$Flc@xP{84-$2g=KM3e z*!B~78#8;}bT9bdg&Y~_5-ojpELSqE-Y%>M&4ZlV?}aHtz3u#x-wT9&(7MM*=mx7k zR7(cyYH3H3xt~2HCKOKJvrV7yXJkGJtu`7U<{v$WV>X>w2YCk-<$rEr@X?_|${VwvRj8u}unjhx=)m6QlQ zUP}+~mlHhRX{E7WBb|iZ-*POd*u~5jTcrj69J2U@RQ80uyk6~4-_?9kXvZ;ZZvMqj zt<*&7u~)0R53R)%-cw6cd5+-tIROqm;a6mq{h>R$#cZ)DNVw}=Q7nw6)NkTouO)K} zG&|a^;X`_VfR)MS#Zc^_7O`FW0*vo3av!)X!+8Jvm%Y`kUYyWt+Wa8>Of)6@aD!6>;gS^>ayDhsR;RAzbj6-+T)lcf)!`6r( zt#cwS(485!tYEt_wjW&g?qhZ$^L`?)_UgzFVWjyVQE$vS025!r_698`ct9O<5o7sb z|E$M}9o1ty$4pUm&rm(A|}LXwcgw z^5cenI8%FUr%zPxv?2DZu2k`Cy*es(j!Uf$-2NQ~KWDGM@5PAYVdAZi1CMVK^@>Y< z#LN!F%kJ5eG!IjHN>>`x~hh8C>#r54K^>cGjE%9UVjZ*#Pr5$u+_Kvdg1ektN<`(MbxzjwMe%Fs z_V8Y(^5_SSY{|5BZDPdLNt_!w)|EZ@j8wIyPIa6DSdJdN^iMq-#Zqh7G{6$ zk+_asK5K67n`WF#I9l9%XY)CRt^=8EQSc|-_MJFrnX!r;&O?v*xiusXTUgMzsM5VD z1ahXWws&AUVPB75>e7p7z=`}_XV{NELrQUYsl#O_q!_03$LA37$L=Hh&Yz^ef$>6t zvUFMoOqXypaeNuaDcAg5FJoS47v(Q_%}=bC>@9uU99IiMkI>H7Pn&wkESm4LrT3XY z#pgv|ofZiFk>O;$bGs9`AW*++y#1mkbjNKHeC#L#4c}Z_e$F%LApX-nEcA=@$m?1M zgP%sYmEJRO01~^DjY{VVDu%ChU7Kq{?32UBf7;$r_F+3!V)?Q2!%!6%lr|z-{6D^@ zyYzkO^6Uxp;Ep=bD$cN5u+@>x-B;u!l;!%#jz^cn#7Co)P!SuPklq-#;UObW2WrFj z99;D!7AGG6Iw*FSp#zPDE+5QO1o#2Pzbywx{&a60gdV55g zH7B%RkBq+@Y))ePDZETGVjKs>0=Xp&)NuHo|Bym}5Yzf8_l;4a?IP^qE|cK>9$)@L zXSyaQ{(+Xt)u^;AAHua!*2&u-FD(!>39(6 zY}Jkr4O7Qw)<{8}GTGK!&K%1;EeGCIGS)HlE%PiNmuJ+&z>eAr*?G3sNM6g^mAI<} zX8a9xjM+V~yWTJ2u{0x&hLoVkTgH{kvHOunYTPCUu8L(BH5NR(JFE$PoWa~D6^5>Fu(E~Y zgnpGCP=}jx=2}`{;$gV~wbBKK*b-7^BX1Jl!;RW}Y+C2LJ*2E){bMxFk@`_RI9k7ffKXPhi?E81Vh$ZsdH z)u!gVElI>iu;6)`}mT&(&!ft zv1^_21#TBne74Cl&Mn>-KOyzf#WyTasrjkq#{R$ldHd(RuJ7$}q2x4a=)`Sk4)n`g zGDpi%Q8XHQKrA>!qQy&eTSGb_i&@c&aFsjJP4j@|EAW^oF+Zu1gDj1Yup?gN*{aYvruj}cpWg5|gwRWM zHL?my@acvAQ)f9jSIWZxb#MCN%HuGnyHe@L_&KD@Ri!;K*T-DSy_dER!=Wvvr|ff} z3nri7^wVKkj#K$D9F>|1Sk1N8(mYBWL+WpQ`tXX-C854KrKh8N9lWZ;ELHo6yqmE+ zU2lC`Q)u_=I1{76h+ksG%~jSHGAKB{G;q&;et(G5ihA}e$glM~YO zrj9r+zUm>~VC>5{=U2#~FQ7wad45=dJ^liTulDetGlwd1y7{h&goiZbzn}?8-&qCa zJYUm%xfNm3ee}Kifv@;`g}H8D%pzzsy16!IZyr!56^&gL`=D%czM@3q9*$V_K5u!s z4(ay#UJ~A|@L`R9NtT>1jxJZUd?(X`v)Z$}{lCiL$e)G1B9(g3SE~8ZBK|$*n)!B5 zUw#A`_e4K^IBk!#6&1m+{B8jCDAx(e`;7cae89>*SAY5v4zueS9T~S^+V}g0yJ9bs z2>xHcrlL{QnfFM&!M3HHYX#PO*BOU6Gw;7+EtXeAwz- zwWojQKkBozsz_)wAuSk(W{;lbla++%E&4aA!tOzr!!EN&cflos@FY?$4}Iu4!d#a~kr{@?^i z1t$v-_)L|^%?~tlKAgOD;)YYjI%qS!$ano@5hf2E>ZmhthWY!iz#xGS+AX)8`nG); zPG*lNR$kSBiI^;*%QM*!o%Vb8gCPQ6=?B_LkEX}i=Jl$7m0lQ`{b|x4u}1{H9i^=} zWE*M@{Uuv{x>XvXEqux~?)nlKw5*FVXeM;hhQ8-dQ41WfXNj$bzuXwiT|HUQ6VEu` zpt1j)^PH|awjQ@#JGaLZlm6_!zh#W5Z#$u&vD(8^7e}0XS5-B%A$5I8^Diz2-xbQ= zzFpb7w3o!bw!x0}<~mS5ht&;hPB5*TFLO>Fe#a+qXkj(ixY={8WqTyFS^LmmKgxJt znye~Ph07FX|Aay-?||GJ&K|7YarwlrbtcgITx4R0N+$Gey?9_=#{e20D`el7CvX6I z<~1)#FFqJv@+0=``-3n}+pL^ax)}-u>((9UXT)j9taIj%YDJ5nSBCw(Z=N7jin^z7 zl-&w#8OI;XYWLw#r~FYZYXbkN^-8=qzv&oMJN6Hyx@D6&NtxqK>nmWW(Qa1oVFvd6 zx_Vcn(g+%_bJ-55<>KsH$I=h!lQ_P3tKbK_P^3&{-pjlG7G_uS3pV!?c~K)V;`jMm z0-^bUlGUDeA55F1)LoZ10ZI_bZl|dt3~&yvDUfDCnuiFkcnv!O8hIZMO5P>amIimaiqDUQU`?(BATR0$^(MJ9x(?J(72 zvuNx0I2aDUx^859C{~QvwntZP!soU8n|!1Ro?&N=Xm>>+q4&Jt^!AF!0mw`1G+ECb-k+&qc+NdNa|cBZ5&+9S2X6Qe^Dzkhqr4 zZ2ElmJBe57+o~2HM;u;1p-(oaa~SF}`oAc5O&%=QMV?U-IKS>WwYOf4*dIeafQljt1_Q)>OuQd^wq>DHE8G4Th7;b z748gFSwP`a61$kpVWm?<{JcIPq@qQo7TPa&yG!RW;#i2ywg-2`He*w8*hPb>2pn2f z^6T&`1OTgMSPyS}O+* zE&4-Zd8=?i6=*Y+}D7TbJTM_*q3lrnv^Kkju_ze|PQmf3`m@^oo&0ofLGHZKEAcXXvq# zIFEN4JC-)Vu%{9T6%zQ;MyzAF?>Ldf1cT6|^OD+Ns5- z;=q1yg8-WxAj`^Dm((-#CvcE`a&mK-54P1r4*0qqhf(n}x37>Hx*9;sTjO%`Pd`xR zbar30yn=(Th8zYg88{1G+be0*)I#{PsmnIzpZsErZhs1_IqONpc+b_GD;KuF{>o0@ z;HyVr+;6M;@kfU+FY4gq(oN5xZLyp8gS-TwDA~50@BNNs>XZ8EqHoaQCUQ!n+6VfV zNn8l6?7)iOzV|IIQpoH#E9uAWU9stAh~CFP%dq42{+3@j2&It^B%iYPVL`GR?A_2v%LW~Sag7sK*I(De;iJorm%O_Voh_3=*&k1mS&O3rP8labSHIxq zw`+W{?uVmkNr4phoC}dGD*TBHD);70m39&Hy8WC6m);o|R?m8^xNerj`>yFylh!_{ zu(OL!5+U*>8zm{+ue6O|D5CXU?+Zpg2o9{7nAzsO7buTg<92kY110stuU`k>LdSUf z#zV`4$()OR92KoVqh7TjwJ4GHChc*Zx*DvyO?p8T3*xp zA1@Quel zyIHd*v6u{*4&56i&3EHa{@FKw4MFI-GtzeN>%{-!U$;(F1Y^&iS+c(!BmNE@Kh}jx z{dtCiPh)&Y(L`QZ=Z~%4nF&5*4o}0UAsa-XDU%j>c2F5J8!rf^$UcWgmN)5npU+`W zci?iz)n_16wO%!?nyBYDUaoz*CT+!nFT}QA~G1RV7 zF%AoM-xyV=G3CKw!^#$2;o)d#UZSFFXHXlBPd4M&*r z2^eN^cu8jEKGbFDGYqrxZ6a&58T%oU>Fzvi-EmMCT_2P_nFtx?%bI3i+QNjGkf;0k zUd;YkaJc;NIvk+ay!G{a1D%Js%vYY>fHQ*0`P-i@VD%{Ls?vx3*hd@Rmib2&ngj!P zOwPvvC1?G*+o$r7?x)2qF-hou6lFcMSXoHqeGDnxHoSOI11Nbfjvh!S;*K2LTFuu& zW3l1T;)r!VVn97s_u=VLD{NI>-#ymLfdxAkdfe@#pvwQmw{TNBHmu3X^9|I++0Bg$ zvafkDQHSSf-Tfulu3C8B;oBH2NYDHDe!Bo|ef*u=6DCkuR$LJ9k;wmhC2{@Bou!wU z`Z;^Q4vl;Cm|Hm-C0U zapFLqTyX&_(1LvBM!!&CKIE|Qhn_1WmVkFGw~ii!-|sroen^`^5!=~>Lu*%(*bMu< z2Uq$)``f_4LZB=$j-Y$5BK$jl#q*yA%c075e?g$$F`RwK@2a3F zj8sy}tsL&JkZyT7LMgid>L*HB41N-IO)0|dqAa^dAj6&DP2TZ4~X$dk_HqkXT`PDI40G4}CZI&H)p1rY(2gKP9nnWdFLP zMdWFWC7n1DS7rkA>nBZQvW`N5-|hEfOC6w8^FzCrfGGH0w45D16wA~OYOvn5ex8Fk znpHx7PbKWNX_bIQnbMU&Ut?TM)pm!T53GO4vW7_6arxnUf$#scOGc{eW!hiILiJG9 z?&UQ^U9&l}-`^#!$inQ>2+3u#mDtavW;Zj+$a{yupI@f)2Bnd_np?rb{{*xxIUqa| z_XuJydCCoytj1Qgm)|8bRH6NQLRzfyQ)m*dSbuIEQ~geuu;R2T^k|0S9mgg6QwW@Q zsJo3Lgv%7OkG91<(OQl5Z~fUSoj8#0J(Ck_6NDYs$6ekP^C0EPw~H~tBr-5thpExu#a0DD?$(xfaEWp;G6^=8#L_DbT%&lS53#=p#u6>*PwvNL@ zuC4!~*h5&VIq+NFiHHx+#c+7lES1BaMeBsMEs1)f9TmD&JF;VNB!b=ZjWUz|6SiFc z5%Waa7t(CFrrlQ(IuPaUnF6Pd=t6PY$lD*mRm`hK;^aN%NmD4paloaO;6A~e$2!das_>ZaM@XSN=1hf!V!Xf$(SwT@!-;b}Yd3kai>l7lp7a+@!|Ye~!GuO|iSt2+t>+W>b+yYx` zPJ7&O5+rl{a3V)8-iDPSFpdmzs73z(=xqeHsP{ZuObetWizT?{9_G{nMUu+ z|B(cg!Xll)S4*JdbThkkuREFJW*GgGB`daG-v8~?*;6=3HamYlG~s{66D&3dJ%=bE z9FBQf4Fjv&=Mt3 z$8|?H;C$WLhgWXz!luuwZrnc2hCOkc?=LkK2g=MR`>d2RSe+Gi`Q{iyR}sg;Tg^r$ zt8i|YYT~liB8>lH>z^>b2-Ck?{vP4^Yfs0F_i{P^wd;G7UU_xj#U`K9Lq~?yak6v^ zkBL$(j=HznxzN93+Zxp)(anVJbY7U@@LP`%=vOX?3RTU=et}0p{~yx+GODUCY6C~< z770bbLcl-}0R;nO9fJ^*5)m+H6a@jhQ3NCu0RxZ}X;h?}dFYVt?#@%dp|AJDJ^LLm z_cz}AANP~P!;ZDrUVE+?&wQTYz-W}p^RBqv8pFGH1b;PO^?i@3I*sA-y9v~K!_s_NSXX6;g^xnq?>lH$z?||Ed-eO*QSaKjKZ@LP8|*j1J9qf5`vPiyeMREO z#9bJ;Y?!iWYk)@2r}*A{>5kT$4ltb8Q-##fUpxJH(oqR=OA)y}_s^gB;S|hT4Ktc& zluD;$U??F*^yBSZn6stehWErS*HmaOuskYjpV=j?0cPr&p&>hxNg+ z-KmG)10h$7yRCH-O0^JkkCNVponO{@U<*?fj{cqlxjOMI@|1Fzk#WKyBCgS2{E91@ zZGK3nDT?8ur+tT8cwA2b-kke@OBkjbK6mI=yvdKlK(^n(xt|gZtE1H=TsFRlO#H)_ zD)$pcZP-$kG_btkN?qx0vF_tAx9QEV$J^`B)Q?qG&PqR^GzL-?AC|$Gq=kn6R#up6 zVSFmro(1g%bAc=;MbMaZ(<9P}<1nle6n$Xq8q_~YjvCC=fP&DY){B{f&}s0-UU#q% zQ4M$$+E=lD;nAoUPw#H|hWcO5i>aTU#W5Uv>vn2?8T_4;5Z&#W0rLZwc6PIWg6v;? zmR9TEpeMk1%a%8u*#2-N{ISbc9yG+P$8#V?51Oxc$(OwEfmQiN(dwvMuyCo}R(;?O z%zd8KJUI09pWn#4aHY@%bv?AaBDw82!qc4*L0rFJez$-Y*LNQ@M41*AB1oXoqkc6G zpEu+s@$_9n+R-x&(7opEOJ=3Lki9W>kN>P8RyCB1y39BfWrF-*HQ<~|OsoOaU3{1+Q`7;uw;BF4pW%j%V?s@L z7$Wct$3MosVKRWWRG${jC$xMycqGiqydQHw&9Ar&rBn|<%G0UMC(hqT3rA1>!u?x&3~0m@{c0s2h%UJ9^D&-WB#;5xcg@Sj&5<*TF2nue#jiR z>RsWF)?;`JJUymhC^_q#?^-LGOCj%@4P$|V!PeSJQ4!P?q*(lh?0|}n*XjT0ruBoM zo?H3>A^+m78OMc?J!`6nvNgx#-Wv^Aay??R8e9N%%+GTC4KaPLmi4qJNpJpO>-wx~ z)9sfO=o5WrIzET>2M*aTxvhxG;aR4r_X191_PKuZ)D&~gFSNoVZpP628AmV8GR7f> z={>J1s;N&_V|bJC*t`K<$v-e-*R{#NJr(N8{hH62((EywX}`F!^2uNXChyr_Y$`?T zz6UuOgP;7fC+cC{XWdbpu2Z^Ogg!LwJ+M{zG8Qj3)XK2tr!vAwn+3<4Gg<%kiDZqj zd-?!H#=T)@xHhb_5%W7_o!eP3;FQ3*CmPdDBQgJ#u$NbYv3)m74hu_S-=C;|wD@He znCKVE{;GZ%=G<1qIn`!R=f!5-9T}!DJnP;SxXT!dzcJv9_c8#PJ+i{Z_Rfa;&fbr} zZMi_h*P%Yw+I zgEdcKemefd`_Fw(6(Q?tzxmT`z0l$|*P3EJ1#6E@eGh(S#<3shUu3N%!KgzXT}UzJ z-x1W;lk4OJP>X6*c*#$!{?*KF#yN3(V_t0N`&XfF{TS|MGASom{0&wgqU;dTwn?mq zwzv0Pw%@cwsbNlJd+Q85J!M2@Wa$$y=YZx_^2|I_}O z1OtxIkdr0MG7uV`-B0_O&I1c^7v)#0Og4<8VY#-yM*s)oPl;YT@ECr&KBPC2*$=b- zP3EaulDc{@{ejsjp7H}bzfikk2agMmef7f#H^(6RsffKC(Er$$XfPJj20V(WD%SY@S*3`wHT6`rzT`X&=K{N4-em>xr3 zumWSH4IfZt;u}8;(fUzQ@G&pW%Gylm`E#i4h#7`snp+CHd^lSHDpFO8%oL}gpT(%s zy3!9%mmSByIP)16zHf=9mpFmOs~_^~3w5Fa?TMiw+y*@nRQ@bU;somtSTYKIADEwt z27G*yGK^{bCa8aW`+LH;L^`@L39P8;U2{qDUT>Ne+qBDpN=xom`% zHJ!bm@omr(r(J)^^~KwnZ#!W?*QtE?Rx*zMIpsOYv=fbSC2Wy&!}?&`cMeTA| z$1$DHNu|S*mm;3g#zPKd!+9pt`fCuS;PaW&ESP>^oP4{l9?ldRr$0DaayUTW6w{Af zXBO0Kt+nlV5~k;zrSe#o(J=(p=`DwLMsU)u2RqtRxm`=3^H5c&=rJ5LA1LCv1RLvA zT^)5sYs;mu`l!RKua!S4r=s*H$*1KYoI3 zHQ-bmno(Iz3=}^D;mx;I5|Xg}f*|?)#N`vT=Y$1O*01{vF)`Mz%8#fQ?{ex6^xZQ!9oDJgsuS6x19P&|()xTGc(hR8 zG{@`*N*kKxi}*;dzYfuw2RR!XUZA1M-sY?08sOtwakA&V zz8)yJtdvMsA&18DU;Q02H-YK&s?3xlu`sz}_({p52}gIV*{J(TIaKnZHWiU<)Wz_j zf`n&-{u%~;e{n;I%RUe_Ghv8GV>NSJ1J_{H$w6KA`oH+F=p|2C?;bSzBUQuVS|kkY zWhcp9yp42s{7hc9qeJ5^B7QRdBpNYbp|_cA{;&6oqZ^7C4^%8elue{XUu}#Bx#NU9(|kUrJ2{-EcJOQf zmN&@US;g+SEgtnf-))y=v4K|-hHF>qmOqHW$U8~RM+<3a_=tvNhF}x)M->vgwNz1B z)%EC*OaVMyF{|7C1sgP_qcDLIzQP#W#|yD_Sl{lT@|l98k{A!`4gtql#r zA9^pOwjhd()sg$Z?gM#=)RM2Z7akt@+0jaeL*xhVv7;p(TYto4I>|$NE~su_ z{LCfJ798syc~Q%s-6*~;b<_S`8}~n|%$|K$Rv+S7ew=6ax?2MtAS1l$!E+A@?@*cAz7-W71i3S; z@B8$cAT?V)PldAu4K*B7mdxgaMX*l_4BCNbYfYUwv^t6Ac(k{$#@;}LvfrfRKAvd6 z=0hVJpm!{`;xj zi5egbmbAvo*TKx#<)k)-Zq%JF-rP4M{LkMBEa4Ky^v1>$i})T&(?feC*TbED4v3uO zBep2=3;LN^a-~Mg(1fbu>qm^YfoPCFqVie`HGe)EJw-i-sJqE_{X0xy;e4sv3Cqu@ z^9g(X&>c-Qn9SzJB4h{i?~3Z0M~cy;+5Dq?vpq1Ew{Q2&U!1UH@8IA?ZHCr?ZCZxJ z*U*0|Bm5-$3L+WTtshzZi>9skqmG1c0y4vFV#-s@E)V{-WKgvJghn?vRG+(G1ElY} zU%GYpz_@05-m+saO6c}id3W0uHMcb#QFZHpIh-Ts-~}5*w5rN(rAVV8$P|A0A{9r+ z8ap1Z?Scvy8XSdhS#Nm%%L;j37N*@Wbnl#3A=?X7;Oh`jOuPt%!<9@axes7aZ~eft zD@0hFcDVl5S_VzvfBZQ8&+UAKzl&$c zs=e#0_W)&ur8Y@RV0aT^@9lKeP0Nt3(R!im4`ydl#@Bl}atUZ?Z&~nFqq~?rlFMzP z=T*Vc4SJax@TWt?ci*N{&!f-)zvo$}44S=zzWenLCE`D$h1i$gcEeJrwWIo$pW+S} z_}LTkWBmvk*pn9mXcr>=vD_NxHG~FoJgOfQV7#D9zbPjx-AobTi!LE*o*O8U8F+ znn2*LjSWAw0$m?-&I_NS;iaH1=uE()b6DST@z~gdL%MN@RCd4rjq)LwM*KGox&`2G zsW8XaeH}nLa+YFJ5Cv0mva464SYY^ucar#h+WbP=V9*4C&H?(jwcR?oJ(u=7>1^gy zLQNw|y7V%s%LDTpN4p8?e{NuWmA$W0n-3<4ZunkW#cKB46?%nMKV1?#P(202(+7iG zjq6}klcDe4SsKqc{C)6VYl^D{dh~dmq&Qwf>wKUs=$AupUgL9{vvkmQx#z>htIDV> z|7xh(yRv`wyz@<&gB8s6I?6lsVEY?|@p5?cU93LQ#oKb5XBx1v}F9aj|# zmu}hum0y+9!5@qFXQGcq1toL<+;a3Tb{(Mo-WaKxU7#8X;ifTJCQKL)Ty>2@)zavF zm<_h1cYH*PH_^12sF^XL1dEez3b(|K3qe_rDM+k`plXLjfd<_Wv}BiJ@Nl9FMs0NL zhGg!cNvm_WO~=TH80RG=R%{QW!T5rk$wy(eZ#d2A$cFweRH=6K4&ge@FGFp!yb8r{ zF`l+XkNv-!oGPHjE3Az2GbenP(D5=K-@|Qbyi&VVJtL&5xv}7=;{xck$uc4ONZeiC%4;FF8ZgV zp{*>1vUfQVX?yOdtRIF?sybt z#RgM~xR+5y_s|?6^j*B8DQXZDVOoB35=f6Lxq}S3(88%yhnR#QfXz-9&C{@a(8w9W z4N<1+Xt?=%t{Wi|maDE3@C{vP=!D#v=0yzWKk%|W`Ny8?|DXTk>D@a0MdG@k))VH zfT!F0__Y6DH~1q{WN>?mwr-&)(V?48oW?T+BO;|=w{rbK^FMnm4_-S2<9C%F^BmBD ziG#Iisn0~wjFLwF@%^`9plI~?lXXn*q(a}HqFavH3CeJLkzz?O40d$MUAuzW<Jf^2|A0g|YsZr4#j@Y0@=_RGOx8 z1mov|h24kgVv|myuKwiL50dBq*$bOKn^1G;+Aq)&y=e+oqA$AO!+T)a#acaN^C+w{ zbl;N-DuV6~XE&DQyJ%dU@+5=L0?ofrEx9Fs4X7!1UM43^qK0dDx-)V|@hnap35S;G zQJ=Je_M^LB5y5EhemR+7pvcz62>j)QDj>F)FvY`fyt2E?JuH8grQEK(___tkoI~yn zJve~yW+{e3R_bV`*U;>_&OE9ytp9fQOCchpgeup4#dsll>og2yEU`YJT6?xr-&99X zU+%TK2?-a}J;?Pm8dnY@T!ZUoKQLVJP-tcQo=`66eE3IU_fE{7&L

    p(pd;SQZZQ zXNR1G@oUc}#$zwQU|{;*?K5@%d)}dizQ`}MJ_7a21eE_|$btH2b>W#=7@nm+^Xb*6 z1u6(nm-0o%Ruc8^5lMPjh2fdT9%?KTRMEhjnKEW?ceHro4_6q^89b9D{VzD?3JWV+x68-i z0KZL9&|cx+_gnP$^Rk6Ap859;Lhi>5)E$}VDDZF$NEW@}yTln$bMpHA;K(}ES6FVR zb}A2z%kA3ClM{?*?)9XL&60#dqVtpG@I+4V^XAqW3 zM{fzt4MN|^dx3(tiqULtIhVh4Dr&#;POi?^0TB%!?()hGhql;bGZT3|F!**y>!Yge1}&X7XNJPXfCDkzoRm-9p&6O6_h22!w)j?p&`2B=%c)XXvCNG)X#TVo^y%= zjmoV*0kYxsZO)PvD5B0+^iuSOn23kG}%5uNZ zedGD^&n{9MvPN%1=ZCh+J*r&Dq+(~}9qZQ{*2!e~lSc=i*`rDJoKuBO&M?PfZOGXn z0wY@Y1gZoBp__5x=0GvlM=I#pX%l-+Ydn`uywKJng!jYC(zt z6jj>!4aaUSEI0Ad{vIt7^86(E369At`|-(h#%Q=CYv&aeYc!OT;yrah6fK)&SPb26 zfbZE^lj3iF!pvY+=jxR*)OtZh)a&QJA#P3 zYQOiP>41hSk`p7aFf6|2rd@Ad=rMa(4q!Toq9 zN{mCLcQCAtvkH6BccSENr;SfU)u9$Xy4$%W;W%cF`(l+VF+g~KR9WAQ8M;pk^?Y{0 za1_0FG^?bAC}{BegR44)MNS=C!v`MgnUe{W2O_XX5P_1r!(kKx;?F1gx2zOAB)ZejVFwx>8&wHvjU zJ#1jn>x>jTLnG?n*YEexupP~q<>=Q%ccB%9n(!2zbSPpd3hr(sd3T_@3upELV7{Tto_iK9nTZQE(5h!o(4h{JF2;u79TFf z0F6`8^=FtHpzSMP?aur)Ap8idH1?oF;Q8a-B^|7ffXpdxd(xN*re8!#Yc?1Fk>{&h z!#u730Ohrub6wt9j{Zn#1SlX+RL{wKH<@}LM@h?xuEBG#<{D%Uc7}IH0#}_81~{glu~|8EpMhr)fulz``z9_ z&sE21*@PKb9+wC_b;At4m#3YxEZnH4Kmtv+IDZ59Y4<%W=Vm6P8R)H_8>z1w{~QDeSDl$a?N7fd)i z5{a|MFgIpXRo54ch)9obub(Lz)wmF1`LZ2HXTX-_|N0*4v(AyrkX2OKh1DkrKPDU= zMt?`$tg&dIy#>vk+-9-aRSP9vlN_x-gz;Mrr1#q-MCd?R_0{s5UUhi-zI z@P2c$?8sx(IL>wFk2lt5IFs0FViZ{l1Qh9RySo?G6jo9a$$8j!Z#f~LaX${G1at_D zk5f<|%U?|^4tdmmQW$R>+XF+mZmZBLOqXr$)R-nmrZP%7AZ0L>XN*XjB1x6WSbd>t z=oF>uWFC+|Ss0$(J&I}{Yd`Hd+5W%R8IJLM)!wI*l>hzvhJ7mgS&wC7QU~fS=jffv z#&{e%l7HE|-symO?(B1qM`^fZn7*Q=t*EgPSDd?3G4(bZ*0;1q&27jtkAx)_{DED< z!O+e1J&f@!h68WW?sM}Gi~kponbQq4NWo;>ii{t%6OrR=!YB20P=jT__uH*vIJz@* ztC=PF$irfEhqsjv4F8-wkh5DB2pOqQwslxT!=tPy-!rGtFk?&0>67a)Csx0(&zl~) zBku^_c$5pooGB@nNIRI=vwlW)f&~#Q^PSRdlws~kqwdX0D_A=1B{|T49#3c46Sk*K z7}olP$IF{81997rp`2U!c*aRF{^OZFcm|VOYhwl1vG;zrtoDp?3+guXW|Mp>3lsMg z&h^-HqJmEYyo|Rcq30(li&9&Or!RQ)Z2ImJjITU<@@nxNP`Uf;7zmg@z0x4i)s@SK zn&h@;K2Oa>HBU_R4~KZc!Ye~M`K5kXkdZ&Fzx(Ni_crlBMlR-W1@;~I&)c%tWTNf` zroDcrv3-B$>akjJi#9x6Vfy1+54f@Q!IQ`F7ySNnU96Qq&?)$O4(6WOPE9|h#W|2d zJty(mDHMkOyf~Li&y4E6el>G+$9O(x2(9VEdxoLGpiugxurVS`iB_o%rJ|XnwKEr{ zRncf=x9fxeAB@QJHK=zFqD6jLVZnRFFk!5_CS>6agbx`>^xlR*y!gGPg8zT$<&an# z*wpf&Ay@ZF?luZ)xx8)Lr2il2?!0W)d2!OOmQIgy$p!t`iBk|p@GOQ zn|jYKSmt%TX#a@D%K}tplE=w2I7qX)H9CEf0ekP!H)oZ?bYSS%t^PlM>Y(}i@f@|h zE>yN>NA)X5d02h%{g83WbJXhm!$86*4vHTcZZA0P4#dwXKEAu+(9{w6flnV@QO6{& zn89ra=%>S}mGybSP<4nm>5eFl`LTqn&#n;2BlUBIiTsO)G)xU&nqmC79jWbQ0`m`G zR`C4hD`J&Urn>iZW>+$hY8)?C_voVmvDhL* z&myAOtUOce44O^7k>lho0~w8KfkmxeIHvC>RP@H^fIRv1Vf3U6swi9gy}#x=%GV+$ zW_ox-K~IypuD=snOp-Pi{TH{*q&fe}bBTtkOlz$RmGNk@)n{cjG8Zk58WUNc3S{)4aoIcxbBzzY}`}e7;*91W4&P}rH zp(I3QuK7w<#q_SbK8C&X(8G8~=Fa78W^PS^@k{#VH9pMHN=MZ^{5}&U2W5R3TRV#I z{AH(#PH~{+_=c80gDlYhmSuHXCKBrPDJHD1=b>F%rJ{*75fS!9jwDWmqsf^gYZYlD zu;e(fp1+uar}K|HAw ziTd;YD78}il8cof42oX~e{YNN1l4#eh;F+fi|Ti%4t=r1{F>QUf!lrW8lar3#OeGU z$v|8?G!Y%&fun0#9thH_fz@fg`j!qonBGxRSsRAw<<0Ta4-A|YLH!bhto^REd^esh z{f(Fj^8^e^UNF_19YPKMLHUwJme6-C%5OjhtB)>rJ~_oqruq4ZeDo1{cOwl~4{Z;G zx-DORfWgzcegTT(uypRMcyh`sAQUHwb~({_IH3Q~sUy$avHg`;SNhEKD)!tXe!EYY zo7kY)oKG3I+Gsc%?^glz5yHAQP;px<@4p4$^ zU~tCOTY*{+^MVyFhxS_mrQTAyiue$V6Hh<+_Ksbj9az@)2KHv5n#tWyj(Z()ftknyd_XZQ?Sw6AiE3B%T7!!a4A zbSX^linOg$+?B5u#yNgzI2b;GTD_vK$1w;E+}*NPh?9Y}Q?+gknKb-6nu@wu$tJU* z{ztq1&~Ym;U+B9brKeooI5Oy<5cZ9+>dD z%>L{`2rMoCrbJ2YMcu1uhl@YDBLZD@7Nb)VOb{5TE?dt+ieswx>t;OMuY1U_ZwZe| zzPyZRJ#`wUClS5Z$Q<+taJ_sd(FC*S=zQzAys`E3vOSTjmge81zQS{Rm`5(6!Z$AG zg?KRgKB)P4+kSF0j5Sjwq$>&$-figjQZF7+og^fF@5JhLvk?hr&-5OJg~2a^u9qG| zAI?y3x8{2oG`U~VqSMwQt6J5yD3>ECMChJh6H16lhg{CG6IW+*Xn(KkV?3yhQqjVz7c9Z@9*A6%F~1Xv z&^V)ouBimp|J7f<^o0+P>5?p-z(*&9WA(!32O>qCn(1i4VSANOlMFP;Z2v({qv0=b zEJleUMggusWOJOe%=JVgkqQhAOeay#rQxfO`my@r8oyfoXZatf$m4XwVLnGhKIX4b zv~UB)^o~ZTIdA0GSaAzwy{^4McNiK%0!0@d-2%ez(ce}KtPthvX0ASt<*VA+^~^j1 zY5nMsp(n3+>y$mJ(MmaD?%9l}TWSNPnk7-g6a6rSE)0*^r1BvD$~wju*S51_b7oc> zbhBRHz37UAsnDiZEBk0TX;giBUTmF!;n+sM-wXO~{sOf!F2|Bh`q1R0xtb)&2fEX% zMvXYI{M%47FY9KuPE@0&-LWi=je8_Bj$Q2=re{%BpFOL5lvcmM)8m9+(!xKp=fJz>GfP&HtZ)~AJ|vndo` zcv}Uns;-w_55{kJ?}PG*=f#;mpvm7^Tg=nv(a`cOkpUsu4abpBF}m-!j6u}6pz*WX zBB&9U?fm#V16tdpG1Sue0=3_lCKHLY{Q^%H6%&Lwf1vuf6%%1oj1Mc3XOptm76(*a zW6~X4dI!xW|F!;m8pGw*Rh7tQd3;9GT5we@inMXOx>;qJC+l%cVOKv6*=_~0a_!Gl zg)nFhKAxbGkPX8Ii)}C5H$f?-?aAv)Vo(>^Kk$U}KAxNI{wwFqBxojv?>?znh`ujg zJ9)yJ2mQ68w_5tU7sfWvzuluitM5R&q9VV}IYF5AYkZNf*@4PMUj-LM$^y0N%%NEOc==FFvs>8KBWDRkE!pOYgW1CE1AV^Gr zF2XyvEj0ZIE3?x@CH(WfzGnZZp@9ssTJOyX)lai>Jr_R;T zB^1x}v?PmP^#+=74*8YVb`Vv#2r{}{PlAl8mF%~M9;ltKaa%Hl0kvzFdJ$B-W%7f!nlLR|(>&8&2P@gd1$TA$ zq4%j>)@IiZm=X=>k!t3flrNzIf=*QO7 zd9eyKhK`-ZoB05xI-gSOVhvMbC8On!+Y!+|ODaM&7R^83>K>X(!>d47rCD=?q%8Dh zdv~dt4g%RnY~iND703`hKI0n11S1)h^TsFZpu;_Jk69uIG!|X;diGWjX5HquwD>SW zxoIqyLBAR8dSmahcv{Eaoq9 zR!pv2-GV~Gd%ENIG|}9&>;05V$x!o(D8gY=iD$Gj|6D48>6Yg}$597f|W zRrzaoXu_(`wcwSLi>O57f!vxChF>SnlYh?@(CS%u7QCEMz?2*s6KvjJvQGkK-&7xN z7sl!am3Q3zz9fp{nZuT~CSGu&t{m}w(?gN#8W_Kp|k$FDv;9Uoz%hdS%}Yl?fOQ1g^(vRw?uJKf*; zQ7!&~2^uix=ZyVz17@CH(C9U%#ld(w>vzK5D-RL%a8S39yAm{A)a1Em-iMY>c4@`; z(fI6fEXrbA*nW4Sp74EVwnj#x@!J;MeR6d`wis_Ze<=aY{bcoHDL4ls&oeH-xjD4d zKVP~<{wG@h40~#Q+tK<@_F$oH8~kC~I0tOGI)q!!;}|q;WOs6{;2G)iWG0?9qN%1~ z#jDr!@pSBZmNTj5(7#pW{Ji%)G{|_?MrJn+w+_@)aYorI+n`>>$LPrc3JmZ$oDX$< zh1O$PD~j4_`y!rc%ufNl+tJ@Yk`a4d-@v+}4_n~55JY%u{_u(HK{VTMl7w%shlV3W z?M5OS5G3j{KW^EA1}$xUcYIVtgtJxhvEMaN+w)=>Hd_jKqzkL=oZx|B+sP}dJJSF8 zSIBI{ycQ@&6FqiWS7P^7Taqu8X+U+?jl^P73ekkw7R^_d=TXzX>3}zbSpBF`e=FZu ziyN#zD{P$DiuLJ^j+csiNn>@0>FLjBq+QIR$Ly%uN3DBk_V`8HObKn6%rdlie|8CV z{4RO4wL2e1F6O-^Z{3SZuH4)f-a&>D36o!8fxCck&VPSIAf{6_txGVb2on+cDR+E) zpchJOq)XG7yoaM_kh*2sg83EGpgPgLcM`=G&-LDK!{&d^WQStj`g!P4n=ju!gw<^a zk8h?LS3ZH9#4pJ#(YY{dInH7vrve$nUAcy;`G_h;EghxHM`Lf(9zNPu2UFL{nc}XT zKrm34@aTzv$>`h6UwfZH|FH57hHq!lM52m)ZCD7jyR5H0qgR6Y_>&n{V%Yvj{{BwM z%MHWvcRoK^JaV1Zw}ci`OV_GiKZH)H5H}Z9tWT-`aaT?}pD^kgeni&I8i0Dr`b#N` zi>UeT_f4e7sn9o@t-8bhAj~a4wlh9(9mcqF-nL$MhF*bzSBF<8pu;Y_ZZ?&Ak5*z_l256-^d-;vAiber`}xp z{PthmWOejr^Bs8@&lwLLIGT*e2AZK9Yv*Bhdsp#@2o>tmBPZ89^ETYS!^Z$uxK@EC zcgG&hu6=-}&2PvCUcz|eW-l=<-732ZbFa#}#Y-_h#?^AkeLpnuP`2aIo)#T(m~y*r zu>2zojqc6-lQa1V$F!1D{@?;5s^9Tp>?gGx@D05mI`6-L>4Ma)Z6z3g48b;**{L}N zRzEogzt5ueFT-#bwf?572pZbQ_Db-uEs&y_^*$SZ0BTU*mUGQcFvb4WP(UXe4Zml5 zdynrJOj{ad4%3}SJubFWv9>ml{pQ%NWhM>i6X_oi?k<{b??yYHi%7LM|!{r3>{fcgdA3O7hjuNSWPq5%{kmGlM2 zOf=BiZlLxC!;21)N+x%+v7p&MEfI9TnQ-)#yeIf7ZsM8WxPSdBrU46KpFgCjzXa-% z%Pm4DGn)EY_~gAAjmHy5Uxxz14_(JI$co#oO2wkOKS8>U5;7=b&_9G<6Z;OkwlUqO z%U?#L20uAovvxqMub_X*@)qb_X>okZG=nD1X1vvw0#Jh+VcjZ@gb31#g?xfbh^T=y z>vXHp>JN=aLPfMbCLCjg+O5gJ1}L;)G3IYufa!#khn`oCLy_v;!`%}{(DLQ87Yyxm za15=#l+F^z|Jmc~zrO}PhnbyG+x!9o5H)7~<%*me8e{IsC$zI8O3OzB;O2pTud;!) z93~i7Tj{e0GFq~HD!1t!9?u|T_cPbG6Pklt1fy25zU}(0A`@muufyW8rOBxtEUsyZ zlu1|@pxdy{DUXglF^nrki^-kx5%GgCH1}9SS5^tFe)4Q{uJi*k>q}9stLD%pDEvIO z(+`#p>{{8G_!H*rDFM#^;-i|SC*t~?m(U1LfpTcecQpNKG9g+W%Nuk&{q@dTdz<} zPrH5lVQP7ZZP3~XwVWH;b==Aak={q0WXlzRVVmPm_5}(6VbM}rlK(Iil48wmUtELU z&X1e5#H4|GZ1j2R{328?^X_nN>_FpXMptKVY~V?R-c>7++tP}7#!%f{D_c553OQ+_ z#=Qwf^^HuPhSBsa(0ColkyW@5$@^Z7PfjuF8~!y|FF%Odn>u_;#5d|O7loTgkTNkI z#R*1Dn=xGj=zkxO^jiVz_n3a3vH17X49ss`G(NiPH0tkqCO!Tq7?J8P{_M>e0?PRT zCyz55?c2x&$23dXKQQy{dvl??x{3%%enta6fUK@`d^mNE;A3_7D zWn6t$DFTm33sdj>PSSLTp^kUwBj&f?V9Y8p)QHn!!}w>7nyp0|bYT3-@wYO&_X6c~ z=9&AKPNA~D!s?Iko3K2FVq=%jqYR+dZAnsneG09yPvYWk(RA-nr$Omh|A{^nlY|nk z+WreYPPsD>DT+Wy+_seA$bkB6iO^_l3glwC=LH>oFipJtl75#1W~W!9`*<8_aXicj zZW0(C>cBHA%~&ijVtRJ#y=$iS&-Os2*F-vuIc|90lYgfPz8+XTviM*lt6>=pw}VE@ zV=r)W+(Ja$xuD-wm`*&Ehr+I9u7nme4@LBQVf)j-0!#JV87fRWti0!Pq49L1ED}Xk z-UZ`dBqkjDKCxpMIy}E$;T8NJzY8_IRCoT3dM*gQcvAZeO-&rQ`*t-JMO)@n|4`CL zec{hLjE`V=+xf`jr2_PC@eG4%4zht}@b_-RPH_$kH02~F$n>KIwZ0w7veT_Z1ApZv z8Ha44(?(CV`S1VG;f{XQZ=*4SXV||wQ8`hFax*4(H1Kwz!Oean2g}X@<@LVNcON%H z&9=Aisx$b3j5GONmX|sE6mAO*oeYPRkP=iZy74Aw7YdSW)+dYw7l?`qlekB z1R;j+wx~j`U1s3w<$rOf@2-HW7{39j#Zu3vMi0$A9w};K*$tEg)0O`4E*#yBJ}4A% z#4{Alh1C9PfFiNOhKXggI2Fwh-+!w-76h>&)(KYZ51^6hlH0W+B}5MAi^*r+0`)Ab zxuTmcz=FH)nQO!D(5-td_xu)PG#-6@#$w(Y4Hfw@czt0&LxolWY!Ply*`}>|-CPTZ z$&kDR2KNH-y-y1&%|`^sG>!ix^=(A$@w&rIVa-_|MA5n@>wT!HBY6IizGP8<=0 zg0Pj<>(#+%xo)28Cs7jyJZu9ZZ`?#9-p-Sq{hctuJw|4+!Ris@rSyKm6Xs~K`?^NM zLqRmXn0Zyre;1B1QBNc5E5>`UFhixkxknWShR&IUZMukK;O|WO{89mp?uOSFH_`He zIL58h)7P)wg#N?CtHNKp5!@^p31_D1N&ta$Ii~$%0}wWqNo~!e;RVppjz<^FBrx5S zwdvwZJWe7o<#1E(zEKNw9djC58{SK5pbXF1?b`Y=R{M5uV zr37#UIAQpug(!=Ik-HqwBm=`%QZN@9H93CViUs4Zn!?4-XZ@Cg%InOYN&YlGVLV+3 z|M-uHXDD#&ZZQXw4~(v)@_uiYf@YR#;rCicp?`hlf>;GVxXg?p*SSyJ*>~qDfwxSLI{xjm6_WEnf zSRHmqFa5QUYZ{ORkB%zTe}wUM?hv>W0PXY4O}b&(P{sf8@r?_65Lx!kzU>jEuzKd4 zaGa_DEIi|6-M8f)P~zMUJLY5j1*5|2hn0lBqMFa%zrS5u#xr?#UGu&H&=oBG%IS0; z5Ffvuo?*uD`vbU`C$f&d(ADOv;V_tq8d&+Yhxa$3sFPvtVT_pGZo5O&soNJ#QQu~Y zf7qxaytrbkPI`>>1CoSQR3qwhVL-pq(p~cu>Un=Y*Sg&S@RgxID+TIddMl@&%J$#? z?2=H_cbTg|k>)yJ&3YG2o-s;RKsR7q%cntajTxrm2K!1sbEE3Fck||5FT*IOz4JR( zO_&RFo&9ueFRDfLd3hK3{%@^|ds{>otvg83HAZhhj6Vh6a@OB!`@Nj#?GR z(z3m!f%rR(b5y|vM(lPLe69XB&I8(FbQ4%zjqoz<>Nb;L=+}HH!I@f%mU^2P>K;FU zj*p&jr{V!3CM4PxTzn0`?Bvd>M3uvgPOlc*3MXo3&zYY5R1Qg|J2X!=Gr;t&pz6m* z_z>B<u2sI-`7pW=rKT`e0F?SSN1hJS;v+Qp-Ra~H0yk!7 z9M!)<`Z$o+V_fH{_OP5+DpkU|41JbnOKVeU@agKCkQ-7uK)8PN;|JMCc)Ek9I~)lz z=!gF8TS}%BXtZrok1fN#zaf{(UvDQ?U{ShBt7Lc%7R-6Or(R*R-@Ru;))aNUkr47x`@gY0>6;Q!`5VZZ$ zhNC-m_z+oD0{W*$UMwGwgYp-)loIwHn7uC?5o6>B1^$kk7osGA-1e~Is5z}q9{R3q zHvFN5@k*5YrCM{+hu|12I4!MY&ZAnu5o} z_)%+$Ah}%ytHnYqeb?D(C8eJKUcpC_7bprd_|Il>2V48ew$ktjN$GT5OMc3eWVZ`&urYG@2~t^QnA?)z`_~({FvY zZb5x*dAikhSUriT-*iJi2w*m#g^}qM7oP6$^AnCaY2oCTRSq0X?`wD z<=ZH;+Lq!NgkEgEv7Hf!Qd!qsZqmL3n4R5~q1=SkImw%w>~!aa5JB_yQ_f?weH5iT z=rz8PZ$X`xEcLG&W#bq`>`&J83ZNOEzG|Ny4F5}Jy@q>Jb_k0LEM2&3x8%e4L0gZ! zlY?kzru}gK$t743>~VU^hs6)=QlVQCw_tt6lOi=)A?f*OCiMn=ekjJ*MVhWp%gMm- z7SndarUXV&SQv`kKllv8RZm(8_D2t4Jhc7xK@_=EtX?wYyH4JE7vtM1J0ITezl|5h zT&wGj#?tbisQf_XKyG0aDwD}*OaG1Gu4=Was^`_P{%_*tamlPodFVYGW4L!)5so23 zl$nfcg3&#^EqTX2p~csB^+`MbQ>PG?j9>N~Y3+bHg_kBBPBgqWj*)t7i?RT%UI#13 zSv6v;_v0AYU-@iKKM7;^)up@kVYuTp+d~Uxlb(?L+^T-$A(potaamVg`M4pDgi*&f z!Cgn}pqEboNVWPf*?*sJyC2q9)n&&Z@|YE?^Vd(LR)5nogUL5e#oo^jp^2*RUx*kk z0GnSAO3OFg5ojM4IIi{g92)ptm%)~20)4J^U0pXFV3ZOu;6v#_0n8z3U6nYXbSuRX z{54?GX_V|=(t!%6V;_H7iNV&N*kjiCL$?t*a7A|^#|M^Q8k}&vCy6@x4(v^=*@ues zhZ4&UrlFd*axao@HbcjsZRN~QXnsAOp<&D6SStk>N>93`qG1Ho86+?JX)_u#UE*i( z_D1s!4_YE#m*W_X{B8{Q`a}QPFD=LAHPo9*xIXyL56rqJ&{0OEC{W5X1Jlfx6XG#jaRf(o|vDx&59Bw8+Z(+WTVsX=8`EbA;5RVIsYG z+qEYr5urkl)81AY4L%-y`eM+c;fboy;TcGfQA0rU6@rSb01yD__XKo6`J&)+<@R5`ta;Yd6;qg_S9{{2^M$*S64ds|L^CO74e?i`Y1L61LRP zD?s~ebyK=$%CH)AFaKrV#`+2I&t;uWKlL|^qp5A&-*?X=BddO*H% z<3m@_8g0C8Pi@Tw*nKvPw^ij4-w59Ulp|dA!AeOLddOeH+bV0&`0gXS@Bz+1<;z(X z8PSG?gJDV;w;f^dN53~&>lJE=ynOknfEz6IPsO;LpFx#J71%Z(qep}HU0hvOWKr!I zKUuMiNo>BR?#%3+{D$URo}6zTy#=3M43oCr@sDnp&;#qieqelR#%uK)D% zLp&o#zug~+IapLGcy&a~9;Qo5c#e3Cp%Lj#XLN(D(X2$wX$|iWXzi?7(R6$h>KDFx zY|5u&!+V+&46r}jRsaKbA9?7n9z+B5o<{@PE1>SI^mBesEblrWdY4e;iP`b2mxeCq zv@rZ>hW-hU2i5z~ylcyG=Fol6Zf==;+yc`H$QtBlFi6Jw3bF;47Mku?!Q@z;q){6q zn%T$pJ%f|R--l9nOg+yWN(NoQTo)$a|j?8?HZf;At7F@CIDg|9QWe==OU`o*2*S z(#VL^I}Nk{e2y*xF=X6qL!a!%=dCkzGu@BnbH*+B81u3>BVy^bZvx@(|DTR?!?^R^ z#&$lfT0l+uuBq>z&-|aq&$aj;zX$67|K9iiTptWven+obG5$Y(p1DsEF}$t%Fdh)z znyT#v#It?dD|IJOUDMLrV=)1Um^kHHar+942Az=a)Xhf2UJaH!mzhvQ?_HsNNk9M3 z>v!;4cpt0ZVE1uMYe&-V6C zn>X?L*GJISx!%UbUXJ>MB0mNhA3<$;hj!qX_o0zb&O0CU(Qqs<-GEftpV0al@pKBd zpQFTS_6)}`xp;=uycJt7o(lM>DGZOgY|+td*hG6SM0j5`)?cxqPK3xlOKRVJH}Hee z{$6%0`5CdGYr}jDYH_LuZyiDNBEbm)YoRc_y>H*ghiX8zj}BQ|-nd^)y;bxucIrWt zYVC7L=IO9-^Dx=as(RGAmxliYB8zyDwTG>JEOx{B7L(no0a2T_fM*Pf+GY&f3S zYf9e3@)pKRR}PN0F+i7JtM2Fc4Lpf>IunPh&aqfMe}4SFMcz(7+TR%$b9Owv`vvC2 zPW)olKtK^Ze5&5g8I3=``9!jU2@$3asOSytg8Gl|L=BH(<04zEE*}r|gyMJnqS6VR zuw*AzW<~t~Ei1ZzJS4Gv{GZk;pM?+q0|)%E!6|dw#(8g^=nJbt_RtT`7sDg6&?x8U zNQeCEi0F`U%z;f0&vZQ``LP_PQ@);Vc_REW#=ElK%bA+9X%-QGm{$hX)u7Jp>TmYb zOQ4mV@>v!y5>P39T+uU13|hX%Xw%ws6eVg#cuvY^0y*cE?2Qk`kVe##Jd4W*az$q~ zZsrM)9zS5sSh2!0mR^{6rlbhu{rK%l(bdo<`e*C!l_==jWy0uNd<)0mDoA(dWfhJ- z@^NT{UN~yIXTSIS<(Ck3e^<7+=nGh_a=X^`S{lz{N=UX==tJ{Tu4y8Qg&V#r`YFwT zesf-E_S$@Glaw?X`|^dv<6Inn&d}G(l%PFrvLmUhmXT=ldwAM)SeqM*9+Ux z#OB+xl9${N(a@`xYrh>VoR-p!)(L^mqnrL@4r2alovQO}P9+?ig5B{XRvk5dhGgL=^CtWgXnIX&;B9> z%@i7@-RZ@2>n7}usy?{hiAWdX@2d{o$I~xuU(`cOs5$44wZb8t4eM_4rq9>EhyTTv zBc6K2zO1M`f9!yUiNQZV;&Yl;(``KC(p;v5Mg{G)ofu8<>>ECEf!6W>>PG zUVJ{(U%bZaU54>tYnHd%Hg%)LBY650sgXviDKw`sm^7U9Mk^`e3C8E&!u-}#jAzg8 zhVd^l!4K#rQDfrGx*<<2&p3baHrunf1vH$awrjD!3r0Q+M{1b4qroHz8T$M^Fp#ek z>o1-R6RoLT3ok=q!Kjn>)o)te7SC{9nQO|%9#8KenJV%dvjcsbPQA0_utu$QO4Ybe zT{!yJ#j3x13ejBmfww3%7gmNZeb;YfgMz__XB3BuAWitoOoFZ@wAl6%Y8-w-A?bD0 zr{rdsTWl#JxBw8Y%#5dI}lazPnvz$aVQMF zxmUHZ5>53dU&fJGVDZjaC(6^H_0RpMMRETF zFD2Nk_mN_R%JfVq&RGv2r{MT}TPzI^3N30jV{11?aSXhq>)bapfw;D>N7t2%XLPVG zMML7K^|=8tCDj0i5B$+Nc$bzRff*gf--$Q%k>4&E`AMTT=<52Ne(J6pP;8Ig_|#qt zgoHBNGVw07eCQ{Ou7D4aEp>VIh6@oTz3pS;&^%DT2|bq>Q$_QOAtFqChtPzk!>etZ zFun1~PQDX8y0>6xp!4Zh%@?S7pzXMf$Y(S}H}P(dPYNPrPTdQPbVdZ>EvvVX=vm;j{e}IW7BV*BXUe{o6O=8%(B)iD{QC5HHfgI z+u-RYEkv1BD7#hTj;A}rk(sjM0z*Yxecbn)$I*YKK4vr&MFctXUkZDvh+1e?ZP68q z=8x^vjvbR^wgOYv3xQ{joRDpo8k(w@kdN< zxhhZ7l>+jUj|m=$DX`|1ert!hJk0GAH&(pMho+9e;}=3!Fy}JtPsgSO%@H~u>QDcK zQQ`#ELWt%!qv`lkvD)qF8|IA~9e?`DZa*|8p;Y;@`Yr0K+Z%hl=Kmd^T{YS&kKrtu zO}MiI!?XVReY)!9emSU_&Let{G|#`d*vo248MA9-(s|>Y;dN zWB%d*^PPH6@=Fb-M z(I5>2Ar^;!JC#6_i8D!ojNwYY=IzVoUygJXv$u9P9UMg`UW zh)%O=O9JX5X*mDJUYKVJnB#9hyJ39m={F8slFr64)Vfo|4ET`CL%v_>X?WD=K5w|A zv3A3EHXo{*%zyb08gA+f*>^DkPoMpHZ1&%HJLa1o#DsnW5^q|q!Y3OTm@iqe4!dz1jVa!7=Y=3X~EebqIxyD$;DjaN+iuUSIt*@M#}n}mU|reLKwhs7hK ztjhJd6il~elC7R^R|rkl997TfQf zJSt$ncjp&0mO9=ns>FC~#(#xsY$^GOf;!l8$GD!LQ6&%Ym$nH&X;eS_o`(r7rPpjd ztp_mG_Vu{I#ZFk^$z8g+Ss5lGT&33UgaUct+hN_TN%T~92kc-ILDchuK|>O;K)sDG zJT!>$Sud*>e&(Dq*sv~YVwnWL4;kUvq*PdY&pM;IKUashNK;`*B`Ej4zBbIaHIak$ zZvf$Zw}_Nl9?U%tdk{=`48!ZMZ5$JcsMyr?sZl7#cRu^&18&hP5QaPx*WGT=cneTd zueLvPPd7}>ugps&^PyRNh4*0*H&I*bg#(Yz2%=2>DpzMXJC!$)evooc- zq3X)7#h2B7;^>t*(kn{|Fn6|wi8oX8pZ~nt07mtQs?~HNk~s>7I}J|maqNU{uAa{> zavON!f#UE*Q!8o>Qbfdfv>j-5SUjs$pEJW*6=<)oohR%tM;*^=I$O2PfFepBZOEea zN1|bkeZFd@<}g0Y_}I^}8I85NDnGQ&hms@S_wIWopg}D^-Y*wl0_iln-dE+bQ1dbI zTh?8ypL3}WzRqy`i(9n9^>jroVPH?uD|caD7@IO?=RK@~<1%GVRNjs0Yfnsl>sTGA z18V0Z*;{N@F#LN*l1+In%$$%oBdN25=1UI0!Lu1beW^CP&P@!rLf*B1iJnTqvwB?+ z;cC}}CI*iDr08N8b~(3dAKHWGUgkDl@!!Dng86&Zqel0o)lv7IecI2MNzir6=WJGoj=BC`ZWGl<%Q>H3I8pKrvuv$2U9nx zpZ$&1!6vO_Q&#xXp~!cbufh}C_wcG_U*WLle~ybZ8S%2AqYO2$w3}^Sp~b~O?m3zF z{P#wl{k4yuZY6L^Ap*DSy{u)dzmud|LtYKNiblS$t9DB0LsxrX*3w-;RMerfsPkJ4 z$0K+9DnX(irp1*UnSJ>GYdm;*LU7bc(j1H**~Bs{`vZn$gs12qbi<17n;RGS(x73_ zOZBKJZq$*|%iMNwgN`RG?HNWy|3|Ry`m=xAek15q2pM_TQ3I5(i;d%0^;2qn=%ETZKUktV&e)v{^8Pr7CdL9X*2I91Vv9h zTN}2I;#f{AJw$g(V7lI<@3K}qA~L=lSa?FiO(BX!d>Oo6K=Xs=S!A7NP{A~VdB1N0 zo^DI0;OQt1Jl&~WPkFv4(0x=r?8YeOch%}K#pw=Wdbv?_xYOqgFrJ-0ugVmRLOQc?(S^6)N;&Rbq5^0LA6iHKP}8sQJE+n zmFqYT!}rU0GE)3teaNsWJK+0GRzi6o<;qUe%~vu9wbfV3P!{SAw|!|xei5m*J2@R*(9=xO? z25u~~Z^Z2BaP%^(Y706+ct+j%-7#^rx)`2KMf|%uc@FidfjFtVUDFwa%My?;^!*}oYr2#x)?Xo&RzQG&SdnERx`uzutZUJfoC zgJ;gGCr(&@c3C+6`DO(tAf(!-PhHvv>w~`p{SSOYou^qkpKYh%C}6m(1Q(rZ2D4Yx zIv+hhg@&mSWq}1xaEzP%SljkWq2cNThDUE{{viraNaUajV)%*N!P3&6Gn>)6-~6|a zLrOT7;8#0&M(m*Dl9H(Q{T@gwkvkv%{2I(IG6*mvvm-L#P1uK-p!e=nWyT&_-2f3> z$L=XEw4kJ0hfZ1IB4IjQL(1;`I`sc499b0I4@+7VEXFtWA>VJt>8$1>)c5k2bJHbT zME&CbP-(1Z!}m8f5-t}R8IK4x7sq*dkE3AbJ9{k;#sbM+V(@#l5-b_6R4YX809Qb$ zzk+o|Ej7vH-#~s=+cUSu_s?IJR`@REghqVDUzDyJqt#~}jxCNOu<+U}-@5|S-)YAm z{pdPOyDwUrd3j*g4bvSe>-&DRe`g=84}fVB1+!0;Q%yThZsmdgY@x4CzL+1pvSsAU zp8g`}{F4#=!oUCsXVh*PrHO5LuDK-5Hp!pfC#U31Ou9}d=y23M7?0t57fKnF(xktgvAk{sPFu33E{#0UPc|sLg!TJW1oFK~2^1mItd({_o?^ z&_()f=)hp^|`~4Xw@d+JN(c@tv-VPn5Qz zfl4`q2N_)|6L^=k-7H?^(L9DfK61Ge~I+s~k3 zrLFXLMOmQBD!-?0h67JeY^}W+O@vvCVY;Y#DM%mkVW)gzf#F=;cNy2)&|rn4LdnDx z)alr)u=TnD%$m`q>`dAQWce*M9FMfouxx!V2)~D^syium%wJHqSLZdsoDistkaKsM zO@#FqwTwRUM^QetVn3WX^JvxMk?Eif+? zw26g$9?y93?UAGJ2zWYGgBZ*CYTEfLbfHTTOx!@Vw4N1e_~`CIZC!_we@<(e91SE{-?57)^{!%AqnZ~5+< zFi0er-S{hmx~=wHK<{XEOFUyzzs@6iX&`pu<8^i}KieVN$!`EtPR3hE|$kTjSv(s)G&;JDrciqJQ=BJ@%PM_4D+QUOx$enl=`|~+CDJ;HSu)N;8hixCMRv$BVW}@ZA(3~5)>S<{b?YL!X0jwYU zgk$>flvuN$#uo#l4-)J5*kio>gT=`Q^u$G=>!Zz9`Ykja3^edqK($`j7AR&K?Q92} zAT{gl-7W8FaU>$YHX_!49fs0JW&7ClFuu?FT@d_!p$pG*Nj$37<{MC4sm%`+$f(ov zLulurT(qF=dE&BvG&CQ%{WNP>6eeCYDv7G(z{qb;jbrQBK9#nkS8?_51017CP|^ey z>jPXY>u9}le-MWJ`EG9B*#Kl?{NcDq-e{ep(Ns@U3AOlJEpjQ-a058zTg)l3={~5N zMKkQe7_BabXFeYzEoinA4SPF%b49d$49zTMTv{lWgk~eo{OpFWFcsJF=$+JcwCuxq z>e&4kI0gZEX$5>fEdMdn*?I32s`%N!^YW7`^eaocc+b7uu%7dt)&A((^azUFs^_+j zXan^q@1ajCet5=P0;BcJUt#g4ytqWs2_UuTCxzMgB2tx(tm&hCpcu_@4w|IG;E~dO zL0^;6qz}LH0UMehgQMrK42ag$#4{bT`g_>|kFA?q{wIf4aVV8P;F-xU85oRg5xIPh zfZ2!F@j^*HzNnD5cCb@#LtYinV*870a=99{SXzPY^~isAO|J5dZ5*nXG-j6KHiic6 zKSm?F0?=4Mp5M_DD?}m1*_@!>gLZVtB756D=&!Plf6Mv-k>1rkG;zgrynEB=W-eCX zfco~3bK=0ieN5O{;qSIxm|dyo99Jd&fPr`)&G$O;K>ofr#!DCDYZ-nm{h8nCpWnjq ze1AvmE|iT@H1SssqK@#7&hs5C|JG?s`Of&_|NT6)`@4SeiSVI6vL&F;+JOAs}oAl`~^gX!}Z3r1H@;^>;48rMQ- zd>JVI`AlVkLLZt2a*CKf>Tsjs<45@tlbHuy{HS~V*ce8m7L}lGUKeZa@ zSQO*D%<2Hce(ZQZ*6pa)lro!~dK(B52__GpK1IVMXX7s+7@ofG#^+u8x8PywmDJ_- z$tOT=NXWY=`V=Z0;u{xJ8c{dBS%Pp@B$Twg*n=O-L4z4uUkryZyy)T~PP$oMV;JtN z%QjTp22Fc6%_xX$z-_=l$xk+cbI)Mnx#a!gi>g2}jo*9d1&t3DCj4(6Ym>i$qqkqb zZk34PhKchU*~4|mpxIjVkY;ubnlH;w3Kkc@)3ehx9Ijt|_47yecn0RJaaqL%Xa?D;U6@NnHAl`l#8)grMVX{Xt4cY{JiK)KVYU|9vHkDD?5|Wj%L&KpJU2L?MC_%)ramL+*lkWt(>KL4?yS7;l^_PBnG4C6!$a{5 zufO+RnEMG0Zyo5}j7o8AZEDM@Se+TIsE_lmKlg!=k+@(jgNumFoa5rqV-6!tse?;@ zA4BWq9b?te13>W{zCI1zkY~urX2d|lSEE(cv#*c(V?6f#p9I=p=hUD+3S)-crYw6V#Ydj-{NWiz99w7L(TE=6wi@5~^ME|FE2 zJVL9#qsfvUrNpnQKx~s_y~^W?t;3WQ+hpPpcE7{X4KCFIh%(CJpgTvqE<}r-{k`WM z4-C4$x_mr&8I5KhlpkkbgXMu;uGTpvh*S{9u&?Iy|9&nsm}x>5_}YwPIJl~-(7X%1 zy}UCh+3OFQd}XpVwc;++njNL^jvl5RPuw(ZIM#F)>cfw-aql3&yu_|;18X$CCm8m{ zw|G5#2=iK|V|~SOsO|CTm4yLiG}rKz`9KFArUeh2$g_WqhUs;6yd}A@am1Mt&o6VK zqVY&yADO36{i-17?L{o^To=riSrpylivoB?hI=$J#XjHQ`pf^PlX70+bG@oRE zuf|_bHvd+K`RdM(;UkMFCNMPCyxGsH3+6>LQ;i==;y9DbqVz!-`l?Cd?77!aZPiwrK5_w6`qCe~FPH;s z_{!ByaS3?FqJzpSpJ@GdIC|?pjs`vmsN**)uI-`m5u>#N;n-_qH_*bY-wlo38bJAb z7(HR=!JTfR|I2gG-eTc>zlL*4!L z)N|H15cQc$iQPZbGLvy-v&PmHpxlgjZmvYjZ=mwlu1(TH+kn)6^}$>$LY*g>Gt4H( z@Qj{MxwPAUqw$?gUKT!Bf6u^?DBM?i0o0PFCbd5Y4>>=q7L}deLAP^q#;@pRTJcdsm{ zp+8%7)8ccm{3FS9tTFAxB=jEoQJ_EbENsyDWjJKZEQY2z zxUz!plJJa$FYe{u!@lR`cXv8qY<;eU5eUI3KEXatD;4vz$=E{a;Q0lIpV@p0L;GP3B3965u%LQA5nV#6=t#( zS6@*uoD%8xz)r1?|6lqYu136IOj_WrX={eQ!_1TfWK! z3V+M7HST8%O7mMOj-@prmt)uSO z1?)Ee?3=yG=)M&a5J)S#GTbp7Dd7Qy^_rI)5O(fo`Bj~ZrVK6!^EQ&8Pr3ATc;bdU zE;e2rS(bh2Pk=a`^?mzNJsRXZWb)LI_I_}zBSD|D9I^FWFhZdz;l zbkki=cX#lh+0&$0O2>Cp4@ov6NEt2Tj&}>t(duf@C9?gmBaA@P)p6Rl)C&} zpdaSDZwoIS_Q$d8dnhD)q7GFYd-bETcpj$Xt6#av=Rm8ou*drskucOku~`0*iQmjP z8Kd^v1&wp6o-=(gjF!Gd?+#(b@-96O--Wrb76P%x666~$z+#hCgUya5Sns~@roy-r zPaiv2+vy?Ri1CBTQnLU8BBV6Z!2_$ zit4dT;c*Nky7T*M?9g|&)9Om!exTKQ_#$L<0nG}ghC5|W0DYMwyxo#ax6v~9s?GspHbj~^ zWHd|-g^`qF?hzymhZp0i_vc7*C@ds*{wm&r;aa-xd=dP+-5d374?MPY%v81KmPw`YyDQUfs7&~|C>Q~&=xKR58$v> zJ0N4*nzV$%R~VFJ$lkRJ+y7cu_dS>!<^JF2;pv6`7Afw|!qLy%4wL!n44tQ}rOMaX z(9mebIMqx7h|<*^Q>9usx~2=>sp@mEczMU)O}S-2X*j90NBjX$<`2y$4E6tuKdvY0 za(sdI4t&^G_1DnXc7#i2+bkk_3w`_fd@l^%Ebl+|6vJIKH&CA$9`MGo4p#EXmcD?= zu!6k}2Ss3d;#7i!CXF8g7G7SmYo7n~pWkB!@1{EaGb6CTzV~frP2$FU!;|S7adu0n zLWhHY!pjV5lm;E|eGL@re} zA(~Dy4ARB!I^DPd2Zq_V-Fjj*^_qBwgRkfuYBFGPE1Re3%leJ+!RgU^!^Qi6@Ylbj zkFEfykyjW0?mG6*KZtR$pKio4-_FX(c`E><0V>p6*X+L^H1z4EE(Wbqb}S zu|v0*>qmFMg0Xg3GwCt(ooN}dRnvmmow1swUIvg>z)0EV{0I8G6_t9v^`o9U%DF~g zJD{tmt3^_R4v})J9-2J=1%2$7eCZoM!-De!eXU zRBgpE-C9{FmRp;6Au5T+RCRe?-K*j-d*J0OietvoOAR+$7cFCF(J0IoNk|2(^xz&?k_D@yvZ( z;Szmh7(4lWc46!@RQKDSGg`yOTStF%Z(g+)j$XpFmcJd-+e$LH^zt2JDpdRZX%eoW z@tr}(m-sEM+%!KB$nue~Reyl?|Do2n!1`s4{~AA@vF_?~VHQ#B`+0pXniA{^ZFEdY zfzC>3fZBOBs9FR1ImLa2bT7kH@|ZrT1UrvFw)T_3<-tn4$f={@mMAOQ z;`xWf_t3x1ePUPjzxYg8qF(YRJ8Bnl*gU~R%cI~JbUv2{U)c#`N>47ce$GXgsX&$J!*#~7@Ouu)dARl#E4r+b!-3P0WMLb1zG2!UE+1t452T>dUF}r6UZom-X zC@X=EfSLo7ow=DaV71t1digycbV<&5HNU}l=mtu}gC?E+?Zd)}rBljsFn{~w*5aXh z=q>b9W_Z*BOI$-T(OjJy)-5&5VXt4K9N+j7w`%Pm67D@~w0;Oza?+Z7x6nMKzX4YUD-NpvzS~fs%6~+dCPM#fhea}Kw@ea#t$@D+CMYM zDFFD7>Pf$QC}pGtzgqbH?(((=RyX%{eeVc}gQ)aOIH=`7BskdnU-^wLu?X z<8iNb&zVHopk!6^8#tmkjGt}zx!vFlkWOS(RaRo?EiIri)`sh&K?#gp-e^OC+sQR&Yzh6Uk@sn~P1!Yv~6PF*nmmk*W zmC;r%86f|R*~B#~3PTomT|aPzqN$U{8Z#NTK$>}-eBg5x>a$y259z^p8mC2YF$%|f zVfehU_A!k-=sKQyruDZaTAG$wl#YJ`O^0IqWb8Ne{n5@J`5=3MeX|T|3sFKJ*s-`E zvUBK-*g+i_$rry=@3=t^i+0@Xg;L63;|n0iJ*;@KwxRA$JMK?dQ>o4*2O0?Bz*}C1 z|2*!m??i%Qy#-Jtb1T;^H}p^bvxlGMCIz0sywPt|{055)J1Q+=ol8YGTn|LdMlyW< zeRteN7pGR9LBqB-SqG0U=wBIqy8EaO5VAD`_i*LIT=J=9;TMuHXyJC( z`0$4QURWBx(*4ibKqE)p-R_>=^`G-P$C$ssIM)i@hs(3b3*0b=(;7d)^#M(LTFJsp z6%>33PZ*DW0>e)3wZ0kO5mEYK(+*NQDlb3qaeJi$TDlxiziu{z$U>4oT%xf2{+#`G zZTZBFI3sF{uFGRNtgmP^Lzg|VX&4O_JbY`J-H3)aml|Ag#@=s}+zz)=r^lFojkGiN z%i>{r^cSTDKLew`M>1J1G{UTjNU!qT7@7;V|D+IViDo9L9B=++z`*V{)6D8msQ0Tf zMTtKWVjD!=KE~ccWBUby?l?_A@|ka6dG_n$=vYX2w?{niR$p=6LyQ2mGIeZzej%vd z=$OVp;Q!F8V7}`2;WCydDzodeH0(CVF)@$%^83(uh@kOB3@0DuBQCTVJDR&38GHJPxd-m7(YixcI{Um2Fwl)KVWI+%PgS% ze|i4u?uDE$8{WeLP9)H4HwlgUkIgvn72;Tgvw2_Hb)oVPeoR{&2xygC>iph8tbSc{ z=WufXC#I88+j;du(nn7;zr%RFH4fX?-Xvu%?Xn4n5qr*}CVqzhe9qa_5eb8w71VQO z+wRJ|Xf!`1=%bpj4dy1K9=V3o_;z4)TY}Xa%`BK7IGFN=Y=`>PC$2`mS=ewr$}ax? z?}`2B+pW#{dxbaTL(o9^1;4`vG<+|r#cLWwJ?KFT(|hn{&oRE^qP{);M(k}szMRup zqd*50PTxe&RdoX4Lg5deJ6NA!R;OiVM1dcqYdR)A`8k5hV#V!LlJ)>O`$OIhSxIP* zYb|(@Y=)LjvG>22!*G0KKVLBL#cQ`7JZ$AlD9!y4MXR-a2sCI*0 zl{yI|pZOQ>9xcq{Hy}}#jeU8|cJh68-8snHaI;g%u zXLiH!g#K55(od$M((#w(h*JbDndPL@?`DS4tonC9yLnLr{lUX4{8Tjac#>1)k~>t1 zB^zFR)QE=mnB*60Z_MAAKX4gE8;`_BZtI3Fi1gsM}$EuU7O0%~c=*CAnMJPW_k&iE~gKsmd6 zYhq3ubl45>JCir)zM*Mr`P{m(DQIeo`%`9m0p@F?O0LZ1;FvPvgScA;XvYoJQaY0} z-@~BX5BjwyMyS40`F3PsDa`zpIkZmK0;~ML4e|mOQC*F}oTc(IjFj`MdEBPiF+8Jn zS|O9pbJS$VAI0XTjuz>fBbxDCFfF~uW4Pfeklk+2C(RmUUX=vSvh&WEDYupMO&nVPNDVlr)^n9ej@VQ6sri= zCmWt~>hYP;Ub+o@4ga^lbd{N1W2haqJFa>D+2w+I_0kkpe3`#e0dr#Gx`GM3j;saan- z?~d7*g*ye?=j=<+{J;&3h16WsACWflJZWmfb^1nP4EOCmg67OG_2&J?^b8tB|L8M2 zZ-{qb;`I6N4-VGBkhodTZOvdH!HZ{Ynp`ly|L*&9hkbEuBl}Amtezp#;&tCaxsx#I z)pKS}juA(1_VMOHpY9FUBLw}F@-;bvsC>Wj11!=}H?wQ#(by)``NhjdJ#QBpzbj+8 zdK<$R4o5#Et)^_O7Z9)Z|M*~Bi{_8{Z%yiM{eQYn(2%rI_v5pXs8OW1>c`Rn7<9M) zyWK_&#dI8(be-=))$_ff=F?(mU?JqLHxmJk8?C(gsfporr%w4@jJCq~#VQSqBNgp1 zyy7IYn#~Z?7TW*&12uC$JhMiG_rvz)O&&0rs9kuqa0-pC*_)Y0(C}2KVVQiojA<)s zICi#mR?iYvonxka&tY}sDSy$Wi0`Rr)`ss(=(dZ{pQPrE*VaSj&Vm(!=oD&A8I#Qj z*Fz)as>?r`P0_ONnXcwwQy_hd;p6unLgc$%oF1n?!>I4^_i9z$h?JgxB_IUD1+}e9 zn0el{MuStHiB_8vQQs@|!$R+|KGzNr6Xk?{4(L9<-G#jP8_kGrWjSX#3N1L1^(G61 zGHbQvJw8RF2@j`l`;}w<&10G$+sCaib1d|{xJEk~eU|IC=QzekMo2i3JwJ@`2u54Q zb1FuaqhVW1t1-UY(4qW_-=}K{QOG`>5Ap11~h$mVZFQgCF(t@H!GmDf?5td zpIKE{f-Lz{_mk3N&{XZGiX|eZ|I*hwYJ7*s2T{pxVqocl#`a0w`6|2tH$r>fZrP@( z#-o*JNq6*$&SQ}cM&nIFls@Ci^sg8WcEa;= zPSTd)|GfYFw}FFMCq{7Wv7=r17!6oiuxl#+!irY#rc8HRv3{({+oJh){mn49+^xFX zWdyBXv+!Qf)k8xOwhzq&U!u9&5%xdwK0~X~trrZDQfOZA%UX7h(Ip3|8hZ?|9kM%XYvb1Ta2(^|Ce z6ow5}NM}CsB8tku8dpgsj@5CU+tI)shNX|wHNC;?Jkio@^ST`7KdsYISI-@b02{8y zL=LQ8hbSkGeJziEfC@b?exa;v@Ry;`|Kw&zrNdpug zVcKUq`4tG4o^hn)$-=~Dg&zf<@4?Kko!sM&en80RzeBu~20hj}`;wNixQxR0alXp} z%S#Wjt%~L!{S9+x+U|x(V0gO*%b2J;zd2x~I(hfqzzSFi`s}@&!3(|I0SU1mrC{=8 z__Nh+jL)DB1_R}O`oP?kHKokMw0tpCs(A2y_lrTZn=Nd)ylc@EN20Ni8Le)OqcdMh z`#gnUwb3TTD_jVzr1XgCRx7UZMHs(I#j9lQ^!!PU~Alx8oS} z1-~V1#rW3hZePB)2Y(#Mdm2TVRQupf?N(d%Fc~zk>4eVfy4OJ3`6$VS!yBfqaG%TH zv7wI#{jKUP^!pnEb9|f3@9z>r#ItkhaqJRkTCggQ8WE0)n<)Y#^CigTo)(j*qbZ`W zU1FJkH;(yz(^02P5D)53GTCxbIl?l)gN0hmB z=r9`mn-?eU|DSxym|vVR8GoqsKY7n+_n8XG<~W?uk0#}0C!W{Q_{yR4AyKkE9OHi_ z6<9>dwfFt!by^$?gErNr0QC;!8t-oW&vB7TjwJete?Ut+@7!Q}hV{=4^_oO$XwmX+ zC_pLt#T`xuXmL9AgM{_f0pTijpYfB6$WPloC)`&D{TX1}Ed9?fo7|+dTVfN(mKD$J1z?=S9`!XNEwi z+?MamDFcMtn+xvwVSWR}UhC_#cxOcJ8Z>(G0pmTIc(7i7-Twp3*?0WhKQ4=iLiF5@ z-WYH5MAT{Bc)KPv2bZtwB;SNN>r2nBAO3G#fVXjKrp5~%pd=`MV%+tg@y&+$?ss!g3^{BUOHCT_jVwmz4! zeAg%gKeruUA6mN8o4$YR5X>L`5x@QZ27O^1i_W*1Mv?#Q^TzgnTi5efBQbrS)_bFj zuOkESjB#N>PF4TuH*C1hM3OPMw}``{FwsqY0Mp4Fkke$Eb08yvZTh3`-d1QETzhQ8 z@-Low#U`QiG#LgR6vs!q9{uNai%J!qs&>x7^mM7>&_V_@xU+t~dIRhCX-mg%eOtdl zA9cfZ7C)alkj?Q7@}wVlK0olE_oK}py>e*bZ|=JP{X3Av;x^@WWTXCDIs+X)qR`Nf z=}S%#G&>KO@d@wxs<>d}*0=4-{yornGWJ|2U>)Yr#*~?jediMP8&2w@^q>AwhdRGKJgfBd-+kT6MR|8_)Cn8&l)tX~SqoNtgI>IdyaNs0 z+EtRh$D!iuTzs-oG-w{z;LC@I&AjfjMP;S$7ykTuq_UAtNchId6dvuOh1`&t# zYixt|0yT`y?rGou@c;LyDB;C7=uQ9Ux*F?Yo}hmC&-DcBk$iMImTzG6WG!9m=TcZ> z6TSG{wgRZX*?Gk}R5o0%RruCj=I=Vt+t2^_?b$G zBnLU5p}|yT0h}Tla!xxp_T@Kf6DaxX`n>_OGgs*&58nAoJ8yzW74fj90+nxkPnb{; z8kX<6!flV?d>-X(R`GAgqoESkYJ8pxn%SvP_QB;}JUCv3TNOTvsJ?A^!}hc|6~}3k z5~%%r4Fp2*FvgaGTlZa zpPIsUxR`8MSIg_Xy1}VpD6rm0yvbe%GWA#7K69KzWgZ`S&ff|Bw|*oe_h0yi>fSyO z&3Wbq#dk$Sw`jkB*e9ocSlrzLQLb?epB(yd3`_jqDOOnjTQk$^`G;eYKuqw`;jOC! z%A1_~&r>$!P0@&$yVfjM7L0TS>a~`0p=DPNhw-?(sP=NVhxm0fbWJ}5KlP*!Q7!E~ z+2aVPZ<`n06Os#_{=MG9rYTjJPCGRwdpsS-s3Xbp*x3o@Pn0L#wDm`1Tw1JY(i_xs zelp4SA`i?)9DJb@oq?KOw{GUpmPM2Yr}a5iEx@Pvg(SztOtiR{yXT7i_%6SDoH9Wt>P=zsq`^>+u4m~P6RiCd$(2|%iKFN5Yo18r#wov6qk}0=uu4C-?!d|a;Nh4&EPqw^ z71B?)q-}V`An`Gkc=`QqV1wrCyhj(pU}EalbH3_X7&-ZoQ*(C}G_OoIeVxCP#3Qq+ z;34;J?6EXWKUQr8Jy*YLrS|W~g7ap67d;4`{eY;J!?yLcSj{Y~fURMF`?+1?{zXF1 zeuQy)wsnSxJ2QlLx>PN=1$~P|R`|C3z-~Xul@iGZF(XiAsZ7`g7+rj0WT)sgq&&Nv zd^LxqOASq$SIXueybX-!(~A_7|Lz-&y}M`j5;_L7TSHXAejNzH8rg7((CQo%Qf5ZzD$yrz-jMpt<`01TvgE>0 zCQ(=ZXHo5x>??VoC9kYprJ@TJFZXmrTJm7R+NT9uk}dL!z$uCMIPpv7xl0` zxBY&>Nq9y)w@~HOuN~`6V6N98nK0yf0%+-7Hkw6jb)V4h zrF`M~I~Lvvy9VN_B7_|}hI8CI zi!N0>+o%Bpm9odQzxBWb9yp*ctc>N?Yu|8jbU|;4NJ>REfgA6h$1fb)oc7m0%4xgx zVGNo@6}L_9AnH-4jy{rI7)Ipj`xL#leSB#N6^p+uZtk!`hE2Bq^qv(s3VupYl?i^0 zcB!g&2VW3;Xq2n9BtA488K&eXoMzfw-rXOA8nlHBVR04>=$B|q#4p8Im0q|*JQFF3PAM@cQ@ z3&gx5nRf3a_;S+xJ%SUSs9{f^;;LJ3OGtcsPgh2LCE@gZ9;vBnZ)|Q`eKUI}k=n>0cjqV3i8y};V<_f4=O^s0Jlv(8O~lvxYgbD0-3tJQMzwO-={TtTJYc$P z`&sC*E=ijI{RotdKkt@)eG><-cj*?Yv*{5K{rEArVb8K>m=@|(iDTSY`Zlxjl$1Lz|^uhJ3xYiCQjj@&a{j^jqpe9T|6;!r>~ zAeLwhO4AxA$kQPy0X5XJ%}?gLgj5J|%IPY&7OiUi&}qqiqV; zQa_&!jl8lFE#hmT+Ug)bDK8mXkM~XP&pH6@UTLe3d|}mpz?e_?`>M@d(6%r;wZ@b^ z|J}X4nR~`W3c5ZYzj6QkR<_@V<*rBUTFl1xi4|I$*-7QBd@+`-x&HivG!Z}TAZaBW zPY=S8jK(`=-$K~#Z|jim4y#|l_WvC>w6osd<%E{@TSq*md@*gY?yt-H)v?0uN$(Ra zHXdKJZ!FHf;7H(lo8sOiQzHei;A4XHtK`p^F|_4)IOP>ICcYiINL6EfUp|S~Uwh_% z=FOnx6z52CZ5MRxjdc8S`6!7)+#=RKSQ>`j>8_wlJj1d@Ui+`9Pa%EZ$mLcA5opNI z_w2U*=X0?;xuMfW$d>Ireb)DfL6mJjMaqtn_kTjQp{ZaZRzbfJ@2}_nqRvM~dcx`I z-Ydj8Ga*4XjTEUs--3*Hg+JK!3mY{CMEVE*Y42mlj)Px1;RVbszkBR%`|v-ni#t&N zDv#?WiSPXIN3pp37QZoestcCCa;#+QA4dD}GPjZh7)a66FOy4x z?lU>bIfouWmyW4;m`^13EG#*F#zP7sbCSvs88F+stK{DLg-G9iD8{MJ8EYQgUOChL z9TD%#+}mpib^UFoJ?3$cs!-?u`eiOq4BhhrT_XSIeCU?)`jGdT3#d=1TP8!cv(FdV z`8+DN>CkRG7}WZP3kz)qpY%(f#v!W%&kf2(F>B4+ho`&tVMoowm&d=e@Txd+VRhZ@ zvu9b)skp7SBazV*H1E1~-jzj}Y~E*Lx&?eF-JsKfj6*l=O03ZaAC@nq|&TKQnM zaUYr#zMNrgh+PLZ3Uw=Qg*kz#wgZf5oLjN&Zlic3wC{*pB_%fs9V4^0Bf;X(ov=Ud z!}S$Fi?|Xi>|g|K?zel$y4PXQIrO*VwIN^_^M@Z?{Q=u&oE~U>?1tWGuk}|oxv_iy z>X$XSbk_IueDTQCGlOI3|5zyh zGcL&1{_@t_{zh{iFknLy-LXX<`+RgF_Vp9ELwa(+sosrQIO!qM>Yy2h)cJ|~f5!fY zK9yGWb;CDi|NNfpF2~&Q_ov95OZRO~zdjCG505T9o=Nyqk&gy5PntWx_{qFc&Hc7G zx_QV)CXV1=>-HQqJKe{IGiG~FYvgKEs}F?zkXf9|tsh zZ4wpB7{}qKpGCK9If7HKZxpHto+NX;YTLVak2)kjp10ilvj~h$=_>1nJ%(u&8dvg6 zIHoPInDE;fio?Zp^GyrA|JlE>hNA2G=ayjWbf%i&rbkFQuf5??eLfC1`9GO8rLfJb z5z&EgZ3^SI{clQ31t?vY( zAz?;EOY)rGHnJ}<<0_U(Z-040_!*2?U3nyZj+O7kc9r`J zKAemDdv5>aRmhLT>c;_|&B6ZI(|i2G;qT4Z$ZP5N{$4RQ6dpgbbs-r$d=_u8_Z}wh z=U0=PN%X{;0q%&VM+EMAO3-I<%$IM_#BDkvt-6uukLc}jTRVK9b;F^N=HVPHjn*;R zAT$7_;)SG085Oqi8{PWTvYwHSnR^}vjbsuy)!Nq5NMlDLZp`@gd%>MJmc9;B7Z>e` zuIj*%n%%kccd+q=v#uN8@I+_p(8d2XKcHD|&*HbdZ1`PhGj^Q_gaN*Oqc}w(<*85L?a$1|R%Vi8M*^ z=pidBuo9f$&ArH`V}nz>{BnlhmqDZW5xdFAFi5f6fl1y+aNyxAuKD!Oc?|V4uB*aT z2tFZN&a(%aOL}q8zuiVg))OWgjGGc-SK-J2_r2}s58UQOqkn`!g{~XRk~p!?eOvv+UTtC= ztK!Vfb~q8^B_7)L$v*`rTxhN4+eNszODxUsm)5w z&f{3Sv_zcZWenxxatyH~U7M@SqAueC?^U&aa#u4nty=s2cTaGvtX9IH}eDf@>0oQ$ zT;%(uvp;IF?GX8S>UOigew?M4P_7g7Ca-!iEu@OWx5@`SiU_=QTY+4%ET0=r$bZ|G zTAG1Thqnbd%ZdYiiHCunXDd|uJ-FpLFiz%skwqDNUV}w91f6}oIgz%j|FPxpA*3G7 zS~mGY29ib{t31{db_?@C;;+>CH_3c{9=g|F)+0^tX~T|tt=Jp(c2Vx~>&Td%_Q>AP z`>%1tu`g+6qOZ3AUC-qHR$&%j4&?3gvA$Myn(g|AOs4jSrhnos*skvtolm_-X3L`> z{myE+r-#|_HaPl8!CS$%0lRZ$4KArf6W^&E6;ghX72m?{Ve!26j{WTO?$?yno&`hj z<{QN%|FAJmJpOSg%BLL19>rPps@)=SW!(tz=S;%MMEV2R`!*KewU-E*7ES&`G3VGUmGL_J| zqMYO7dO2;1G*josnbT~%xWsw7YQWLS-9De6IXHiS1u<7hAd#af}YnV_gkj2z^reCg|8F--e?vUYo9^5%pY* z+g=ZQ`sJ{rq5XSdT0D%rGn{G_(!la(jbRM_l{miTV{zvCQ<%5-nsR_Q(O$Y>sYWQ7 z#b<;4yY=+W+Y$Oq)U=^z5qkVM7^GSKLxu2DX>kaRt2N}KXquiq;nv{;PeALF!2QrPr2|7pi@DSI;RQ3A*u z?T=r@8nNPUIKe-9?sO=jU+}e7EFxhyp;K64lb^8nur2mFm%n(Iz4@Qt$Jlnjtm%#n zPVQbR>2Prew14>f2!W9fqk9(A6Duh&OEkEKV@lwBiX%o zL{tgBy^&o4f?9G!-lEx(L$;zdn#^_NV2PAoG!|9(zpCBM;^8B6wx4c^emI15jd<>( zxjn$-wFmuCiu$90hT&x++3T_EONYuAnxr+jXtcKb!WcVO+e*tbL76PRc$XE$t%W#pTQDW|}1OaUv^rd}Zk;>{!Hk{K|n`82b3SN6P*vFx_SaidS#M zvO7+ur|!z*Xu8>fEuW%DeAN}N=l0CuP~(`4_)9{!i(bB1SYo3H&_<`+u7##y=arPA zLXi+Mhk#h@q)tEdJ*;d@i9QT{rnlc%PR(K0={JF53tXV`;M)dj$4=Hb@~?A$yz=&6 zJLQyGXv@b{NV^qC-_x)LD7GaT5iu*VyoUE|xlbdGQ%3!QzO(ADu(`IMbCipS`_xLe zT-e}v3Bm-;Z?;_Dj=gH9ZkTux`8@hX^NS+dJxKj>YqYR618Q_GT+y~$h@}nYLy!uqWBjJCHC~J-E=d%7^>kcQj#BDyT7JwZ`MN-`E9>f8T_L^wQZs=(p9hv9e zgF_MvDm?;^;z(3S`PPG1&~CtBRMm|HwI71GBPDGAuFtkruAq*CBEJtBBoh13__mPm zAN^I3k@Z%zyqLhJO>X*gdUK%^QboN)p8h$9z1b&CxYnG&DoUMBYa6S7ak9rKXnP@{ z|B}Pw5j|1*n(h00@;2MAmFdIMflKP@O>sE3t)q!DJ&)KQ9#uwcu_Nr(GRolc7z+=0 zwOT*qE^1+q!|0mzrt|;%xg@S%wR}r2OF~2RkpK7sHw;uQGJa-H=y+AeUJ@yAazg42 z5%DYKMBKE-!ROtZ5h5Pb#9a`(`|VB~4cc&}_HqvPEM7JC76tKpoM_o?6DrgN2s@N| z5PY{qn_@|K6@u8>*}HC6VG;)mpMXQ5ceG}WmSO*q^9Al7&*P9-uVOS83!jU_E{0mR zj)b1~1XEEhb^oLPelD3Sy}dV_M+OICYM-5-TZ0q6Yvg!EiF{?=202gf04tcz7js^~ zTn~&!Ql`6~Dg1b?-(s2E2w%7#?_1tDjlGsfH|AF+Aj4&y)Q);$uLI^s0*%p=!|ZL zya!@D;sR`b#lP)AHKyQB9V={mcQW%;{SBm&(|wLRw?OSp<-20rx&C<{&2rt3xc}gJ zW)%EBa4nTBE`}-8RC%0ne0}T9Jo+^(o=wr#(FHP}BeOR7 z;9vWDySL9J{|GGc?g)$jeF`g{4sYlbBIBs_=cX3Xbk@A*3)JClzJ3KK^cx%RJ&?ox zPZ<-l6~QodXZ^<+SN3-GuJ3qskGT^WNBzXh7L<{B>_ipvT%wVA;9}qV?XPfRtDL~R zQMNi=oC@%J8eBPt?b3n=cRV8a+*^iL)$rR~$BLKG<9>%&Z^MCG?_`DXAR{UM&Ui~K zH2+*CX1Zh?+x(i=r(57SL_-!Xk zp9V)tKdq7yG{DJgA0;HxCLlQ6!|C?Yx7hC$#BG($hI7EUv%>;+5Aj3Xx`fTPM_BnA zY+NU|$6=5-cT-(*VpDIh#vPm0NSi08*z`+q?6dT@mEQZIVQtchr(gK6b4$DaP&r{I zq%;K9Xf5Z1yYshxSXv(lJq7dRJl?R|RpXBhH^nBsVf+7hY1a+e)kOW^@b7)rqmNs0 zj87(I%!!5*2UmuRJ!`-rA-i*lms?;^*LO=@FHx5^NfJ9->KBih?rHwbrYt-rwkyiX zUwq2Kf8wa(sus(z7dWzhENxfF0vOdhOVvzc8xPhvd{@*DjJ6m5U-JUWniSXG4!sJ4 z(O(K`Vj8hBd#w4hz618XUwXH`jO7PH$6A$m`;l6BV14<)STqS|d`69*pJ0tUnOnW; z#zUTynBQbIdO_$cDgt)_T zimONv??kb#|L|b(&4kZau|IqI=-8SqM8DGdm3%)HV14Az!G#{nAk*DpspOuw*nYxs zCf`NouitY1f|MyM4ugfmUbf#kCvfV(8Vh5=9T1moMPGKM7zeAoo6cD)V}bEK9zH{B zGN0P4_MT5tQ0Zd*R#Qa;a_cRv-5uEZP8pA^OOM~0U>ldLPa*!%hvwm!QfIbJKpVDh z)|m3daGX3oCv0}<9F9EZl&+>GNgOg#}$_m z^NoH*Y~AKd4md6pA|`i-Ek21;gTq#L&#~#cBl${vNs0SkJ8_jeW&4eDBrY-jx5mG3 zkQSznEzCZig`>fu{+&x&Vc>C<5Q+Qm{`XrUbbF{I(Vj7N!3D}sA?WdM_fHc4VQ&OU zs6-9gl6XS|a`vug)2j!D^{w{zKmH?s>9UYp{{vws)P5~#{~gJm&u3WQT));bkLVxc z>TBh0?nwQ+_|EeYdN&z6nx)US{l<Jf3_acu_k2(ZQLa5VYR0z(frd?|LV_Kh|SYK6=w1#+W!8TR{bcuv!D z#DPOCW4+=+Feg8zp`T6cf6RjilE)Q;phHG#l<)I19ABory)%f2S9DAoZ9B8C9rLf} zQu)u%`|tM={nD*kClsDT{Jpg}%S$E-`hR_Xb?Av74&3Z)dsg(E^?U5Yo89wji;=!` zIO1d2Tb!6$7PM)Ei5-T1sRk~qpmfs%1NugmeF{TK>-DYfe8tS0@$0N|`>|`~M9*F0 zVVwGK{+qD{;5f7Se!<0)(DY|b%Y~rr$e{3SoN6cP(Ym`1>7V%J>Z_Kn@rT+S`jgN0s=I@7u=`wu2!10viBd?L1g7teP8pn+4P8eDTxx-e!l zZ(Py97{{9mMV1w^@Yq$n7ddmGq`AGZ-XsRi=7wxW%r&GL_~0 zn0}qb>j~6-+Rt6~?nCPJ=U0~RDaX-2&ZWDpHjp@<%q{sgU5T{w7tUNA6T#{OIbPZIDST*BX}z8#ly`>7rLd6bu3@SIp*!*o{5`pX=#r$PVco_BmRlQXB*dM z{m(lc;)(je+-m+#htKDb>Qff%e7p$9OrJ2jli2ILN14`98$N1b-on`qWvf+0KW{#` zarN6rI3qohrX*{Q?G0O01h2B{WN=8QTuW`(1bW&H-xZkFAtQW+OyZ^^*tfZ&@j@fZ zo`pP7(ac>Op_nF8;8D?a3sddu1>)|n#P08L#R;8>NWSZ`KhuMC-ooeb+6S8#iR0v( zk1nxx3OHi#AYy&97wOrP`K9N~u)W8?b@M(7F%L-%k<&|JpgN$?V8QJ}IC^E`xyLmE zuU$S7>Ah8Z76(mBlFghPadZjuzPCSd4t0H@{#4eK!;a&2_iObx;h3|@OKH(Htlx`c z{0Fy(u=3eZt6#A={a8LRKlC#H2o&;T%3%v#g|Q|7(|+vwJQ|aZu0XDSeZ+h}z=Uej zLF*5!yacvzFWY=8yaT@QNZ3ZF_rS33cuVOCE@*z5{?J=&-i8Xjhh56ww_!!F}*8x6o%B()?!32a05K z>1{Fn(06|Mk&3eCP-DCfeNX>{PP=a|eBNJyW;NlH@>3isrrRBile2O9_rS(q(z~#2 zYh?c;-c2N-%MTaxeLjkPicRF01#1VgB z+Q)(Lb9?4Lk_jte=XH6zEULY|VjLN*3Ri-b5OHh9!cDyMZgxQNtF=?&lD=4^)cC$m zh%GNltWVRuT&*E&{I59qUR6p@^E*td)_c#3X89vf6{W-V>d^NKOg&q+xEe6$*;f53`#9 z#>yV^21OW^-efuTn9xZYsc49FA#-By0huq&hZ$_&Guk3(t+!JKYk~#lZU4HE#P1m& z>}0+f8hC?kgJ0Tn54*(Immv#K_>(u5L>wr4h0&}ujm-~d{hl8-=Vuf=hJ(&LN^53vVQTWnZSF-WSXFM@@yBui z%NkFMJ1?4#ufut~jr&_koIWz+Sue*SJG*h|wZmf2AF*!7EteZuEdrdPZ*sBQRFbRn zBT=7U_O7HwV@o)eE;>bfRc3%=A)Z#rCW2U7AT_mdj{{~;GMolERB`y~tNvgsMXZir z(Oe(I19>9LngfzwlDL}2q|bJgV)h|F$N6tEVQ@#V+%`4VIR}lNF*~>J-UH))(g8ne z46${W7S(X13OW)LJ46*#kS?J4P3X>f>>w>Q+AlB(MKgMn?`=Nf&q9}=vrH2F`MNiF zxz=`=zUKDvfbJRS_rDWGTN?v)DU914TX`VGzyIqh0X5b+%c0UAcp&;GiSKAsLTipM z^e;F3pzIj_*DpFcdTP-Hs4#LhIU@fD5-+~pX!MQ;7;)z(4;9YAP|c>Hpp`qYZlc4H zyS)yY=dAA~Dvl#nm_Pr;3vNiKiPdM9SK;uq;R~DkMr?~{;XbiuFBa<-ZhN+ii7n)% z&m*M0k&+^?YqgUx(to>+cZ9@}xh6f8=I)eX+2iwPmw&#>Hony4P-*GcPSC)#$lp-2 z5h#UMZolJWCa22x8|E_i|1hN;f!ULi6U`i~{2qzBq$FtD^WLW{H;<@1A0;y|$CGz1jt-w9Z=;;Q@#fc*g1dgV57#)k`Vy z$tLQfN9GBBwBw=S7)t3nml$IE`S-E)wT?)Azu8sJKMY0>tp4#-#twcS_1<#HC!EBg zCUfa-n;?FXSW60z(S=%=lYF8egv^)2r}yn0L}sQWFH_42Yv+ne0Q`|be1bA0U940%0K2~06oJ%`V`N57_qOsy?yox*2+G; zdi55AB&5MVt=TpST`9hs4{iJpzJQ-k)DF7lNyQNKMb5WWA6(z(ywbSKOSP|dnav!1XjHR7HnNk zH^}{lZLNMcYAz@teR_uy$)+0;OBU?W;P?W+MJ}M!2XQELT(rp^310TrjNjQKSy!?4 zlfV40#sB0NM2AVeKQ}UTVJcI z@6rnWkm|pa?f+?RJ_c!2Hhu2@zxVH+XTE0!^a-3%`J}dJL!vDV7&mT{%xHnJs4LMN z{*e%}Do?6b{w*{;n3P%ZTn)X?#5=hQ5%s=pn;)3EPR(MDd@s*W4H=wx{82q4lsE^* zf86?Y-e?|la&XUv4cx)D1uBY1Cfl$Fubiq^GsJq4=byO+RG_L&$S$v&59ZoS`3qD1 zvD;+>@7Auf*#G{AIhnK++G6N?iXVN3maiurL_|BVMI&$Cg%J&8MDJ=jnED;M+!kb9 z=heqvNP89`_%(XwVPq=_;G`2yGATKvva@EAf3e7CDD>}kib%`5=Xn{IO zcLI}dwyrv7G761fZFM!Nl_ajT+{y&aVKS#iv*%9}L86~5&hPxZKNmWyPB{m>QHI)N z$CY&vgw9PzkJn0dhg$4@+F++c%yS(0p!)In=daN5r9poCtv)QDzN#2&MaCrUtn-;i zG_YQH!6l>UzJK;JrBvlaU=Ldz8jKx#bZunuf9P08tV!7;OYna+xXE=jPR+8lcY5C+ zjqGujju^BhTbF%(Sq&Y!LM8GF1P+36WF$3Ak)3a_hk9nyllyG=0BmYj3k)prhQTAI zlCwuP!*JugXVt5{V02ws#+MC5eR;?2&sFCsdeFZz!2JBs1(-a&S^9(DeeCW%{oZrS zQb6%fTR3LY;HRJ*BlPtpthkd?U?nMzt!JihaXJpeQ>(MhYR_w-E7{e1d1DtA=IH0u zhpAwlykJGUwgHZ0-n#Ly`3L022GA}u)1bfg*5wtSA0T6T)_ZZ4h*>D~91go->Tnt}_OG%3083yx zHFfuMgJkU4(dMz=IvFUU+RT?{=AcY_o#IKQzx^mHW|2@rjxO@Qu+J7NcrD5z>y*j7M+cMZRc>NM z@gFf>S7V^B(f*~Yse(P*OD68fNCAV+x$jBQUhosL{q^NjBy_%D7D*3CA-=lvauux_ zt2ml7cU$xzm6JC_^EL_Tb~8A~4HposW~%e_mnzU^tsz+p#h)p4N4zV*d8?Iq&YS0r(iw#&ERK@PT* zZydX+bO9$8-g=_y!2yFk%QlW(_rj#DKNZe4$Y9KFQoNQ+IGL++ba@l0pTxt*ao0cD z8^_E}2i`4R1>ci$4qnRQfFuB+WM9OmVa0lE3rukN=Ql* zKQ(tC!};XRFUA!lp5AD)?eQhR5a>6P-uDxFU9a2n@%KR6`Qe|h)&cEG zB@6iIgwAMvUr-q(!49aLMy^XQE8y^?lIiiE_n_wf_E!djf9);fd9|6nvse{rp^$h^ z2HU2o6_3oiVfN0Cz4KKBAnHJ()|+}34+eDC-uU_~P7=DjmVS$FVeK>UWr15d<;7iK zwtETsS}ugqrd91acm3J+q4t-rFQva+2DE!e2cIy-A?>j1PmRX2Ff*Lt^hs_#v=_Nf z5C0+Zc$vl%G*f%F_&mgJI<201kibFq8y_Fwa-0DMhlVK!m5I&q5!r$K`aoOroinbc z21mt>%m(D`+2%W0>-EfJln#FNzD?uV%&OynDYKw$L$dYIXg0UgpTi#qY>!<1R=ylZ z#DFe!F>vMniO^WSN;wEl_C=A3Cg>$y|-9-fCPC*t)RA@{GeR=%XEU zc799b(I$4SpEL38#fY?%0w>0E$=uzCyFc9GhO%x~<-1qSu=s}7p{)47{_q9Y(Bnx2 zZ(6ke9-|F0jX0#*r)a`u1{o*Y26hz_^%_w!Npq1k6f#dm!o>yO+4*ftw2$A|6aNs$ z@&;DzU9OI;0T%M(3;vm306NJn)L+B~lM1$;xZOtROO1IyF?F4<43DVCUsWry;US@e zPtwa;q8*sB>c95zbN}o4Y~#mJ+GNaqdk2X>^s|&g#|ikecP{tn_;Z*PgE3kd8B#X{ z_=lKH{r7z&9sy}VjX%o&d~VOUvG&CXDol&laEuH&Lhk40$nL8q#Jth>cALDf4K)8;rFn)PsBwuhR_L?Q!37#!QW)RngvAQEr-6`Y8wQ>}OLXMu{)g(bbkIO(qIi1z+ zQGqHmm*Fn#t~-3u#AE_$&ecd2)tF+f={<>CyhNU;bhGb;s}e+Ba@fmY$#CvLEG_aJ zKbU?We+24|w8ai!X8g*u-5Yiyb( zfFx8IkZ^tF{=e&r=lN>+Hf$9=wM1bP6XM(rYC{9QV8@uw%!OmaFlwIf8#4A!{%UAB z{9=WNzyOR`Ec6U8A@te16<*u!j4^^?#h#jZ5;V5^rdJ6r6<)55<&WIwz2>ZhrX5nH zs=T+6R-$vOYP&AdPCj^WC`cBmF~e(Sq|g3e=QW^rTLn|;E}8Rv$W5&i12DApW$&qV zuUPY!yDhE!+X;eyji$b8Z`?&ENYhwmXQtc?X}`MEpHGTHG4)DuL+EoD3IC$~E{)A@ z#8KVL&5bSQtiQ7z?q06Tx(2hdhf9a4o3T&a+}5vw;D;NtRgrN3qTa3bm|MqFVRn9G zs%lW(yC{O!yz7u)PgiCbmUq?0J+UM5MvYmTANTa~A;pY2S=v&LGg;&O`%;K_!szDh z(dtuq$S{1nEaTo$7#eE+RY_vi{bIT<<68jN95N-#V{9Cfh;h-N-T$;U633RV7k=hI z#Gyyl1vFolX2XL*9YSw#7fZJnM%ORUqfUup_eh7rBSBS+o!BFleA5xKY$as4FIK?B z3*qIT`b4o>(kse*@G{b_Iw?&4vSNK#?Wo|ov1~RxHDUiwd&oZ2*MuTlNjZ^NSLhUT z-7&hR778zZyQ+7v1ZXv#fvY!4L1$hJkH=?!tR6EW*_(fWOpTC>a_`!p)bLZqPj!|) z1ho5kUn#d>0%PZeThy_1cN+pU2#*wkFeZrmmanh5>%A+DL}z*$CWCU zhdJQ*7uq{gmmGGuneyelyoc3cg(X@`CNW9BsBphYK6;u{@9IoT!EF2hRqoYgs1wYr zZvCBwEvEIdm%WY<^VDX|iaqnWi2k11KAJRZfPHC~Jp|@=gf^xE=iRx1ZlqpJqG4uOjDrMu6bOzMI?GM`Jgf8dLn+r+VSqvOq%&nu< zR{^n?m~o@$FT;4}QkfO$1mEsNiL26LJ|fRBt3LO2U#KUHZ#?ed`1vM@WB7m$ZOu9o z$K6on=o6LD_t|}YO1CI5rg+Uly=GzV!aUvM7Pd&XjXM{!;ss>xv<;c^<3P&DADh_c zSD?SmFmNh~#VbnUomaH~H1`*ntwf~^O%(X)>RkvjxHR%XU%L+(HNK;^h zF|uq;%Th?&^<#he=RYuH^TVZ{qYi6&z0YwZ8DOVFY1~q0#Io<)xu@#CV%u>ME5qFn zv8Hmx@pVd0SR>%BeCU8I1Qr~Z*dewZ2b)w*jV|woM`^DTLy-fA!_Jz2i%y50%gXUf zu^*|OChu;A6d?V=JlC+G68z===9JUuUM%O6$ksV?7zQ@`zuUW07&_^n8@&r#lSMbdFOuka=tv2B|B}GHn`hwE2Z{pDL zy!Tm#dp@R=zN|`VTL#0QUU;f;vgO4gPDo>KwK^-Wi5>C_3#~-kk-AH~JwVI>D?E?I z1Y6#~PR6H&&WrUizia5A-C4B`x_22cDvP_DZd$nQyv-9-QlY6=JK%TAiRa?l+9UByOtqc?O@%C zgB>us>DQ{Q3el|RRs{Ozt&QY@vbz<&#y;W5@OqdbIJXc-mqp(TA!i`vxzR5A+ACP~ zbHS-^6k`4Mm7RH#@aY}SuDKFs!a?ZmQTHD^Abcbga(Ql_^AeoEGSe$2e}=~YyPsI^ zuMc^2h4h$Z-vU+n9xZrzG4@=SP};TA560a$?z?Ib49)IGR*w!W`{#2QTPV)CE>v84 zETNi8hVyWFwwAtnp(5!O#yVs9elE6jPSAJ2|K=r6k>;)MZ zxOmgla(@VP_v*F&%p*f1zq!pr*67Sj zQaC;)w%X#|EvO6AimJT&KluqD$*FmPzTOfT=dZkW?5QUV+l^SRPz}YJ-vi@Y=wUGW zBI(Qq$8n%L*35J`zlXM+@RHd<@MaAzyLZ&fH3TQt2bqp2Ey0pcFqoi?SYtLUc*vnzu|pPsu4Bm1qcM=c4!fzN`?TQ`0~Mqb3~o!6q-?r#+f_UG#{#{ud1 zXmRdbw%^NXE(+N@*m1p#ji$5Nk6HRh*qOMAvp{$Q)>#P$I?NKh<@bHIo~jM7gCXa* zec9G$VaP!0xdr!jEPhvh(oBQUf5_~goB!H|;B)SmQx=^?7C37za2-S{7<~hwT9uMJYUx(XE9-sVoD=b4XYO4Ze3^_jm1As+Z!9R>d1g{ zbtH9JLJpZDmn+NI#~y1f8V|iZw-fqT*&eGhQp3{825z3kO;B8%=cw}gD2|w^bIPaR zz^=hV5$=}-FwgBBe<_b8Hq+O%22E>1&n5DzJ^8HpgGn=sR(5PR#Nr3na!aS3p!4^n z_$NCqpw`ZdecbsFn2gWV{XXN^(fy9szW+2%c=R5Vrrkz5!zeI5_yN=&%-#LSCkEOC z)OEW?MKCGQ?(u8w2w)x)ry7j{b{qFN+19?pfc;ZjdX@iKUxB5|u0KDwk>#HORrHj_ zjsjME0_c3f7qsUQdG&_sX41J|Z~ywMbMb<+>j*#QlaRZBDj$jK+F=Vx`+w^9LDre` zu0j3hVYcHUkF?gqe_l76dRJZhu{Lyh#;$D9sUUG^WQf(buhPjWS$;ACLB45qaRKWN;Jwqsp-eh%xmoR;%&bQ z5APg3hV89E{F!%@;On|-3A5J(KStmF&&NX_h+*TQw+b4c|2{``J#&8s%U^;Do3NZT z={VNk#k=!3#+xRve}m4lEj{0HbX)yh$^oKo?&}dH+hVgqXwN9ku;|XkQTd3foh5AR zo^^fv{DG%M*Kdp|b?L$wwB&Gg& zen9Qyz^+wMf7g+v`KxZO|IM>M{ARhf-pLQpY7^a0Hv2mdpYL6;?a|YJUf<&!uQjDv zfHel*wmUY&K}*U->YYEiB#sqA*FSafV1I+|Mf#WJINhA1pxEIEjN;IUI$9z0r7dJ! z$y-k15EmPdIAIHn?6W4duNOh@Bfi@M67!)q$4}5)ZXFEEyz*+|%0Y@4nbRVw0Ao8Q z{}kjKlK4UkVx`-xNj!CvKeE3~KTfX6=eMUqMhZVIv%djjZ2 zR(F*`&R`4og_MESEdCb~N1Wv!g&3mVcB)Qyt*{Bp4u?^<@cqX`zCkHp|1VP=1ANt& zGGg6OiUZx%I}>=hk#XHRboj&Hb+CZ?Gdmri*JO%!;UZP|t^U*2-H)Bmw<+WNt=dBu zu<6l_wJw>tP=hJ-UcX>$+vKIRP4$uCS*?gt^`lu!TTt@1GXF2lR}{92f(+x@*# zMG6w5@W;}?R47YxRqIpB#=7|{EOa-p>>U_c>HXw|Z^w&%m*N|+PPDNea_0@7qWOY$RYfiloO#2g=o*2Pz1-#(efnsK2ee2JDg z6Pl8BaBB?obGys&xO8IPeR8nrB_%Aqvu-f_`8Mbj(6g}^yadsU*9t^Db%d#{Iv?%5 z!*S$MKws9T8zj!-E{(}{N>DII7uD1Afmu`a6lG6O_}!spd$(l?Nyu(nmXG*1Xcg3W zr~RQ5yqxzQ->!HBhs-{>?t0|%&-GpzpI!b(iKwrs&x-Q+=BWkDJ!^mMvkS$kb0*&& zcn)Eo$gS}WHwa$d?z}h`Zzm?uc6WI$x9uVG)NIz{u}c1*`$6BAjX!qE=0GMvg|gXN8Vmq{OS%{JL;W7Zya#3`d{y7J*V}H ze*4314Cvw7!OSS0Byq%S^x&IKgvKLWi7E$of_uYn-X;4vuxGt!jeiB3e}tI}UAP7( z2;Jg-UCmu7%JszatIN_;F^}!NcIP=Xep^zZtJd$goeHa6I4CK){6eZD&WeW5qtO-q93JGk)fvC9Xs$0R~|U%^GB zwsoBH$#KO}?zioVV@HruTVyCjc>xpkM>a&9y#(}D;cwQHuOPRyZrrF{l+1l$<*GBj z5inZ%b(t4g1Ih!vGLBx5hmsL%{R8HfKrv~&{JT(!%v-8|GmeviEs6A7Hg=}acRIU0 zG3Y2{7wl@H=PSeX?m#v1hitfCGUxVZDyeHkTf8xvAv@ik()1kayAm9P3&JF5s7wI0KoP^Ob zA=8=9dN{~ewP8yYtKJbGRSqWV@)a)$CbwMCH`~YA^Nx6BlTMk1*u6D4NH`d{`0yytK-J=mJ$AH^M?YT zj1X^}65#TBZO_Kni&V;^Q{GN2y$R@BD>OGhon05bL5femAdIksr;bgoBA5FkQ|wIS zgM~WKruFdN*o$r)=S@;-<|A+%EIbtuJHa@dT+GICIMU5#?JsPS)B+aJ>zL znW-J0p=Bd>Ij7z%H$P0`s-bLCUy+Wv`LCM}%Z0#ny6g3Oza=rzWPgq1c4Az<9@3Qh z;_67^$p3XLNsvt!ldT;Csv57QFW!OKte}cdBvxJu#vA(`Zi)Ysrw1ry$*!)M6kxtd zFqn*GI5v&iuSv`2-4#JC0sH z-GUvdhnFrkaf6od!qeAsV}P!5Dtl5d6APVQe2+GM2SW*4^mD)l%chRRpTAy+*`ZF3 z;yXQn!ZZ4;TTKV5!=^ZvlbtZve3jmIn-AFOlDRghD+H3d&Y_vS1oosIou4)G0!F6N z6^g%gLuKC&hXVypNVzwCl8bs1Tc?IqtjQ07-s!7oe?|i`pB~9+t9+=%gzj)K80tcTR*l{s!GsK_!(p}}f1P8=7&F25P1&zxFZ(So7!|3C{ zO;SGfI8~C_`mW>zbl!YKG1_Sg#qZ5GFejO zTib#oXLk9W@Y(=$sXZCFXCkrX8JTZg*d(@od1#oUPUH<}2d^3UEo1(*_w8+$?6Sh~ zr6whzZ+`$y+KngjD3OP1`>=;Z(yW4>uWk?Sg>>N?E47g0n~j07Az(owUpTbKtlGC! z!V@Y#net~yRl{(gSYL6eJPi3EkH(6PBreNCE<(aDkhbv~%D*DweI3e;*HwR>fGUzg zOifTD4h?d?Ts1I(t#6J+y|t-=Ha`FLSphPTPUl>=PIfuO{JxWycd-K*r}bL4T6*F@ zqWGnWq9p9FtnI!ZYmO9tFU2$0uOjW4^z(-&g>abaak%Zq3Z#mNe!t{422D))kg(mF zFnVNpfBH@lDEPC>%F3B#H$Yq4JrVU|JxG72cuD8b4ALpxJJvm8>3;(Arc-=)sw_~y zgeul=`UV35*O$EC&X4))H0si{2wYP1c)sU4>L&cQsOLk*+AeJH+W&p)&1E>gR{j| zgVFo(BbLS&T>p4&J=^z>?ldel5Kw@Avjl11``XxkvO49P!Y-h;7AAY^?FE|sO`X63 zwm2KxdTvwDugmzej_vt1$3L8sPuh;Of?UgY*L-1SCg54n2MXJDeb@YqOE=iV)H&Kl zl_jDus%1mxOIwB0=c>yoPXuxJq`|ZJH7;1!x54V!&J&2QGY;rA3t*?N$%6IP&g5IE~^T(fk0)MfnmVBY@B(4pL-TDVuaTFZsnO&mrg#(#sg?Do# zIC1ny$*6|IMP!T|3@x6iM+D-3MC_Bp3QO`;gIM@nDD1qx?P?ZLS2HsA`S3@>3)nT$sdVlFi`Nc{ z8%AQst&*WHr#Z)6lc*0HOpO+Z>?^^7C%jkH+L%b~EGk@_t%EZ)3QHNiuYe)yn7#2G zi`M}knjY&hIBbmN1GGxFW2`zD5=WH6)WzJD*r*_VLt=^*kHyc=ukbD?euDM;7QcE? zIf`j>Gj)R7*m&BJY8AXqE8HJjXs)i?uM&D^6R}6OOnaGNeye`;roi27_jMdzVSD>2 z^MB1}oLwUod?$Ab87YVBeVyOnNcIQOA5~8=_16Rb80(!_`(x`e3NOL)$k4fvJX9)* zW9!F?(?T_o&NpwUXior{=c9-Bmrt?SWb`fMY*{dw=TrD4x%@<=KhJo4oLmHT>T>Zr z#aw_9;U%!+I~6HSC-+s#O_K%01s}MLej@f66E_*bjC`ah1g6JIXJPMQ3-yG{#IT{zTYc)S^56b; zA81>~!vDeGia%!?m;A!M1%LEkmH1)pqqWP^w>0ADaI@B-wQe{ua)!$Brx$+g>xdNG z%YkM0x`kY?Tt%ul?P}3(LSJLTCv21Y{h+_+#IfCa&)DkGF!N#Ly9{qlq=Z=*_}tu# zL!x}9^1Z@1Vb-=c{TNXn$PB*tx^dq)_7?Jei414sPawvPB5~}@$CWU9;=QB8pH%D# zx+dHpP2dB+otu7r!2i&M)1Hi>eI z@6x$##Iiy%IJqx?`cSb|nn0DIV6M6S$m)YI~2ZYi4o$2+z^-+v?bO zQ+>Z-4N+&3sZ;SxV2F)}FJiI*h10sb~nTe6*+(r>c4qO^zl( zh46Ro?zPL1_OfmBLkx2lOpT~dV?;FTo%lA4bZ64f$w@w}J{a@bE;1YG$GYyH zlV;~Dp5R`}@LgvE)Zf=6oZk}sfRx#@^cIP$n8>p>N>3(1$T*f6P9R@g&L4W? z@A8c+r;)jr;X1K#0$K$p&AEcDu_{~r&f=+;*c;lL>c6QQlSY-rzQ)`9wNo8Fk~{*S z)H^=dS*eiBlQpl%yn^5-M{9^Hnpv*q%MMeF_}PpWK`Hv{Z>){&O8~r>hbQ!v~8PL5#;@I9ZomvB)P61D?eAg*Lv*s2Wz7o`3-7*jf@xT786uD2}8Pd1k4q7eG;$gz}p%d%G~E0l0{WY*hV_%O~zYWao;hy3-=Vmmrh^I_!LsUOV! znb>n{Pt2Y`I!+iFIK7w{LB{sinGcrK{j;B`i)}JHQwaUGPsg|@Om(o`0%R#w=Kf{rR7YymL(zq=O#u}wUNIct z{>D{)AqSY|#Y*M7_+b3JGtd022RL$!ugrGX2KzphnS9#9hL^xnn(KS%K|)79uHo6V z&vZQ7cvXj`9UJOq#Z^f>;!j_j$zH%IkCKi`(HtC0wY+fd7NNg8nkh0ut+)e2%ByeK zRYgFwq{rCr&;cAfmPiYZA?&%DxN{dnT32HAVok%{%lU9jrKbL}3xPwZomghJOrsWS zucaXwA5v@9!0!qENBk#Qelrewo~Q9oUV<{x=&|oHMEr471Nvw<4+U;=&m9dt1kE<8vkg4hZN_#PzZr4BOVZYJcM80k5P!hND{g&;ldJLqQURonU zYhkr#R+cMmdfXEN?@BJ4C)$m3mrfd)Z|uj8nz5_=hU=j#etXT8@m8Ev$yt@HJc6BP z#<#a{i{lU@{-m%Tf%74Myb_=Fc@6eWD_fN;R6&N!x~5J$S)3?0JG{!2EuVx{rW!o0 zsgg+D;~vrT#tgbVkN_hC$W{p->G^Ee`=e)qkw5VjrQakfgkjb&A9zb#QGcx-z+v*hP3CHzGi z)sIX60LvaEa~+coq`K;3_9K(!U!=NmEYs!Q7xFozf3cRWkNU&C5pkMoBKCQ;~p4y1lP7zY+g#B-%etL&u7!I6XKcU#H zfQ;F5%+jQP>Y@>LcCF2m(%Fx_r`ITbKFf+TVe3H=A5^(Z=HN+GNEbH8u5&*>-gDf5Q&getZ{AR$ z&-kL^rOCN2jT{5YzL=9|I>d1N^20-t5+OLQ?9p@U1wYabJ)9DHyZ}3|EgSYA zbf=LvX&V>vtR7qMw49DO$7YYPuA5!^J1IATRVM=dzQy}!Lj`QtQS$VJUrMs#cVylK zN}{s1C7H|L^t#Jt2Z*$vj(pTsjBT&mYox!O!_h@W!l$_QBXwCw?YCtMa76rU{*H~W zpzY;0p2~BISR3?(bGfMPU;nS}+4e1aF=;`I+&beMSRHiJA#IHo&^DR-Igb+cXO$gN z{JLKVJDL1=ula-LfBlEUd$JFf1!Kp>ydN*-d10}}+7<0bj*&SPtIZen5bLdfb;g<( zCr)77{G)TnLvG;sOQDZ@GuilGkh$=?^y#gu{y9Eh#ppXbUK@v$^+=z2gULL}v#J572!8ni+mtQG2Y=#}hJ})> z8epx*b4pjA47QG~(-`djh~q=c+uvP&hD=WLwUnzvIIzngk;`@|G7nVn)K{lqv7%s| z>@RL$?oXl#*|Pj>q;eP^U#j$)%z1@Fgxo^lI(n)P4mma6Li(e3nqRqlvAuVtK-AP? zXx?ma;7!&_9A#|odcj4U|MVAUl#dTtB6V_PPXB5mc0FFMa`E9#=$lzz``Q(tJMV^%kuNIl!h@L|hl>6#HnY3~|16KVhDfn8Yd3 zdGd{dn{c9p>!m|5H%`3mia7RA3EGB=C8mc)a8Uc6Y4$iv9}8-RufACsF^=6*ytO|5 zPcY}x#c%r}2z#OTW~9oqGCxSCFAL$`whh}`%(w0cW@{ z_lCG%v&bO!BNIWMws=3b>$_yLU)7b*;>2}&UiD65Kg)h|z}G8Q3bHv(O#%Y`KyN8y z7gyjUPHXi^;4p!YrHh2PoD^6|=5T*3d`LP0awKLhpIN96UrIQHGX)xv@y<%}&2@rr zHi8jlRTLvm<`OQUyxJ#>v_@xCwD?WT%RTmog6+kz>nYcbRZEtFqfcDVUa5cTKR`l? zV*cL#Hnw(oJ-QZtIL8qfKWWc|iiB{&UEBC<1Y12Uj&Ah-@=*EKU;8UT=h(cz=UQ~y ztDNq?Mw-NBr%)Y2FP`hwyUjPk3H=3b8ymj!oFp>;>yX*KY2RV&a(;fRiZYHSYfL$m zvc+w1L{e_~tziSAod>FNURn`&m)e{|OA=jJ>z%|I+HL;g;RP7ult^hcXXmRQvoo+A zbnwGs+lI8ntXU+NXpO5~C-f%sTRWoO(ao^`O_jX#?m=RIdTBaFjvj{c{ZCeUUyXv< z%eRluY&?T8IcsNY-<-g%=C`hWOc$&ys-!0UxP%k;UMw|z9E&Z5`ns*hegB@DG4R4e z2^-{Z@<}gF1=_sZ+dUR6fqa=6o{Ypv==?T!hVwBC7fTl6F!GVhVEL~wnw73m<6w=Q z5v^@=vaE9(I#=JUE4;8A+evTgzdUooGL85*_J>*iDpKi81>bM_SS_>pedEboGOu>( zHj{B8uhnToO^eN8)4L_?pzvyekIzIf@2o^|!#O!J??msqbvp^3=b^?ruC1N{Z1XGQ zshC#kFE*V%)_vJRD?@VK2)<;m#UmaW)8RPlkhpBzVmAbfedR1|(8unbcyHljBCgu{ z=iSN53SDG2ud~loTg3L>s>mJY`=tqf*x`K@dOh(|M1LilehDsM)nP%}<;tBUt^?>D zdVSv1SHgcxPbjqz4PeF9poe=$t60?oGS|S{qr*=}U@&2Ut0nCOc1rSdeYL*=ZIU50n9qqkNdb*yRIq_i@`zP-S0T}Z?iMytgPB3G(F7e{;DlDVxN4K+a<66 z;BP18n^QCPB9&|PcCdih&54($KQQKB`q2zz=KEhv7(u9H}=&K>*vTRAk40xE};Hb8VN zP5I0BK4iX%7yTsr3&(#cuS5kRPf;)HmdI)QPaYp=m2OT$`sXIM zhHUA>Z2VP7xoo(hCFI}nk1^@^vp7Y>CFpk_PW9i9!?DMozhqt|c%;*uEo)m3v-8Sy z*mZ{8opVLA_XbO1ACj=}R@LWUi6PiPaeYYPBk&FV1)GQGZF$ai|5S!aQ_;`=h-<1W zQWYry98119z3vXdk398s)2wa`8H?7QP$HRhU{>N`Ki+j&NOyR20R*F<`NGdJdIM2U zS^3@3e0;}BY*)|@4?D3K{Xah2(tot|KkUKr#m*au^SVu)5m~bSEDYvdjka9L(x-=> z?OG|*{;5zmPf#g-5d$YQ8~cnz@*(WRuQ!Q>wb-hmVa&CKg1K@H4UP+>k?sKj1M`(@Up~cn{<(Ywhrj#p_=}X9@~>DQn<2ZpO%9m~PvuP0m9b~S`CHDG>oCV$ zJ*a7>h(p>nzYdn%gI0|+rw_LcvF~S%S<}K}Sk$1Wd9{Fe@AU3vm9=#Q-*@sZubTB~ zA;{3%AUP{{2s);IBx@}u>IP^LWvO2IY`S@D<5IV_d!>%S5cX@>6;9S(|DV^>Rl?4n zf5S`6hXu~~bg>89b)FaRg?fjYs!vrf5P8YVBf#@ z)&7x1TTAoR*`C|;FfY_NQ3>LeS8ik&wnF8{6Sn1J4*z^k)ordrWfIC*dvdAGF*(*g zhgIs<8Ro`BUX*SUdhh3c!VaS!-V(CC`wLKoH}Bl)w1gzkdA8EVhw#hn*EXHFo4Xx1 zeQt~P<>A0~b5FG*vk}Zqx@nk|!mD?x3ytdC+y$QMEsU$YU}Ea)z`9w6J%x zZY{-qD~{Y)_*L%aXY6*>60mn=;V7VI$?^^7&J%gmxtDfovPaziHQpqiO^3t{!+bHz zbG>4Q)>kxHb~A$SCXu%sFyB0XB7;p&;-B+U1$l@M-nS!wy89)9CM?)dxFE z*~YW;?lyk49wX?O-mS7G_$^R2YS|5v+42kj+)qkXHkkV5UxgRi3$|IV-cI7|)jzAK zMBoUA)8a(OY#GoS+u_>%n9v~_4)a_Tc!!1a!kQ_=U6SY7^l*?WEcJ0~u_{ClpBgz> ze-nDw{<-*(yyTzXGZU{LKgcBvlhK1_Plnf%IE;cnNcquO&!er#+w%T2>%7PB?<0OD z(uw?&z+S669;a47`sU5o9=I>Zrn;x+OeLyu#KYZ2GrAmxR-}J8V)g^ZqfABcB%!O; z)x2=XMrS{ietE23o-Kf525vWU?+|sE(T801OgFO5bLd+ltTew&24;jqD2EoivbCF* z6}u}=F#$_1rR7~&BMJpsVY%A&uSf!)s;!&e=wOa&(8qu$tb7+xn73sccN6QnGo{%j z`=&p1iSeuM9VYkzx|*iEEH?fJ|DUg0)PibO{0rJry>CoVS?3;!^Vopo(o1Lm+RH=7 zkKbgS2bl76SB>!{J7{pVIW;BXkG(F74XC-ABrdOUPB-N;OmWany7y@mC@=5V*YG=l z`*Qgw-H8b>Kz-r<-i(tZz?@>*9}R(a#_NSBSo)P3^>5l4(-t9i)0wFqxF1DCE_HLu6vlJTh@{|uUkI1P@IIR?r9Yn z`wp12_f_u5z6+zqOOmV&gP~{riFcP*v;0+{ulSn1GJXS8SV)+P{k{s+{SJz2Zcu?K z+|@@-^uq3s30JP^T!pD#k<i43ij&y>pSq); zw<(6uL!e3&>1N9>ohse)(fN3pP%EiIX2jA|v(?DKB1Ak=aXl+*V#k^w|6m22x@|{>%wwP5)!%?T>`o83xSGV<${TFa zU5#V(Kii8J^b+%&!|V13&w8M4sn}%$H!x3RLEk(%uD|~C5?jkWN1(($GA`i>$JFm3 zL%~~!b={a->ZICp0m%s~l#05DeS~^_JS8KDsxS8^+b+(WaZ=&zzx8SCl!P`S0s-GWC~w{k;}o z4ttew_J@OG|KRLG@9l)0<;LY37`c(eS(SXg*7^a=2}DXK-(u6_C2=jDD6ApVU^3^W zU!cw#m|Hzd&g>)Nm6M$P6%kt6IP5roSNp{mgq_;5Uzz@PKZ$F*Wh^=1f7;b--`}v8 z%SLCM;MJt zIT$~`!Q`T`FVH*oIA5h6g{}f^ zaKTOUSTQJxlbw~BvL|uwSDQ1HXo4m`Gp>7cZ%LeUDx*zZy+B`nHvE9wNo*2YI;*T6 zi0vr~rOo`3FwyW>bGP?57%OkiIrJwQ$CdATS*~9UJ@X9(Y^2QD=GWja+IA;_uh=1V z?1x4&p>NuD_sH_wo78YDDfs+QmN`(CCyXCmbPC)I#HC8u-{dzPmze3FvBkYbKz8OWcAIMCUU+^L8S429L|g)(;q1-n z!WRzEZCugLL!H5%$#DB%Q$Z{{n7MJqd2gUeEQ^>(I7H&I`o8#a%mfUbt=OPba2dz< z`48MJEdqw%N;mnDF37e%w}A6mEKIh7fN3JDz82cG1y3bfH2iI6n|82*JBh>MVnJH; zKXK+*rIhl#!BY)8Tz_uyUw54C_wjY69$ed5Ja;&hGVm@Cl48C|~|euOC7;EU2A6VF`zIUwDM_ua-Id1x_zr_oYr2m{CXET7Dmgl?si)@ck+Ecg5- zCy~Qim)Lt~F?o|}9!x$E=($GVM~Qt%u``}7uLaemE-uUnM<{<^ZFkor4w&ax{50g7 zPpt3DEwtD@%VAPH(KY;P;NNq?lQOo9g;#`Ho#o3n&GtZ(=I&K@BJN^&jg`x%*K48w zkF4YopEtmKl$!TQw;h=G4HACb+5?O$ogu+F2dXD14`f_7 zLa)+1(>T7*kh#rhtbmyXGkk_}4fi$K=4+kzj;PX`Y`7BE?^Sg|(hnSpNj&A}FXwmj zlDHa{ty0i;BQ2EMCb4S6hrjDFX1if81t!l@znyuTN#fNlI!lg7MlZ?t0%q61ViA_gW_< z5|_bR7iB5dc|hV?CM#%Z{t&W09#i`9>k>44uUx|P{C_yys28{Tlp%X2Bze927pNav zpyk)chO2`N)uTpFMyW6~?{IMR>NaSd-}U*V`&OXn`2>m2lfwE-dhPKESE2UA!{KV( zE!Y{ayerR*6PY*n&U@X;4VB%VZj3$YP{lkc`=r|#JLfSL1vC;mN!40!!d8b50QH`? zyp^sGG!F9QnsO88*$A^r-%%j}${1d!)@G@(ecu;`s=}1yk%-{0CZ^y)WXp{}G-W0YOh6{FIm)J_o(~L>+^@nvz&}rjC<#QFnz&L;5 z>j_^`XuzA5dUw}D#U}Ukn~7H;@)7kwPT2>@+oIEaR;L+;hHhvk{S<_jgOz2uhihRf z=xXoanjWM#>PBvSz8x5L`n*q-=Od#iQ1A4f&A=$R_G7$IM0ikybBqdFBn$k{eUXYo+V2r5^p{gViN zTC%F|+GpM>NV6F@)Lu7D;{1Ihv&ygvnq=j2v(x)vbf*5#EwU4~1<8v=3jVcAL}%~X zbVdPVbDEV6p9EC&@`|3QItZgzT<>kub0qQnT;k)q_ti$Mx=iT1T~M)$9EEYpTb^5a3jkejp`A}1 zQI|F9DXhB5i{&3e$HJaErT|+WfjB1yuJr|8>Bh#=AnZD<4IM&S3a`i6>cvRB5#+2N zlvS9$GxFqQaTScjw7<kNO$?#m`~%=;-!UY>)KH^;MG#|Hk{UixY20uG)}Fdo9slWMRMX17Mo zzw4HYgZ$HC37xB1?`7;>H@`=}9fmkxS>33Xz>LRUFuTE7TNgQe%~nMe4Wbrul^g#$I9`q<)68DGMI z+L7#I_|pg*obQm_p3tHDyvwo@>lUcEZ1()a`!i6e;d6EE&_3us)sVeS{4hj*9@t{M zwHjzkgOGGH$5EG4 zH?fNI!dm0i4nVzl>-=;14cNdFbYRO_DpD54#aosygejHhrgx`qK|bHhz3H9;bTNC>FOBIysLJv_%1UFP zar~Bf8}0@<_ovs38F0hcRnG254J^9|s0;NCahEaDHHJqlf6Nvbj#bx>z*c}^g<5c*z{vQFy7L;rW}C$cw-G3I^H%hMe@S!M|?nA5-XU8$BmwjF$BaPs^vY&N8; zE1%s6sUArejm{fjR*`3v--Tn4ynxR_ zTq%8gx)GU;*&a2Mg3ur_8M>tAZ$Eii@=RHW!(?dJl^>T)vBlcK&SJqv9Bf#+QIDJs zv_Bpms`hlY^*7K|yUi|D2Kv)?B)l>vlei5@ySKOmk~n*H7-^Z^FncKfo1!Wk{uDa0 zc1T_Jvu59y_I(?ktlJIE!C@R>OD>VP?>HRY`BWPi`zC5@vU#8^&?HDoe;JfZ=>1la zxeJ|Mla3dnequ#Vy-~@QyCi{2W7(1~-a}_n`1hB;^I;@SyZfQ45HQ!Mp4%(55&Gur zW+Ij?!F>K#`9_9XzzE+e&Mhs`fPi*pV^ zZC-6ycf}Cc9eB7ZP)Q3W!eb^L@oPg22+lW4F$G4#v#ZZWF5&o~e)b>`^>gj3{VM7o z&%@HDPj}uwX4OwY_QsEUPQ+NE-`320hGi%+`Hwz(bMUVpaxFS=q+b= z)*-BG_U-k(|9{p2f^nbAV!JL9-#Z(BazR0MI@@)_f0VxVC}d#^UMLvqT?;i2C`xZQ z-$7!FSB?KF4rmL%6(Rj-KNP3_d~nX~1#};CYI&<;1QheVr=A`VC2>43AL1W*jumU3 zuZitc!}6HLA2ip6K}z**TZbP+UZ?BS(+w5QdYI*yvD@C+0AuHyTiz$|s6f}BXRK`o zz=YC{6K;o*HW7TQJJ^!wZ~OOeTDwkQ^2f#dg6Kp(gE8FTsm)}IH?w|c%A47$29(3J zqUKovSAFQzSie;-?8N`ZaL8Tm$At@QeiY0~z2o=3Hp=$@waN;SQ)&lrK#9lM zWYG`k3==ypX2+J#!tP4t$EQBA>psREFJO4DJOH&f3N0@+pN6`c(q|q&+2S)$vHbCJ zv&#-pOBOLax^W>)M0oFs{N#oV$2i@tBSd{-^{2Jbr9aL?CxvSLR_G$kGJjuuD>ndB zYo3{odJ_5vvmEi4oaon({PVY6sNEG3*Cd}&BBd~fB#xKB{qJ7{b%AcIpaX)(*P6=B3f z?2=6`O9vCCgO8tmxreArZTq4BDZkB%%#pW!)3doXB(Aaso{x#Y+4`r%FSYR5ZdQN7 z0IXRe{gV|x!@8nOW}??s9G^AaW90l8nm*5st<^b#tqCII)CcpR{*d%O8|kk=U31v@ ziCz}APNrdzMHWn@U+b5XWaC4}_6J}nRQ&+D^{9u`O1#*{VMzRQ|JKtizD8)~4BkV! zMd01W&h_es*rh|yfRxM|aTP2-q*%ol!@?5*<-B3zOLu|~r)+Fc?vLPhp!nwNUNuO8 z7MW|$NxwH@p@j#}Yh=Q7zC1NOnBA{$d#=6dO&+Tbg2b`^ZQ!ko1ixGBmTKcY+SMfP zW2-;)1N3tn1OUu;z@#U;F;!E|(7^LlP$}$mVT6N9Gw# zDUbPm0ZR8+EaV@`$Mny6zqr+!VN~S$h0yi0FyVUd?76#@F!JihGFtjPn6#aC-tlb~ zC%)23vSe-G+wrLQv0&Ez17m%`9sMaR{}uW+Y^mYY`GFZU#m(bpRP0I+uYq1gJn!gl-fK!3sMYa;j)$4gVb3(jwW5x$+H17`sS znFns%y7wAsDh7fhS*&jX6x@l+OFSV_9#h>(vc_Hr4IebhPTPnV;+~05?dQGD2 zZZTitAeAI{x#KZJH1)VIT4jVI6C!%i1<6p>(kA$x5%K@>bwlqO^Po;<$pLr1OF;8J zyijk~DlB(jyYlS*L>$YN>bi8aAG`KE=31A*gVaY80|ET`kZH6?C{|<-iEGERVXH%x z(0RycX{yyC7;<_$EphS^PCS!QS-s;BWE~aWk=a`Y{e4O~2WBmxGDptusPP75mWtTS z))$k6c5Hr4ZoC8|?T(uhBPpzTC3slZ;^+cV5*N+DDMtMsiD$r2Gv6HzZnJgTA&c2fp@*!T=@K{_}Jm+4iU>8a*6$pcoJUjcq)$BO-HqKFlLlxJd$kTq``iyPZ zaDqRTt8O`zOH-BH?7v`h%a6i&t61?FjO95-)o@sirIA5uTor6ORmAx%n6g%WrijEJ z?6TWQz6ZwZ2MyMk+d+HnzExZ{yI~fk-kTm%B=KGD*Bjp!2xIS_smXpDfcCPJdE4%0 zlenzkNR7Nb4C9=a^E5aA_2YLubfukrMdFXsey+HR0W$-Ocz)Zl>NiN7Pp;97^!}qx z02{!qtqsbL#@U^#?#H5?$9edV^1!TNy@%(kd=mGq@G}1=StO1{HY-X~-;sDCr)aH; zx-jRg!+SBln%F-qj=vHMdi2kB&OLnb&A)OQhVFK4k+|axnT3UoSqF(ap$59B?8-Q0 zXmz#p+^S3X9TlBw9}+8;!Bn+Oul9U`|8A-?<@KOk4upJuJ>mF!8m6sfh9kR(`pAkU zt52)$6^9X(d8Hk{t^WC4(}g_y7mcyk*OV8WOStqO?<>r+b$J^f&HB4ZfwFKu{{aFL5y|um5(b#JoXNguC}B5?MB5{J*I24&-zN#8DcAA6{s-MlfB z)&XwQY2i#uDI9qvV?FSl-EM1lO72wH`4Box{Uy$r%VLMZKvH7$LE=0%85mUMXRGH2 z%CC6svz1#|&-tlsw0Gx|RIIT5Ca3$Is532m>f5@lDuc|iz%j<9>aQKF`Qeb;mw0Hm zcZ{pnJ&M!~Q%xPN2e5DL-q&-=HCS?GrT61wzDWC=$9q(DHPVXYILcDiVSTWPeOXKB ze}4ywbL!sGU=J;9I?7;9FWmXpFWcyR^Nl-+C+_kcrM@U=6B4R7*qQ=Eq$R_?>lWgl zm05QSTNYw&NxG0ibT_0*y(e3i{k5;p-P^f2M;fXQat~(+|J}E_r=KN{Y=y2h9;#23 z0-?uwW63^yq7H$^xncF5jy?Y`uHQ9#;yKbNi^Hc0k1min%iE{MG-82C zvG)@B-UEF$q4(A%y+=BSdjRiUDJWe_Ik(0|2!?`3GQQ0b`ItXn2jeB{q;UF_2IWQV z6~H`uTlHQ4)YArXz0Q+`0_&i6ofL8W(((AD?P?DybU>@=oG-^##Z1M~SKdZ!#a5sp-cn|j46#!w*3r?GSMRxImT_n~EeH8O*m)Z-o`V&uGxm$F`x zKz;YREw07|%Xtk>iQ6c`kh^%q;68#Uzc%FQgieeH3=T~G6uQliu6bRY69U_@TH~vx z&ND+SZQE2hr(g)JGtY*nE$%_FIduCohae_9-9ly=Cr+m+k;lDzu_OGJq06)X7ulLHLFHQf^4&umo4fP$R3;TDJ+I6dYlV?6cd4oo9z&dS zUY-Dd8Vqk^T&L?0`+Y|0X*2T^cd_ca-OwjrA|Es)w&oUX?M0YexM0Fcm}O^RJlCJb zW~hWhiVzLPTXktvB6JSROn8h1L`1Q{y{M_jB^tZ9?S?K}6+rIjzJ~r)?%4Ov zw$bJR4My*cC?y?z0`&DdDxc1=@bOSn`oSbySOl8x^qf*kjDvb_+0U=;gh9b9=6CPz z#RGE*XcU_P&d{WdM{P!!vxBi;?+mP()u*7UpjF z3XZQljG6W)GZG?i<4AI!r^Qv)K859GTYTJy^P#@a!$5_0AB&wFdZe;>q58nqV@F%u z*yd|Z{Gnxit+%m{eDk`b;eM#Grlfajv+Be!j`NhlCcVG=?0x;$*NzxMgMgx>rk@8i z-lg{bOya}w!oAP>7ypLYMe}EdYT_U@T;+lJTY?v#8pBOl^Npw*iMvHvuvbeSrY_D5 zYkyOO&e%I;a`Olt@`|kJ{9i`FP{;N3cByn64pZ$-@Ba42){6K;8V9_|Jofz?t2ksa zwtBSY*3X03nzglQUe!~qIFUB}bzU5qlkuqb!im!`@OGu-(9BjC+OuOI_sB^cA@8tn zeBK77BdcpS96Ew?8^~1}!Y8qHjlE>*zI#|LwL!*D%n7Ear}h>cB6tCUa+9QvN%TN} z*B8o!MgY*dw;4RqT8G1{49qsF(V_QA{hIIAFMyG^=tj8O5lDF0@_6yfY#3?knVX{# zaf()1k&_D&S@n)2{{3FZF4UgDbeiXJMcYGAyLtEVLjOaMrrh%+cE>ZMMaN1X%SwdK z+ZP-ql%lXpz)xP|ohvpUTjW%o1da|8yZhwc7L}#^tN3WTIwu>)h+JsFCQI({K6H-x5o^_@RkyDN^&jKrO3wI zv3iJQ(z6{Q7dU`+{>m&hd=WNZ?0WqvK@*>^+~nUr#etIz=l1ZWv-3*R9!#XiND#Vv zLvPMCA8PoIxJS#CkW~dkSi*>9qC6WenZz+WoU*!78ySn}7l*?KF!I^>>$8hwfokhu z;%aCK)EYscr0o{&es>b9DR(UYtQZGoV0F)6vnEUr`F!h~p2ndji4$Y32;J|52Af2x zh;cJe&%V;yj}#vrpC#{RVQlQx=g{U|RmWR})W(T-k#vFEv|K|qHd!!8{@_jANK{`JT?BXJ`b zIP*eJconOj5C8NSSsW(vbI{$p-DFXv^ndpkTmRLol&m_pp5RyM=J3i|CmoL&HmPE7 zNSUnn^19wQv^z>3TjrI^9S-?x&lwsX2HXFtL3F*K+wHoBXA6X>gd<@^5H z2@_qf_Pc%B$u{r0tOs&7OKc@^KiGb?Fiz%wwTtckQM=8#P(I1}yJL29k<0SKKrLUk za(`Djj`nS8UACL3EAGl^5AzvVg4E@-oa;(=!O83s_nEp#Oz~frbmXNJ)VweHT3^8h zZI8r@qqH=UdA*_N-AWl~Jf`DcrxS`)`N#Ss5d}o)z>99_to$-EFN-7}-0zFbm~e;U zwbEF7Iil)pbpYn%Jv!|!%ko#TT6}@=_brm>eW}s4x5gO$T-d$kiYWue)^rLi&~%5k z6KRvR_K8U0b1m`7wa1QXk1NZ%|p&JcAmA`uK zW;F`FR7zRNSVq_fjZ2+}Rdtz2)wi>F@aP!Qx=pXXuvCL4+7t6=J5jLl8?xMRm;~c< zcY2h$xM7&?y4L;%7nD)G8eW`?!aDDIp#}K_-g0>TOYfHFGdQZ_ZrtA1474A5Be> zV(9@&#oZ+zNL=}D{VV(i;j{CIQ$G2hyxdUZ7R-HGq!gH=uhV7onqiRCp*(x>Km4eH z-sa0p0w0i{FLGo>1q*ja;xhLn9h~yT0%JD?NiKlV?4*j8`2BQzk{qEw_0XH=d1%>Icc5)(wi+6}51j#RqYJ`W@j#&Wh3|gyAN3mTuLh?ajBkS0 zi@PZwvZNsWK!4*MMmjXS9M@E+__u#D9-J!K5}ykF(P7$TNna?VU6Xf=KZ9lKoNxRv zoDbCVj=wndJ|kL07=QWfgVb2h{GG}~-DzUXhYH<-`%q)lQMyfOB}|D@&ULJf#-63# zAO1hg{dqW*(f2=$n}<*mQIgW6fhI|%)=8OC8H$9YNi--_h$K{KloBN>W2VexTIPA4 zXU9Bq9P|9^`^R(c&-Ff6*Zcc@p67S{et)^#=f3y7@4eSvd#|-#>$U0(X?7NbJ-=nO zDE)WezH<6Y^S)@96SeHWF7XK4W}H*UY>i<=Lfz8o)=`|iuI5@K|>`Z&fl zw)beNDm3brS=^JEhH|FHcN? zv_J;R!lo7*p$4cvAlo|io&&oZeX}BL2|Vu9l7H(P=AE#_SNz0nnl8>jyRUp%TwBdW z!XG;I)mn%=FAU@_wq&m##w12AdmoK&7on=mrm`y_0v1Fbem&R027~He`G1beK)X?& zz*}Bz=-R%f)kIzh=C8f#5%)UwU-eDf<1@Kq^TdqT-!ebJHOjm=zwU__dFA}5Xsox=qgnc_FI6iXXIXw?RxAv_C zuHW7;v`i=B44)p64HOulioT*i?AQ{hzx_vD+{-3m>GHg|~ix8Z>c*OWI#EG;fO&!h#NTW@6RYdj+{zRg?`pLKzz zy{T3qu0k+jo4l4a+!oqK7*_el-@y)Xe4H>uud{6j!A4YLKdq-J+XIYUG@6RH`uo^KeRCZA|yV%EOPWQEw2r`R=P2*POXHcmU`1a z4YNq$Ja~q$devXKHxzAnDTv*>nIxk&t_pdB;?1( ztKl&H^Sq+A&PkXzh)_?6q1CsB4i@TD$__K2)TpgAjoAV7{(Ylcnv7s}q509Rx&`P| z7+V>z>^Ig|Je?Zfp96Ca?SH<#aw0J=@Qhoqm&4e%%O;O?R^sT{?K4Yr-7vb{@V=Mh zDHuL`xxmA76{df%>U7b~fWahQ2B(+LVEW^2n|(hcU|?gT#OY&&F!E>Jpxd(mm@yW< z-W_!ohI+McWEKQLhgCMetig5?6U%Ja4u3<)G5%BZdqXXZjcuwcv~a}QkJ}c{FKfrq z-M_`Xo#bFf-?n7$dpRIaIHa;=ZGkBdU+av5Wh73PN`DEfekk(ZDZNU52q-r~+8tY0 z!#HPHNm;=M{Iy-{>9;OR=#D;NZLm88a@Po!f(a)k={5aidRztLTmDpZW?iCN=P|Y| zW_l0l@)ltHJnSmYqT$M6THvSclk-HpW|%B+aEvU4$)R>{17FbP_X5>4JfcgC3C1s< z-@1F?0g2HpaV%T25}O~$PPAm%k{I`{4&&tgCw~bf@2xtNcTgOrr_V2(WZ3~(i{@@c zCBxXz5ht9}>jd+hTg-QfvOq?9W{}}_!f#4d6U$tgzz?1BOczsyX?hWdBwS2IH*)~h z-Nki9Ga0%~EM0eAa)jE_nA=emw@A#~7S^V^PcY-{q}@TL4iW>~?caOtl}HR5rF|}4 zdRsg;H-f;2gQ9rQ!EgEcI#J7jU^An*B;Xu1ev9@;n|9kG^ zD0KBmb0#XS!`}0IZ|w;%#L?1S3T%JRucP0bcCj8SfzEK14 zW8ZmrY#j`~IQTiBoXFo zCZ2|>3C#rWV#2@7F7&+f9xEGErfQ0X$J65_X5Q^eJ8wobC1sHeob(*K4i)NG4g-+?rH2^z*6c{qmkFuKQm|H*4(SS_Y)5F5jX zZ4ZMIF6*y@qWJmUxVjUD;{xOlEldK%)_R+wWCL`Mum8QgWhKOMtk5wxC+b}<_Ww4V z_2q}f4fd~gX*XfnN;3|XOi38L(kNeFv>)^F=b_4+^HBTVBR;kM?>Of5@+`~TNn(F4 zeDc@C0qAz0nG_5g!1>eB1)`B`nD>5t8k46A4x6Y61Q20ulY;cpP)xd~Mq1A_)_IfOn*sD;)xqcnzjjvEx}Mi{(zE$R(ZI#wU_ zN@`S7gN_reN$;3)A#b?)+F9#0SS=Yjo5hlkOYfQ->%%%R#`;=_xV0<{{9$I6=BR}B z5%b{riDsZGEST_JBBim^k|L?M6QJet1v_48g10bH`L?NIrxuK+G-)XLJ%Mx${VmES zQb@k#lX|}K5PtohoN~sV0oz-iU0rap#0i1UbPn7L1(v_hFv|=uWA-eM^H58;iS;enddm0;FuBI?G?jkXq$+cWF$c$;Na?ZYyPhn>BCZm&;G`o3L4|VB6_mtQLhM>rYOv?l}GTJVd^= z%Y~MYi}|8~f~s=X@RfxJMdUX?ziCOybmSGR{F3mAxv?KQ{tQO$GE>3)QTWi$zXStK zaSK^ttWa@!TX$pXHtg-<%+yU%f;3H`b6zq4W1Fv^YLSU!pji zJF*qYI*q!yQq?f{*{n6;cQE!Xv$oRWoq?g93R=4^@j*?_%sCA~HoEoHS@P2{@}n4b zM|Bqp@Q6b8Yk5}Z!eJPA`f`J&$$q3%do`XiJ_lpZ>R-RP9*ezqpB_lrB}nKE7Qbsw zVNZzu=w9Ejwl)CT4mclXYF-PYR@#0ymtMnSQ?AYEZ(Ar%;+o@Tu)+b=~ac^^c$EtXf_yrI}aCcKNnczP4K`{0*z(!-1vz7(pLTR zT;D8=zTed~yR?c<-*hJ#j3{<#1NGBFUd3+BG{&PY{9DfGBh|y~*zr$VP=?Aft9uJb z3^&f!{#ek&0k+}F*4_Rvy-)T=lgTX@8wnENi{ghoTW9JsMQ+f&drEKG9Z35Tkh zj>$xmW0+?)YAC_T3cb+va%ABE472->S9N&+S!nr@6B2bW-j?WImU$BT)_04ht#!hN zllPAInw3Dur}C_=-1ngSn68y|CBd^=bysI{R7!@#q^*7To8}Byl^$!Hs!sqxK)%0& zCiSnUKB@%4Vz9CQy}f_!D~T^=kt)}ri>s)<`=~k$PIs9z-na#EtCsYp73uMWRb;h< zNcU+%|7wV&fM?8?k9$s?6>{#PulA512-wq+Z&zmZG~(aLw3TL4;!r> z9*o{fa=pKnu5LE?tE_lciu2G;x+wd>UL4=^9^JBTmZ%#)oRjI#XS^L|`?lxy*AV)7 z=tT7!=}e-2@RZoW_B8EX7@D-=qa?$1Y^Zay*^g79j zZ``{Ei}ow8>bvq1N)DNfPhZ4;?t|17lZG2qnCPyLM;hz){2=nqsH-#IR`xmmH+z}H ztp7y2rT#qhy(LN1xarfiJL_k^gW)JG--5*2v0|Vw$b!T;T6DhKIuzy|mXEaX#Q^o; zg4w{Y9nh6(wEy^?M=;&_IijR20{Z4QD&5{i^IN0R9wD`YOLj2bTED$ncAmr_Iho<8 zOM&kD49x0)BK6+1qJ1l?|ga;Y}qveSR{W%S`m=0@$mc1_j*gZm3;)RFZf z9Xt+{+KVx_%}iieC5G|lby)}xdii#UfdyttUrn4I(T48E>uY90X?{vbul=H7$q)-; z46SP@Dm31H7|{3c%jaDOUB=^6T(+Na=ru#xv19WP)yDy~YI6USn^4`6YnHRSF!$rR z#6`a(?2aBUydT5~ixNYr7CeMrtG~|Xsi~O-Nr$8?ObXp$!1#XM-p*bULqqNF_KYwj zH#NWU*bxBadxm%23m!o6SeyIVD@482(HDFzX5Eiqu$ApyPY$uq4qeO65(roct+HgM z`vqtWkGMqM!S=i8Xq4nJXT`Ja^r z5=EG^;pw)kO?x28t_rH>#-a6H^Dthc*;klm(?21Uc?}mLOW%KPZ^kZV+vLxsTX8gX z_@)Mz7EpcDJwM8RfzFSn!X_S!FduQ-H)?GbG@lF;(qXN_`*l_q)yRM4?Ec2&@>>Kx zYGS4^w&%uBq{^wW zoMM4P3wKQiMxMgRwdPgq>x_u@ie0E1xrjU{>LkbKLqZ|gcRYCJ7=;#(gvFp+7NjlR z(4jL%ksde(HT7T9=HFh0Cb49;#udRZ#hd=kX=fFQ!DCYH#I+(Q9og%B&TSqVV{>QT zc}qjr#@f?^;pLFcXXO8BK?9}>S<;_8B6!;CzL~tt-z!QdZ(~8(6r0pa@Y;Co*`S>P zlAkSzI&#>7gZ)FUwUuhn-pqqx!eX@h^E`64`mnfTa*Y4XqQPnCoA@sMm6uL`!3RIz2ov_rP;s=h__{|x)s2|oYC8)9CC_erXy1<|KRnBx?n}Wixsy$Y1PGo%ip~x> zQoC^)hs91N%8qF$5 zb54wy?4hks7&iBDJsxon%2>C%NI4yX(RX~nw3oV+q>*i*Xv@rQ#14x7TaQ20ft4BNkE{(N>g z4cgWo66JRz@QlMzTc6&acnhP!gBY4Sfpg<=qoH;;p)Y-|{1-oApAGeQWpe%cM6~kOcC5TbD%yu-S*~R zxo*o}xyc3-jXd1*o)C42+s3pvs_tXLPU$RVeZ4N|6jt*V4qL>m@2lV6js{guTzUb!tAOI$LJyyeTp5iQr4MDKDK+$8kfCh;Q- zq|Y!`thj(Z(n|fT)>TN6h=H{U6*!V$`gWasI5D1;(r?UeDPeBn@!rEiERc5WVC|dj z%-AUzagKvi634zJ-A`Y`3{Js72&Ew-8643(owV)m3tddWyvKIWUyBfV z2gPlwl$#fyh);dMq2(#d z)>UM~$R5R+=Li4p>*S;2b+WyXGj5+7yp*%%7$^D%3Kyo3k^VRR}fi($zG!fwXfE{X_2CtN$FQe}trd;5l; zj0cmLcB~5a(d&d!{V<^eyJN8I;RT^zYJ?qA52`jXLEO;ypz4&2OfFD3XVcqN#36o@ zo=eD^7x43ipPSk5zvp!w_ZrfC5Y|x3d(X{=LlQ(^>Z>XLP-|lNbO=H-k8PG^&gZ`VwXCAF5 z@(0?!5AA&mMbI8BAlYLe34RU63zi{dY!mpx)1t-${cJyzzgzS|r|+@slzj)W(^_Qn z4p%Oyt~^6k+u-rnK4_1&51q%B!8iSq53a&eQo&(c?XU28>9qRN-Dwh&ctFzysrT4r zC3G-@O9eV>0w;KGbt7dN=i<55ics?`{CmZ*2RNGb=G~6!KFB}PZ&@fm0kdhPuYP+H zdbs@d-ZO>nGDvgA3TymmcVR!wIW_Yv~8qragdl>&pN zcO>~KW6*CO`r>nn2gIoOUuS#Q3Qc?|BFnU!Ab+xdRc8W?PoBh7UXkKxQv$v5K6x20 z2wX*jpl4}!v_rrKz z-sh1;Z|F;SnJwQFk6j;j7=D!?;@K3>mC5Z_zeD!#yzCLmCm7*dfAyA-B8*y;NDE)3 zLfoaErze&@hSI2T?J`|<_%*-G@igyd=yF)Uq{G35JsNJTZ5^4=wR6Rp=Y0i`xAtzB zw&`o=Q8SzmvarDNYyHeQl@EXnN?klvZdhP+p~Fv`F0P-(C~o9tpiaayyGFPeJ^0>2 z-(E8Jo}us1bFOc9QK2-9N&Tv;airnmks9p4ex#DXTV_vm+^^XxiCKS!ow>IUdC zcU$qxh33a5v5ddX=YIVVGNu+pY}PDdZO|dUz&Z_B{M70UT+>L6JaU~xrt$BC%kikS zsShKtSKl}K@lXIZYeT=;b{!~xrbn1xznGpqK{XzP(vzE2p1maU$j5JYv&VaHAoPx3{g1rg#W46S!gG${ z9_Fmx4H?N|Fl|%5E2L!{N1I+ZnP1GuVe$2&E10FQcd{Xy9+-(>mtD3J@GR_jausH)dV#)|x9ygMnQi(s#F-W2e~A2=_h^ymA*>J~FvCmlAROoaNIc#->d`@h~PG-Etmkf0SuXcbvjh`O`CsvR+W> zW34RV>if_AI{sro;zVbfVoT}i`-VUOgnzip}$cCIjzsyjLhwN+X1Ps~q2^sPr3hr$b>EAjy4 z+<-7Nojve*mL&rFQb(low(o#WE;*0OK7{>Sc3oA+{ACKpI5TGkErjEe+{ny8*nY_A z?ik2nBX}Kae)2g8*So{ihwsU!VmD*STun{iD#E`P`>OgHYwJ9WN9sP33+~4r?a^)f zuF=L3n`=ATBySVEX4PUm>r*`Vq51WI>a%d7?k08i79-OIJ)pw-fO-8HsNm0R-CFV* zGPb4|3Fky(XJca|&mfVnGq`J^TEIEuKgSPKP$Qz~`}Kd$H$^7-wsJYK5BEer-_xUf z0V@3%2i86Ki6Qez1Ir9-p!-^jY3?>JSmmBScS@?-k#ik5QJ82nf z9W0E^cgs2#jvs^h_k|}`H=0Ag!!ZW^c-pxN>&;!fk7n3Fg5me|z1i(Bap9WshRw2A zQ=Yt0!S5qBEk)fNVe=rdaQCO=tXl?UMV73}XP3~iq;8t=4w=LloIS;I_V0Uoo14yV z-Fj;pKBpDmNc&Tj(>d~Xgw^7uySV4h1rZG3ueXoD4r`Alu~YH}kChZjz5 z{aFfW`%0)`JSU*;H1&CLN+(i!m5l9+!=UkK(R;^Y7EGE3k9TirJTS1(`8dwk?E-dX ze0l#Wi^zY@GUitl+;SZ#y+d^$We9ty-jUO4sEfcqw~en3zogHa#&E_^bKy=7jBl!N zy`jYqy))(yORu)V!sQILL_b4_>#DUhEhBIj{vlDF$+f&Bj|?DctMp z92)6(;h}Bd@uiFjI-a`!-M=u&+*mG=ore`nyT95eJ3{)pFPGd6jB!cQ+0us13ED<# zMU|wl5b~QjbMoy3P3|$!hG&Pj_I(_PF?uDkwFaj;r>IgNcyVwbo$nU=8v6HL`Ej?q z&=tDtBOWcK8{p?T$zSWT$vE)#z0s>H1W!qDKpY0^YpBR4-PdFiuHGN@fzn3299c(|H9gvulsimNkO&KroCsFw&JAl^|4=T zi?O`#W{r$Y1CEAl(eUkXhc3#VZ5}>Eof(GfXT3r@>HXwncj|7<9Xqf=LWXbQQ~@qV z9=@~r>0zjRI3D=Nz=p(P{^MNygfCVr*RwyM$4%nIuNT7ug7o!{bMN%s?{FmUCb>)2?pPOviDAYh!<5ZeLa&#!aNPB z!>uenbmKcS+I(?Y-UDc7#1u7~DEe{w{oZ5O25lJSN%34WZVqIPUfFY}2^>w)+Skjy zk93jPnSPD2w9@hfv0bxO_~OJ@EPQqKtMltxXuIF|41*3KRmCzRZMPIglqEzLNt8j3 z!4vlkK?%&(2=chH%APncPrlnf9Uy=M5A>6}9PMGU-q-8E>q}UBcX#GThh13n4L5%h zcZ8{|OCgHpV$gp2)ye!c0>{uLd*H0JgDkY~G`x1v+7t_YGn~t8&cVd_?^TJizo2%9 z7j>>}8BkKKPv$YPV^;1}_Xs~hq)Iu+Jl9df=6mBG+ILH0lkvx^p4a(dK`UN=dE`29 zcYhIc%2o!OAFy@H+_u39zuIfRFNj0)>Mc%t<=0`G{gZm_&vh`UOTM6B?+BCA4*orJ zr-1A=`$pvK1Lz*od_ys%LWg+d*mKDXP_N9k{eAQzi8;O4A-YcjBNm)Fm72&^|lzjEM_#V;@j0iT;6866^U zDHY+au$@nf#BfD4cUZUsN1XOo9Wye;{&~Gd+hezZ8mI9tmhb*Q^+UZQsrK7}U?`T9 zIndogtA7I2(0GybbL_AnbDEiI@&Fn`4p;TJRzSPt@^B%g07x#_Y4wfw6pUch%=uR%4KI1+Lv(Ka=jsMx7>69|Ej)Eel=S>`yl8ndU zdlzihj0;20W;{42$426)Uo)tq?*wzri!PF{Y4{BMV&BlXZPhg0IHtGfuQyF;fq`Ro zXLfC+s~?9!*D{Qq?RsIF>6p(sR~l{>8XZ`|lm@4v!z_8vxrJNMD`(IdXV&_7p47ROFM3bNHFwPKp z#r4D|oY=2#ajNu)!5Kr2&XcyD%m;zpc}V=v3L3cGUs6|>>Hy{n>iMUY>t^o9S4eT;tjXc z>@c;a_Tp!f5VoDM;xHW4hlb<<^Q3xi61(85_?|tFU{Uel)Acg5*mYGt%Bi{(+V*jT zy)|V3vNb15oO2#@Mh)2WyC^{X$ybLyzi5JP;ouzEfWzqbgCn)Y?KBp@9yycw${pL& zI!iu1t;9~Xum`$=7mzx8S&rXI6l`FUk3GE1-xcx%p)J&#yYfQ|^n3B{*ZF;o z#B_;Ghrd<}6P>ffQ+i6^2ZKAMHfkISDN{mHHUXG+Z1uHTTN(}m##-;aT_b6i#>jH^ zRYnuxHy%;tPTW&-0w&eB36WY}!%U?}=8|JN2+$653aLhX$m!VjhzV8R0Pba(K< z+{{kyVV4FNE3c?{V@B95jdQ`)WEUd8q-9veyMUnrYDIS*J{u4W?V8`?#F^a*JHSDX zY%VnnNnS1~f~U=3Y|R+|lh+C)rq8yIw}?=Y+^BZuW)3ZG42hF>Rl5Sjpq`m)=35#I zR`@*h_YSkdA&)Y zaRmiNtk{R85Y3S^E)Udu;VYbZJ0NOez|diI5XMIof$h!T?_8Fd=x;>AmbAFTg5LZ{ z{vw>#vFaFf%wq1U({cDqsk`iOpfzhZ< zQ?&ecLn@Dn?oCc2Z-D$|gIjA-8@`car8wZJIVbeYC^0q7 z=HsLl!~H$il(B1kje2}?CY07RWQm;A#(crYXU?51r@gmBBsjn8hYXBOSx!7ypMmAK z*GE4y+6psT&Nhi5$DnIby>->V7ii`>``D4;Duz^j*Q_abs&u$~jAo}|FGbid zDQP<_y_YzJic|=%RlB)yw-NE3(DddL3LCPHq=yobIRbKsZc2|j5c=MGfzX5GN{)a_L) zJFIbn!md*s>`Yodb7!skt&d3g^W8)se*b@tGl{|R+MeXgEim;-^Uo2A8MKk+T>{F) zaAYN;ZttoP9BJ6J-1}k?w5|BEmL;$briVHU{hEn*-1L*Y&=S_iP?|PzTKHJjf42Y6 z{fO+?k+SFq%&wU^k)=^Wwa7`axKgo8H#i4 zpf4q_l;_(CB+GtT^W#rCMm1KbW_%fkiq2*;m0CuqW2&1TGVFv&22u66WDe+e)W6F7 z(FFRw++4T*?3MpE9xyY~b6qJ)4?0|yKYk@n)T8Oy;J+?6t^!8ev_Ca%V* zio_PjvEO9-5onN=+4K6j$KQ3}Cw@Qf#@}7RR(6+8%ykh9g7r zvfb`Q5Hm2Kq13qn+d5yds82aVicBb*V^0rE&AYbTv?_w}#YfJCw|i*gkmwjaeXh(C zD86@tGW6)`zCxEKf1>M&5@<`~HxdXJ#&N5g&);vBCNbAtNN*rLf~C3BUGJ7`E|PuNd+J7v`j-+Ici+Mjd&vblWhHocm@eRy z0GPX!J_QP6pq|{#)7XCCWcb}HWl;Ir^7B*&7nEtn7D73Jcc)T!?s?OnfP=?oKf0^a z@N}5-x+KRh@F^q)1d5h^rppt8c>hzn0o!PGv@RqOZ5E?IQM;T3(lIHzsQ*H4fRd)sl{x3ICfCspy{Kp*!N-4 z*Ys-4Ups*(@J;4=%oEqKxL-RC9d({-HXm2O{M`4$Vq*D7zEtrPy858^__3IeYTHSS z3BQH9Gf!c*;rc(bZ!MtIuG;#A23`HW|Ek9tS$>A;m3@KqFSAJls~WNY<;U3pG_U=T$&C!nvYk^yYuw zm&RaWrls;#4tq}UWTf4s)4!OOly~()7$c62ecf&Hr5%g6c2tByHS|@H_lQm(!V0!= zvn_%rpyp)%{O+#r7_U_8c%S7x)OOsv_il;6VT>(W$ECMaVVf2?C2;Bn)K%4??w}Xo z`@Lf?F0F^+Y8^^h8|{0cZWo)Y*6}JrPR@K56syUF1$|eZg(Jq;G51Dq=EwZm7lad-@ zawe>p?s|Xu@Y&S!w7QenzsHgP?28}R>9b%I|3(S>4O4WS{rCK{-CE63_W5sq|5Z-r z$Im6HXl^6s?Mewrieo1X4@zZkUHgl!-S&!Gs!_?OvGcv|xi5DSCr>gmy^f2=#cMq| zWsgFjDM#j8a#+AW-&a*0C=-?R8q=nBnfR^I#9?9g?SHt+Fl|}hFaD==^(blYowfZ` z#deDu8o2H?=}w(W=Mg__`Z3lRDoDdqkQfh9w;d2^ z!@A!r6z(@oSh%C$K}*m!%nm3pN zMlOFs;9MKdY=eRVf`2mW5Vw+fLm-T~8tvWjKpk;VKGQ9&1g66bf_=J@F^2ENv1>y# z{3^yZmWXPwJ;nCKR>_4xM)*9O!x5ob0ga_DCOglqh3?qZt`dh9vChQoOHTgZ|NWaJ zBZozBIELkQ>nJzW9h2m@f5n8YVO|JT-zdM z-+J-KS;7Hgt$RbYV=^EnIQ*2!>JR_<-6V#Oyr0V6(ee7wjZf-9U-53AtN*8Z>HlB; zV#EDYYKjUcah!!~d-gqqiNhhB8ipU~?#qrAuYqS6YmaC3o!i*a zS*K%b_!$dc{yHAK;vhD97^G%T5c!C`yH_$@KO2Q#N{lePp%*d?4zoz;nB!vGjB02K zQO9&BD{B8uX+FC57oJ!QJF#5`dld_OvdFagN@H63u(KqW8HP4J_S(6Iu5J|Fe2vL+ zZkNxKq`O|_vn5bqa|Gn_tb7(^WQDgYM4!)Xp{qAVCnt>NZA;!>ba-`IyXg}93a*~G zNB2GbZ3$y1+fPEd^;~8G?@8B&Fm zv8VIfR`OF5EZ}8y_Ei;zzCJmIlD!|W-eOC=j`$boSjlhPDzF|0l&_^d^K$~vj;>t3 z>rOB#_>()5fv)aS8Z&iWQO;gk9bjmCT4FVCNbi599K3gC{!1kGO+Uyzv%eZgzeE)J zaO}nDS1z$?hdz^7G(H?Owf%)Nzs#8%&fSNOZ;>2qKlHJiA4=O}C(~HvDkLe6V%W2w zI_HwCj}hzM+XZ^i#W}$1YxLcHuM%*E|0tz=J%z+Dd}hm(91+Kz2fw?@_50}hH!&>0 zoUb5}#;GHi%a9a=^AncYKX`bs!(~)&Rr^C|x4U{`T1X7%myljj|W3~i?z+Gd7ibBpkp0@+#N!W2QczoEaLRUavXpD#l7__ef^4vZN2WVwdmvr{+znD-%o|`(@dpL~}dB5@ddhpSyt!Svtrv&Q%EiD|n15tm$-?*H7xi1R;~Y(fg?_zW>) z_l5l{)#&(4uz`cMXLj(PJo$9*AJ_W8^izN?zwQ6@FSJz5dmL5$g5$aL=bRNq<{FiRYv^c7+2+b8|s@@e0R zU8j4t9XNCa#zL%Acd-E`K9yT_*}mfce*X%q4dOP)(~Vczfsx_b1v*{~%!dmas@~N9 z?e;x7mX2qV3IA-m-@v0xpGq95*u?%|^H!MnKDnX#6m7lYps+kx)F#sYuia})>AFMX zrG@g-`wnyVFN4KBTxrj~{9EppK6mq2mD*$fHU$}d!DZN`Sugwf;9ICTy17y`hjzXu zF)b{KhA0y2npzfkFTt1{$fe`wZ_20OL==P5I)5S`!aZmM%aQYRb)5g}ICN>rWTiF_ z|L65I=G2^RCpHEW-@7|#Sq|`GZ$YAgv8e`j7BtrS>>~g3I|pn``L&)^+^Q= zG?x6O@3D_4(Dw88eBb8(;TP&u?a$m~^em1>@DU;GO>y^h0~2w`kU(2*H2qw-IG66!5jE-I<@@AmkryI@>1yT z)6dUg0hPZR+i7DjRqo<9%QhTe6>&yLxB#)~%GjYP`Z|@J7VSO$XXc>E_3d4G6EoWN zN%hEL;9tOzP3{#}id%5>om$mX`8lY{*!inWuLg3q8Ewp2BY7;Tg$9feNaRl?5jR{7^M81WVvn@MsIkC?>ef3HM>PkV0a{nFBH*%#WWV{~+!mdiS{Je)vY&jhEr)cW{C*JDWd6~t~&C5dd9;xQ> zOsKr2U~jEx2vfXkCgZG${L<7aQVZkoL8!Xh8OpV9E3_@Kav!^;4%sGJmVEK0fA^)G zoOUk`Lq!xQ6^y+=vRcjF^NL4tz+fNya^rPCy=@t9cv+3al-aQ}KzN`C@{7 zs+QiLJu|Rfe9(9@P!HOCQ#RKA+=q2nWxQXS)9FiWG;mWgyZ=v~qkpdVoQ7C!sj{cg z;`8x}Xyzjf%68uz?6Mj&j{myY6GY4F#+KZ%5v_H0SkZG{V3QA>p93ea-CQhN?Sr+- z&z$Wk+(=$8dXM*e22?8D4f1v*>~c!pS@rUi`%peB#V@h(7G#Cx>`lMD1}k)Tq@2{` zz{cL=C--j}#?|lvzi8xZmyYfJJr8;&53jxUKssX-W}RSj=;eS=g-R>Qh^C}PM5iuB5ZQ^xH*62 zERr8?%vp9nA29akK?bK~bl=zKY;$&0wh|~lrM6zLSx8LRpX`h-8Nr&1LQ<7#sn~hy zoe1lSOIUfcccH=K3{1Xb_sh-J#Z2alA44j1u+~Cg+mCxjNO`LCWbkP@j1BVdH?Mh( zJt1GC#3zPvsPj{T%q2q{w#m=UiP!)|Kf|=wRtsV0lZEg?ZDmN|^ublTI-zb(YuSZ_ z&5)^mB3-wI4a%LjtpjT|jPR8E7Vaeu^*=9Hl6RWWt*^QI@;g@>PSIT-WGr~SH#rV^ zFMD47cCi&Fe{R%2^D!0c5-zaF81P|+VC%yH7GaE9^|CuoWf#sb?uxOt zlUQ`8Z$C{!9l)b-9L|H!B6Io2X=BShcQNp zX99+kP<`p>=J0M7q{!ULUo>XG2QfB_J5HqGU_Iv(mY?o_&xtk4$`*1s6w|)5boB!m zND_4jd&xxCe=2#~hP&qozKPiih9=X}P-w_lkotLOJ*F|_S*>cO=|^aLYua(<21%5A2-JsgmTOpGw00pv>lV1@AtV&3CzML#O>V%vUmX10q&UPq^^T;55q zbR0b~Jzae07nH?R{BpuS*e5!%|D=B}j1S&ARxjm>F39?pJvT3NO zmg!%F-Y|X2;TZ+uJDj(#ABxn4(*E_$wYT)>#%=Jd+J#yFjnMA7aQ;n-8xGX;yj^2H zf#X}WHZ7l`%M+zNM-j^Apy_A=k!Ml6G_P}1GXh6f-X1F1T~60dYv0V_z|?W*Ph@5L z?4t>zS|i>TR=coA-AO;xJ|$t(DRZ5X3Mm{B z`CvF-JVG}wEVBnwx}Ow4*@n%heD3x`;m@5{uS7oucap<_KkVc9QglqcQ>hGs&-op{ z(IbfCMmK6!exHI`jpr-PYn-5G-%+DXl}P9vm0D?FdLQcd_*ZU)0?0hDi~HTBqyHKo z?DHqzdwPKl3-NB|fnPMUhQ@>Els!*VFe!Fty#L*+(#V?em_ik9=V|vtm(hnLTZsSyu!ZBw7>s%H)px z(qCcobmQ_X?9Tj?0F%G+pqP5=RTJIs(@|gL(|BG%?Y2~lE2cTvzFf+#wT=Rl$KPLN zF%iS|`7O)D4!C0L(@R%=bxC8rt8wYI9>UI<%t}2USLOm_KSkMz95rk{yI1q8vE{$M zJB@MMu;|VCfBIqnIbJ2gUi;;TvM~4B5s~by|N1;u`y97bTt*j9!gt#1Ns5{T?xZvB ziyqr9!Y?$UJ1l*vfewcU{i!R%^(nga-@{chTqV#EA*WAZDqX|16@_Jic^g7bnSO|S-%+mPUp|3wbLv8K}%WppLQu-dtFDejN&$m zL5fo9NulF}|CK^=_LF@pij;fyc6A!6SaC>YY-0H>sPDIBs+1YVk#0R9bIY?(yo;IF z=pzqQDJk>T>aYDjov;7b^lQymjy;DsPD1499x*hw=J9!)-|75mq{7EKk z+xbh^yH5k!zMD+S|Fz$%EgbJ=?j47|MyDe8rtbg8{q{jh-WNizJ$lL4`+%r#Q>xORcK`N6q#ATMw7>gn z_pWR)|LF4_lXP;Hi%X^<1%Jc|4v0ariw?El<>7y}m&SC?#NTz_vNWca=GL6>hgj3@ zU_s_M1%0^KTs21aa}mV9ox)Z&Dbx$$M(I+`9WNP z*vivg4orpeIBeFk5@y$x$^L&Jzap+P^34VBs@p5ZTveAddnn&EI>2`PeG zo~`*s z=N=apXs=FD;{BmdV!FFMOZ*LucM^wFw(xj=For?)gFBqE7;)s)Arnt|avH-=HtpZH zY@pRJBG!0w8cf?fbEGiv(VlAz{gd?V#Cx1}+_5ZbBi+9J|Jr)3Z1#UPdEOq|dn~g`JCz>_w2p!&+i!H{Hd@^pTKoYE;CR#pmrWd zw=xbTTxO>0=VVZzZ(rBN|KstZS_~EXqA1La1!rG~zxw|xy-ng0e^kjJRQ><79Z2{| zS>TW}z&OE<%Q6x{(4H3~HTiN7V(x7Y^0d^4rdjUos~B0}PbLfFJ1>s^-F7j z+~@!6`@1DR`3gS#r~etJ!`musinile?_7k_2D-Q}(z{ft~0@)Jv*W z%U>LE{f#4gWf!Nn7s33M5T0KOfe_11Wk`Sa3QC;b-$~#n@+c;HzpK1SO#IJzBeAG! z*JwMNAk~&!xl+WXnC2*1L_ga0GO}>#BiW=AE5AkEaj+El<RjQd~Oyf z-4~k0%pT*Q;Te6VJL_ON(n~MR#Sy9f)5h1%eZ)oGS{=WY);Ox~eeS@Kz1V+w!I`AF z0Y(c|uFS-*z{$+bQL$5_bmNfcWSef4ZwVQ^ucDd5Y5otKFN)hwZ71qTRW#erm<(6J zq^e#>o_`vN?ZbM*mu&R$h|wRnb(U*igbp(Qi}kNKVeW*CionMmSS7}Po5QA(kT<4% zC(pjHBC%MVsb6u7kFK5h(tEbf$yXq8CM|!~K@!?eH8wD{Rl+#`jZ(Jrguk5PQu=l> z-IQpT2tK{(#Omg8M$J5#K|7d4Tfa@N-O-7{6J=xE?0O zl<%2E5WH+XqJgJO*JhDelmwOq4@AJCVAZ6G6&*h>iS_t)@gG3s;d?Q>SiAlYEuIAP ze|D(Zulj-H5T)d6_fI4F3`6d{QWX-D%%}Yqwk<-VdYeZ|EK%Q%dZ_8{JJM$yZN1F? zBhvsaE_Qu3TTk!+&1`G1H0_W^Swy8R4mntfLs3^RwS5a#@< z{kO1=)Te%}YaLLx{XBXkniIRuw2z1j{vgIt;ohT|UgA62TsePO8eYN?yGF~6x2tgM zy`eJ$lOR<2o~m*#W5FrW$isOZyK&mOhbP|M4*fanRvel90L?u+yO!_1@%Mc9=|wno z3Ms~H3QDXGaq`!CAZK13=y30@4-{HvceI2EJ|aNMOWyW^N3F`k$HWP5p3VE?_?lMmg?p{wD1 zxn;l!97~z046yur{<~<&^(?WSIQQ^xGYE^lfz!^eTb3R=!orhwT(yJ7vx5`HJ8YdM zRzhq3m+bFmq%gAyiWV7o6{K(Bhmh%gH;;egMs6hDZdJCGi`?BM zDiq!Q)o|VD=B5em9LHjD-h2r_t3(q;E&Pz|Vc^ z6ildOJDc);hvB;pmD9$ybob@nZE`DFbAawTd8f~H;fq(XUvW9RBFk;4&-f^{HAMrd zJ2>y!HrS9@4=Sym+xpM%ptaK~DaArgrsa)7Tf*k@HJM*w;QFi2w-fD&@7p?8$$XFf zU+utzUt@Aloh??>28gecq19c-u|ntfYuBV=He^=2CptlcvYLP)|8*RzKJT(F_A6a~ z24(ZL(z~0WCdz#3=hpwRo-GTJ)bvFY#aXKq9I_%Xy`im-DRA{KJJT|99k~2aNg2fsSh26&U#x+#Bd@z^4^b=Z;|}dh)-Ab2}J6>eAl7K3v~&nRQu8-u+KB? zw-grz1M;4{YfIXTO$rR9KUkBI?9Td0*PMc}Ol@wYr)6~WJ@n8);4Xt0p-(T~T^Uvv z2_wsXDQOEGfim^}nE(M1>`%{Mt^d4&?)eJoJMXGjMc{;bPv$ehY>deA6BAc^gS|?Q z28OHH3Ax(v#9cF^pKe?#GG*4gY8K&yTi}i2$S+u^xc}qQP$My))dsTjZo}AdG-` zJ5h9gvvM~MsJP{?S$2kQ{2Q)twmf_k@-I23*Ebf!d%sti(Osu|wPIkX&`5rjGuAzisDiu??f;3lNu!X4ug?pM*=(Y_ig_5CX}`|}kFZ793PJVp1D%DxNI3Jx`}JLI znm)nNs=FU@o^GLg?|8?+H^$#I90T@LJGNI0mEyR%)OeH$kuMoBxH?ME?LAhGSSroW zzM>n4`SG_Gt!>X?{Fg&qUyrAf7%oIh<(bg&$6@*Eny)QpS};nUdGp~K!OORBBuXK0 zJB{ZQM-DxCD&F{s_@D1!=Bb(=&?~fXBY}Y~FBQiloT5L!Y{cOZnX8Id-@s6`g!eHD z4Of5^kt*9u#Y~tc{-S)W07*c$zkwI$b$r~ab_`%GpUbu)&pnmAHqX$mXC^wQ-j6lKmbR6#L6Zj99gr|)`sfc*CoFzs%Cq8t z07d$IERnZL-T8+lu%!xSDi2I6^tR)e>PGh?6=^tWT|9q8Q42df)8EL4A4I=N+tRpZ z51iVux8D8TviX-PHj)b|*k=T#z zQTh6vrVrp}Mwj>9ok}p1rSvOG>mEe7v9hu2)BHKud1rNt)Mq)`{c|7qRoYvMNz89# zILyQJ2|1coRp(MPfkwWTgG=&6J!3|4R>rI!EiMX_3)Y{egA=h=bxN+@R}zQon7p4` zI6%F`H_{*^A?k(#I6BZnN z;dn#y>q@LauXy!TBWRo7CVhtb4u`*gSnaHz4Sily&u()Q?RK=j(7zYdo?;){FsK35vvEdvI?}&p@qV_9$XmL55KEHD>rHaxP-21i2TIv zpCWJ0MDND2uj&KiC(Z+3%S#Dm~lEnkF87I{iVv{kG$!Z|ra3xOcEL9bHO5{gS z93#vkQfzQk#P90VpVKhkMB02?{vo!rp1#pNwHkYCZp^Wrm4%7*rQa?bS?Ks6aAM%GJ&8m8Kqo2}kQl<|n#(l*%G>8x;`Pm*Q0U|!8S(9}o;V?> zcFN8TdsH?t{M?)g)Vz|NzVDaeceA(L*7sRqhSWdUlvRM8Z}%L~P+Jcj&Al}qLQ2@p zKkaGv#0rM=TMW*f-HPLnSbun>I>H3?`OD+$tNzO82H`i$nsNAq%kY*6CLHbgRN16& z1ogt=0wrPu|9ku75vk~eaOg3VS%1ML4Eyf0Zxc3JkIe<)2a+hL36JL%$0#J*X@v#z>; z4qpT{YI#`xo#uzdh8#Ua{VZ;npY zgwuF1vHGszDf`u1p<#Zfy<{~pehq7Kj%JqUVNF9pTYTXI99i9bf9>~3>{d`(VO#Bk zeP2&jvqtpc)au0Xyc4u@8BT@Bi|pCpgyYM4kL9UufcA>DpT?%!ag_6Dw1UD7@Gp3! z5i3IMZ*4R8H%YH2;e?vToY=8^j7uAVwidCQMP|7`CNo)z zh1X;8svR$m9ez#dS+heA?egjI?-QzS3+dCe{1a^9n0<0LO#us89V{jMSYaUOPGTzaD&Bu2laRCFuZO*aYH&Db z?n3wMA-ear-z=U=X{)1~-=_X6Bsoi(9AUHTJRbLbglz&RE%m+V;siLi!dJP&;h%8{ z%wf)`ioE*IdYCw4V5VN^TaGi@@wUSTC(y^PZ&&y3W@vBT`ruw1!H=4KX}n!QP8h2m z9N7PC;UV;2e1B2!9#%_|y4`ME!k*$=$GF8O2zgn{d93z~5{`w>cQEgx z@dRU}Ww@WuC3Q%XmMEBal!AiC-U^{x*JA0|ftTwu#Q&Z<2MYD3>j*i1{PsrDbpl^D z)pk#6A!i3nKYdPBeI5ieoW<(-VWl|rl(Zu*Bo?V$87tn&5P3xjcN12B(DH-vrSsbv zQr|%HaBNvO+S84Hy8uaoVU+G%43%8wvYOHY(D&^7?)ZYE*m1sh<*h_kY#eGhv09)F zJ2-B-I==s3^;+eQO}VyWowT9k&*t^exoyQ**||{?6PLN+Z6ZGgCaj!Gdy9gpJkQBQa}5Jn1f1fmpL#jT^R|&}a2~`%|(b zwjUZfDmYJ^mxhum^G+M*;jrzv8l!z7^ndzwO5|`OF)z#I`UXtnNen|Lj^1*zM0fW~ zYKw-eabmCFG7%HvoIPyVTqHVu7QS;tqzlAsATiwEaWFB1@cWGln^JDY2ZnUl6MjfCMMu^@;uGc{>Eu3Vcn{_uId4kST#bGGk5`LQ60lHw zFO!W>8L>Xfp1=LQn2l{szrNT%JBpn;<_72P-ojB|N$GP9Y1pDrDcvh}AIAMT{jE2Y zK$?KQg@Y=<)KAIq4S8BHyK*@6khC~eX#yRNH@E?3H0u?vTb%f1wB?W+L!w6U|eaUi%a4(Qg+=J zox6GhiW9h3DrwX7Ft!{O{Ql125_YdlnTBavTm$pB@;m+#pw)-L>AmY+UU638=pVH| z%a^6Wq}e|zl6A+>=)mZOCnN5a6;UFSM3TFq)y4G-r>!F zyxlU#gBN^oOj$brbAUch*pdoH2f69LXOr~}b6IiB>Zcg&E+O=A=LeF8<-}SXOqah_ zxl;gJde?aewEIC3Z`00(b}=YE@Nm87@Gw@*=W(yRI013VK8=5NKT2cZ)1BP>TL|j( z0xVbMtpLilD|cPETd<35`U&%HS{@ztlrQWiKdQru&Zqm5eyU@gLa<`ZQx@z!cv5lF zh=}v_v|4vgU%m$moX2NYR zW$qA=Ph7sa$KVc5UNhK!Xfg-q_&mIvarjj=qfNjk5NwU%6d;>GZ}ayooJiu0wcp6HbZB9bO&chOU0KSAPi_VL@HLkzvYP z65H1G8}e7}pgo^Ae09smb3}ce*4JC!SPv8D(y1Qj>k%6iaEQb9u%NgMWGi3hG^(RL zj~(mWBzAPk;b7M{ISoY{7=$5}3vY;ZRBRTJuKSMWZ-WjFAKsM{+_12^z3JSY7|2Um zVC7z&N#ZE;`{^Zp1+#}mP8mk7fV!vI8L~p_u^=+twBaR5J^*`lf^tlP^g={BmI_PXMqnCW0(|8f@*wp`9 z`DB@GOlbLN5NR=7sH&9&6!pW6&k|oj{jGZ?Kew$V-h0RT+pU^4(CcFNnfZDIW=j0p zcqXuiuHEMj7DnJ^Lnm*Y*L>xsxM==-9N%~~O4*r+i*>(djd}Lx53wGlN#ie9AHeug zmC&mkadhu#7V+gSN>(A}U;CmJOKb>EY!=XF=xE1UL*rbd+Z%C0e`5scB3<4McCTQX zf^fQgeW>&n(z*Y-8*9uyYF@kk*WS4{#@~8D9CNm1>s;pXho$PuhF8mJ>kr0@^}BX7 zWkT#5owGMiBp_u%k88_x8IA^otW9S*W^$_f`<#4dmG zyIPM}7?7y245Bu`yr|_tzAj(rugZPcH*^JiqASEsrK{nLBB^IVfwtao^1GkbRI(?w z@7Z~h|3(zOoZPC6k>HNOiNf=Zt2&=TpTXDu_12*n_T|}O$yIduEiimLdhC+x0(9Rt z&*&D|iuFu&uOHugfqm;3%+PlRfB942@~778pDOV1|JjoVB#+p$Z^?AR|? zThOrSGqwv4oHjX>jDvC~PB2zdF;0N5yj?1Y?zyr;{(IPGiPQCOq^&hJVE?_u-VD1* z>@Bq<>-#j}P}G^b_kXA3>^J#{d=VN>0TyH~?zCA`0O>cjTK&4Ji?hofhS`*85PDJf zI;(3l;eQ?7@?^v=y%@VX&XR8)|2JQ3)~UArj|^!%bVzxpy=qtbAPzgIy{kRDoy2{( zHcEb73hh0kzcd@I1=o^T9VtqtO3E-{_U(;|)zAO@9fv=;hvLdYn0~IlOL7NcS2W3r zX}%dH*6-4<+iv(!D8hD#$rlFg z4X*w6e*yWr@+R!+v~xX>6C|hlq%R@WD&ctH>l{d~V`*GU-GI}xBq_d^+CX)TdBJp_ zsEaVr&nv&KjE=7ZqMQ@9tX2MoZOfAGuYJdd-5fvth3*qLxXIx#k>tKk?6B72(_7_+ zJ>!})9$NIgACoa46S8s|+UHsR_G=$bc=c9D#EB5!8>}A9dUXa8E%25^7e?k_5AHa9-#r03m$VpW-VybC>myw4b=2wb(X`*c?;mVtSt||W-%f?- znTev;w~~Il?c&fO)4TCh`*NJTz0c&qbt2E3`)XaW$rcS9j!^l1QI+UduGVV{(+nnB z|L0!FJTBWT1H*PJ&d#nS?5i}#Ep;95=+8YJRdYY1yS|bb-rTr47)98j?Hh%LRct6Y zXf(msQ$fc|NV{)hurxw_b^@ydVkXWfFT(uIt^lt;2XVTz&e+A)0y~Sm@>E-%!2pxu zw7w(Fj>l}-fYWDRMdREpuV-h!&*GHq%U0?J0qoFBKECGkSsa@c7BkdcA~7pJ_x`rc z9A~&2uh$41#@c|F!!z4DaYEC=)NPuGhfr}-WvC9#e}s|kyEu|I(DWf@Ra~gelAy(L zNz8f2o*xhw8!p#=e1~rQD4bhfN0|;| zM@?A%=yogYW#RpGVIz^>*#9U?g`JbOUqey<(bkv?~?;t~?G&wXYKAzGaw z>>PR;q$NPe(^!gpUBk;$IJuNx)vddk#7L2d)cJZ8XVwHhs*mZ0@=%kRnN}k1Sawfp z%@(C{oQRN$KOD9l$2GddW<_rU#p`g#l1nA@Q^O+`#mq?zQbSklpp)G@Q>3@ue z==T+;NBKxhsXNx!ny!QzohjCPH#ms(aOID+ISgPPdFDwC?K5Mh~ND%m}*5~g7%&yYoFY`Xx*!m3LCgyiD{$V+>_wV}kSv#t2 z4V`WY%3Mm@aLE6sz+1I2%=JteQQt=6bAURQKQStQ{BUrmdR&S-9ls0oHZfHkNnMU@ z#y=I_KBGT3j&n5J+WA!zDLr``9VIj2w{LyiI!?mRH^wlJZuU+v=gsC(so;cTwGLCC zhF4-=&47673#|lzD&ElvlTfP`i8sv*MCT8o5VH0P2d$%K9 z{R5!rWLvV{zXMB?`*b8_b#ZY2?mN3m*WzTtBeA+rLawNmSGwmm@%`uX#JqVk2AnT{ zj}!Z69X0Q`;F!`&`+^2KyaqA?@lPOx^g;{8Ow_D4uvwOvQ5Tms@y#_v4sTrOS`q1aED`uD0zTP6pGx zk6glG*3;@ucfIfO>kl0FnEv^`F3<9)Om=!a!=S3t4~dB{&}010WSo}+`#1AHi2Sh$ zVmpN%?kD6F>q`RK4E2co8glJrS&J*q|9T&ZQDlXP6NX}DSU3BpW(J(-G&!~Qu>rP} zSQVLE_-9@SiRn+ou^r2ac;jrnqrQ8J38t~1Q)0Uph`Ep-m7_;H|Kaq4j*e+=AX0c8 zw6@L_W3}@Ag>BCru-Ep^=)qg1Fne@eTdM!bf6iaS*mbQ|QUr-X)zz>qgy2OWPv2{> zRXYnKFPS)>|6+v(olJvjIZNGBhLwPB!?5Vn@Wb7@?eNII%J8-BCL~oGNXs>NpSu1AbMF`)@^I_-aK7hE)Xb zPQDcL8wZOgI69`RDzKY&z9X>?$z7CuE`cc~+uwxbpN94#NxgAdYsl7#*(C(eXwMJ1 z@+!Dqr|EyJv*3+6IcNlJvT?(0P4u{%srx@;tnbj_1qr>hN`6>j(H9G?x?;963*k&w z+8HK-PY>&6Ti>YuChXYKMJr98Qo>)=dqelY4W_^6u`MsSZ#%MM;e16ztqdVY?JJ)! z8JWL=nfB8D^7k4zn6hiBNwEqdH`mw~8Ek=ZdAB{A3hjX^!W}DznNYKBDn>We9(r3| z$eGp+VMXj|3Dz4{ND1~<>#%mfp76SBuBYD+a{u$UjEYkkRIM9i>@SwVxeH2>8F_XX z;2BxA#y}0nsra%^xD}@HG2glyyf84yJv}(J6?^hDu8_L?;fJ=`eM{aHoGO#pze95h zN0#|1>+;q@zsyYcN)9?c8R)hsrbvJvwxsGcr6z60&cR25JGz~qh1(?X+P&}p?L1Cv zXH>xUR&%%-4uzHOTRv+C6YD~zQ*?;>h11;r<;zPAai~<#=J>s6La&T?S8;7khS3#E z3Y*??;$YoctNwfmod39!9GRa0hcwYv~w{OefgNyITJ(7li$>B zl0i1McKTfZHbLO9slAp1>pqGjrLeYR7auL(8amAO?;SP&jnhgi(`xt8@hcMN-yBQD z)$5N#m{9)h40R`r&+HnAl<0vXuTAUJo)CErQ%0uVa(>IPGmz!Tj=eS{<{+~(Xrc+l z)9wjrQKm52zyJK~lM?LMpl+gSY7NvX9>?@anw|j)$1&wkM|ZqBz3X&AS`ScRNB`sb_5U z*m;a?Ewox>#OZ+r?GZk`dR&ReQZPJ&jQ<-u%dbCyOQ$`C`jP%*keb9xg+Ku zN*~kkG9-pe%AD(eFUP^{F)!`p3I6Wc7K80Km+X-8YxfWKB^xN@IU*C{BLI|5sUSE1 z5E_QR4)X6-z^Sd8p~*dNfGc?Hp!xj0LhnNAv3IC7^ZlLX*Kf;m;i3 zPCofgnZRvNp4-sWdME7fIn?R(Z6xgQ{t#PEk@Qg--9H|L0R{#a}n*~{J^$95bO37T6AQ*NERuN+Q- zDA|JpZ%&0n_8F$FKJoe3>}0{zZ+RNq3v8tDPYY0gxj!xRI*0?aT%@E%@ft4CY^ zI9O9omT&t7g$ZHK%zA>DaiUL;zp4;&c>Xj;uvSBhxY_*w1)GFg^twaD& z*bb}<<|57k-Gyc*!Vwdgb8Jn(u2I5|OeP)t$gOILX%BS9w>l8`@X7k?>XM$cydG$| zclp}TEdn?3Lq5~%=LRBAyUcU9YMRCk9Gln{{zZHR#Hci98m;;b)5q35IGb5OVkmEq zwt1+7lxJiz*vWl-t(?^gn1&u-me>1<^(ABDFtsxY z$OdVQ(o1RB`~JH6upbd8Z$H&L7;(r6$uBN8P90v4l;WXgn^0wpjgr0LXXk)J_M4i1 zEhHo5v`65+vI-nNmQCH0PFFV%DaJyLJ?Ch7Sl|>d8MDfSIKL#{-THdyGZ7~p^n1#= z(~Oqik7?pp750z`JCO3wDq z9^Kb%fEpncQ`sFvd}ncx&%}wk3$q;sD}Q(Yn_qzXebc0F9wEP7682HMS+j9$>H3aV zaa!F=NbsC9Tr)wdvkqkH0As$tK2rK}txM)OapY~xhsXQfu#<7i+8N0V%%+Ska})dw z6y>(B*NkQVue?*tOJ+$;li!Z{=PL_b`8N^cYT6fio8c6RS>~0)*791Mb_dPS-grz* zY}{q9eiSnzU7k9vUIfam%pDuw5b?Cy)(5gdGuv<`(2Tc{S0D1;cSd%P{ zP1h+f^)8=X$}!`g<1nTBZ1Z8!$Iy9Qmgl(Td+2D^3W?wyAu&ID=malvVC=HeVG*r; zIMmX4b1&~B5`#ahv&(*q(Pqrm6c1R+BS^ER*I{72-${w0tCUN|p%V;^b z8haYQJ23v}fXS;pp4!1@FtOvBci-h`EO{Wm=2pQ4Brhm7Ol7QwrSwZ@qDBwHM77lm z>*LiVX5)seSph$!DBo~6z%+_6b=2Jl_Lsx3Ldfo-7l}ADG}(Jf_$K6&_!o5*gJ4`y zVM16b3(DDJKaK7lhtWM+4Iz>LBTv_tT_)tJGUR+{aeaJ{m5}q9n6@p}hp?;2>w@Ro zJJ8?7exuvL4CWT(|1_D=;!9APyEkrI3>_ad)~1|s;!!00G4s;o_K^e{zXFU8yf{46 z(2LEzN5ub3)8Z%4^Ot~4WYJJa&6M_+ z1&wb9Cce$|MK?ai;ysnG!-WC6(sRAdk_nug+bYH%p=()?(%`$qIA(&A9@VF&W*ZKvCP~7OxE6m-!x5ltv3tjT zb{2+f76o6NK1E_Yr0p`@c?ik&cMBg@#*i2k^7A%jv%#Odyc7L{bosX=7GBdM^_NSa zKY37Q+ml}~v2;)E^lo#^RZ+|A_+SctYvv@iG=fPiYu+6f9^?Yb@9;lQjRQ#xd#y|i z=YvQLiaczLOn=wG!vyEOM}puzYrF4$8`^tdVeiY>H_J`|Wx4hSl}kHd@kLdA{O-+rZ^E=$TI|NVG`WG!`Bh1&YFI?p~!SjP!qXMXW3k z29z@oam%kFu_Vn%<9)QpP~(Djf5c8R2@d1d}JX$>rZA;wrN&ynlx6T(MTN zX}oGT%`XOeM#UoPM#O%0bN;L{b!P(Y|8C~cvlp)Jf`MQKb|vd*_>(E%b)`&i1n?D|$k7qy`aso2*NWIS(F03+5?rJ3TkSb3y(N_Kq%jGhlZvNfB=6Af*H%WEC4 zs9;~jm|nqx(%-M3UbK4cm*@ZH zFYfbUeEWitmY+#tXd0h}S1+OGsN3=g3nKp48#KO7s>U3~{CrlQk64C%udc0O8`_KL zJj8xrrx;{O_Vo)~J`Nqp=jV9hhZNlEBf%~N1f7bP+>sN;Jltju?x_YBn zKAx%XI84|RbcT&V8qT!GPOf%{jj#*jsF?aZ!>5>jj*e z&wM-yeRnynR`SyLen>1O7h2LQw&Fm?NQi!0HS`vLP;&t@?4x>I`w*cEZK1snpQ!zU zsd>kX-YSlmt)QY65pxWws`&>OkIKPFneTyveu+q_TQ76lnEIdn2J+#9awkdugr);%{m_El^rC;c3o@R}^}7CX1;t%eas>xWr9RGd1rc5098 zc^GbTt5j`T2J`nKvkF6vp*z_otMS%9{nD6y%T%iD60M#!w2^mhm6#8~KJU_i@n8iU zKM==Ze9#XkH!hyHV4&lNgE@Y#l{KGlKt_y6>e*lQIPgxM>={Ym-KJekIkTQEVfpC} z_pZXz7}@6L{)#&SC>^hLuKZEP(cKZj1I1U6`ti-;ZX%Bt#_oPp9gkzicrg=y`+bwN z>#f`5`j3j;$DtlKX^SMHp5T<9#iBf$8dl!UPIQ0G2R#Ob^3wSQn8|xS%4kp@%3p5f z=WeaT6kg?D+uzgX1F3scg16T`1FH0&ZO$E`B*rU_+INhaaC9Q*M%mOYy78JGHy8~_ zJA@UAVi7g!uc0AgDe|^+26nk8_ls@u!BLk`c2^g7q$)Yv6zS9I8)NsL)X<@vFs!>{ znqr+HgZ*z!Ev;KyU_SOnF!$bD*uP?ThQ&c{m=}mVWL{H?{oCSoqw;C>Ja7zNsIKg! zB6U7#(&|wGq+SuPjJZYV-;#|=jm|%gV5(av@0U;-|0d+V-ybmcspG0;Sa|J8D z3SPx+S}KR+k4vSpUPS&?M!v{n{RsycJF6_Iz2Y!&?r~8$_vGk4>@!^xz_Mi-knM-1 z`(k-uGW4}~X~hee{C%EPLG&=pko;dCef9@(VpYixc`ri)f36Ir%LK9a@Gg;~{Own<xR{kRH z=}uz!qO+#sUN83L`*ZXhC;ZpbHmSk#brMMBlp5^0vk8VyH@d~|PG(34>1KfI3wOVTU9PRCC`SB1RiVip~*0f{ZZ;EAk*I^Fm*=Uv+8c*h0$ zRrRzg>KUQ@^W?8ndSsxO2MG(Ve23)@k-E=0_TfNN{RgJf2&{cr0|&S7f!Q%zc(KAdLqxg2Hk5U4Yis)qxLaN6!lmUp8M44+?kn;~wELmvnB zFJ6i!V$6Jjh00f3xqy+8qNMd?43kOi3q@mVW{@$x2&3P}p$r zMnH`PYcO_mXT2;dA$TvBW|ES0cO_xPXSIMOn>ZW~jXsbWoQ%T*x0H>hLtyd~mwk@} z5oe!zY;*1M*-)U4S=n8&yh~zW9(dKkM&Q5Z1WM}{o<4+j6pGHF(&1YOdE>g~0HI|{!PeHcvs(daevl`gLoI?75F zm!A7${f3)|o}Q^9F>X?LC%EZ8&XSjFr)kmRJ|vcgFTh~z1`}~hyI%VVV`m-PUiKv- zUOqnTyY{W(U;QAzw$75-2Zs|lnfF`NVb7n<9D_$z!faOQae;7F65E7FVphXSAmInVezk~`@5nrQu7yfCD6AzD$Px426FHD5s2~pgniTJj zGl zF^T0`o;D?-7sfef?1V>+apE+m%tQpSUybg$s=l@442*YwSm&wW0zHmbd9vb4VY*TA zqtl@j9Bs;Zqpi#ihdvdikL=i@%{fGS&x|7qC!hZJd3aV)etqo{y&trpR*L7+ zN*0(M)Oj#k`32k0&aM`n)giG%4rfS_9AOV0UfC@qr44mz@o{8bBL3L&QFV#q{ddS@ zd><^chv4BTZjgU2K&JU$NX$$lD6DTwVlZu3^`(*EC90QLsqkp72d8#j(>-vQu1+R& zOE}oE)zAJ@ew%{g{Tffv@#kXsRbyuJi@&j+`}$p@l|&qMykbdEH}MqatyaCA{LB|S zjYN{zTLqw2#Uf;LFBLjXrdju${tkWCduHmt8j)C9+c?Em8vyxwtxk)V8}z%eMjdx+ z!15E_EaokYg#LbOZC-KR0Or)B{r8Z5;plmB)94$te#62~wgKDQtk@R4b%m%>DHOJ{ z?-O{OhND(K&lDEZVIX~r>awC0faLQzv7Vn{;kIjsVfPYHPYB}ci5swZdz11@w^xvv zB){dEJs%{^zFcwB-2`f0R+f3`#gW*KSVfl9(CQw+(p{4vJC6e}ckCFa@U_2s?Uyp& z;g=fNUdcP~rg#X-f9O3euGEEcX`^E!O?ohEvkbU8zd;>!>a%NRA&EnjE%|}`AQTJl z2&gPu!um4-U!t0yK*!zYK8+Pyq5pD;*YOE5*01ui+;q1SCO(Ho4A^;K@6FR}f``gT zEGdS9jf&>~nm3}K$43qxJG%(g_un0Uc1wxGQWIt~;YaWz&No)pWFT?=ppJR%w!gIk zOT}IqJ7v4W@M^Uwt3_Mr&*a&ovy6@x8&kdd4Pomi66-W`@R=wgUN|}Q#)}WP;BRPF z__}pDPO|=ZId+K-F9qG5FAi(>b3nUqUzGm{Kfd-pa*(|+0`jD@g$=HX!i2f7liEuf z?=3bpyjzprn@iWu+{D)Yfzc~4>P3B;x|}ZW0ja3+!O@zYr;qYy=+*R;CtZ6Z`C`)9 zC&TFOFKb*sGd(2+bBwdqAEjvH2BUwrX2gWg;-q^()y=C~I2-fx+ju#lr)n}kj!(W9 zz$xL69#Oe8ydV@w9}mC1%LHfleAaKSPlL|yg}IyS2^>Vba--*u;074kz1Ymsp%29- z3M=|-h`i3o2i}=qD`67#-hANGegaon=<+mVF;NQiti#>0IS2L5d7(vW!rnX8XY zKW~>KF+a*)+)$Sftw|f21v?B$tZ~_a=b8=i4gcGqsa_hs13LV(-d^Ru22q_ZkNABU zAkV75@$!9X5=-Gdxn&zzV610r$=HW>$lLaCt*_fAnDSQ^JSKig^Fbd?X7PTBM9Qws&7cwJ{7AEe`(!itu;wpH&8am4gQb?8b@=s2@UR$*&4 zCRwu2coxqR@;Enk`k_57zChv>eEbdG5^;)%EyDq6+e~pLOEGaBNg4ZYJ&f2@{DQP> z^Ra$vB7v(ZIof>d!qaY;ILn_gAaxMuxHcDN%(LQUI#b%) z5+M3U@V+|^LomkI%d&kVftQF>>S>o5)} zwYz-VaZs0R9HS?OEnoaI=eDvzOaD5vxnqa1>pk!?)e`>Us-OXGmoaf@cdu8kQTdEr z&tjW@$xmX}+|l$y4{zaMIWqZldcgSO!I-YWyEv|DyD=q=#wQIkoH~j}WmgB`rXEY{sb9otbsY~@ z(-ug2Au98|CA7C`#rErYR7-7e3 zW(!E#AqFp3HY6Ecpyh?(_}X+)kHiN!(Gh3f^*{I-D-*WwY5o$2m5-RV|M^JZ@W(z! z$ezp>fj*TxgHk(#aZKRzEytgK_wx%09Zz?5U~JNMi=4u1I5Dah$WwC`Cq65(hm{j? zg}Jv=pI9wv=R6W))8Q-TdO`TvvemY;jNlQfJ{^7E{@fho+CKTMA}0s+5zAMWt-OZg zjM1u_oODU-=b8-|pY0(rbA9@{>EZ?G-*$F~p!#3ExSsEDv;Dv88Vqb(ESGV8j3ZXL z54WBEP2%KNyR!4tAX2orPm$tyVC0fV`I^fNhv_-{nXf;g4pL%fWkPeQTk4dqpHg=G}LYqW}H3?->5O9d~Hd#9?FNY_{&G7742Z`ZFB1* zk)JRSd}o5QxB@D;J0@?-o(Ae^wVP)K#bM@T5x@M;?eb<;$+0-BcT3OSYNrPg){?R_ znJhTz^xJ4Q?-LGKL^!?;=Elq`o=3uTbx3SgY|dtT`Egi*@*scz7@fRSKIRzn+ehQS zhUrPKd%kg8*vX`p%X@=X*Ak}wKcxM4IF@l6KaSg@VT6iMQ5i*p(xCHhPc)ULWEDjs zA%%)G6lGMhsf@~usMFrE_uiX(-?nd`Kfd?%9FOBUKF9C({j1~Fecjh}y~p`J$LoB( zhM~8(*@D<_vlW}R#h1a8?|T;S`t%*hnSuSsrLB+>@{wBdF$ZS7+a^xt|MRmP?_5n_ zh(T*l@ORle*4TQrXlv`}JiiP3?$p2V(G0|f{U6#khi5_T#oR^rv#HqVcho>hQ*fUB zsL3mSU|8S)9o-j(AB(Eu=;VmZ?h{!!a#O21(!3SOHas1^z8Tm%SG?-mEEyRZ=B783 zRI%%w%7@SOgumInab{_8g)vNoIA<*8UpDW#7`L1*TSnZ%F=;!NTn9f))84F~@^NK9 zM-~0tug#&vxO6-+Q|m~3J@0uZH3LOiPfxKw4{az}qxg-$-F)%gbF(|49P>&e4whSv z0d4)ENr$w1NE7L{wDQ$PMsFRrl}iW)z^^OswyNQv-;=X`VV_~Fg7ZPz8p5v7uRNmL z+Hhib`NV;1^=p73@MtvFrWK0E9G$MQ*mVsO2T#J*{uxK?^4@G1psfj|wz1-S)%~$s zafoBhT5srQdB~qrFotrQFW~s7>x)ww4o^SMtTF7E`ZsI-v zFD|&k3UK{=0OzC)6)sH?c)&5qu&~rK!^FByE-gz41UR;OM*JA_8O~vHPhN z4yi$8Of#Ah^N z*mz)rN*w?Z6>s{vf5^%bmW5q48 z9Wbjz-tKeiK2WMMZ!|0W!&H>y(b_B<94-8D-P!yC;Ro?O8|+a23#>)V#7z&)fi2YOXO&KbH;8U@)zWS?I$!IRU@sdX+2kiF%C>T>E5Z#wy%&C(|_-^iXINBZVIWiTcV&54qRw+D(J5RIl`%!`bL{>AO*bkp|8!`nLQ;#~gGiKT0(VNr2H6p$D>B zuE2EU5s89*j|n@QZ!9wXoaoaaf368Bx=-Mv^G2A3p_L18{3+<>%=# zL3NPS{w-bv&b-IS=GTy`8Bhu;ZztGiVyEHzp$DTSNQqyV;8I55@S42Kj~~hAhoKk2 zzuMf{eM~UYzs+2Bi#+yD%4>C&?kDlspY?psNAQJY8hAJ3kD62JAxqRp~|?nCXUhb(-MgYoLYn-6YWnK!?qi>N2?;XnJ}D;@Cs z5W)M=KeyoV=JQ|hi>bB3hC)6V5Z{qMd7bFvAP2WvPk%WJv)_w+S9-DgvZ2x|QEmCI zD43$v*KxEs6Z^358C7HdAsovC$?!o*9OCfRSP@I`i;X#Nvkng!xo^ zAVX)6jIwmdn^_W}OC|W4DJB2F`M{5-?>94K4gp1>V^ex>CW&+S+d%KS z^H^}ZK6m+%2XCEnZ))ghR@T(?EYn-oHdfi@~xOQq#Zt1J_z|o2CM)}K(xRA zWC$T+-77nn!7o@^aq4ftwfjh&_{}G0Lh#WHp4W|C&1Zn*$K_{^ohS0EOh)tr8SS4y zu?sjCE`I>h)v4wiH%&nG**n7GZw)cWT73D$ilZd1lh@XAR+SO!BkJx;b2B~cOcvg< z&U-PmSzdPOjG*IqpI~@?kP)&LFDXv)BKXWGDF^qee)d5Ycc9TyiZ#%hmKLsGoCuj- z4}9B%lc38vaO)n=X6zKy|9+9r94ikjv}>4dz_DAC{D=NDLkCyzL#w{a*jzCC;mP6W z&=WrSI_9I7}L;sF^J$c;lIT>ATl>F2E+PYKguR z|1+O}1v#%{Vuy(JH9g?ryDzes#AWdEiMZ6ic;?>&P98B@2lH~OLCEAoptgDXXm`<9?hx_G4z?I+k{=!PTLM`N%)`7SNKd+{q>1=(8*gn z(SPDU<|&U6rKG$UhJ(GX(s`z!D`19hyRHg~F20)4j6I4&bCI1xq(@lox>brfI*Q+` zSS;p20{7Fe+kbuF_aup9bz*VxWfi2y`Zy&%{DzDrHDbMPN07E!kvGzX&;z3NX5QEk z`v@jT`+VGO&SF&&^-}pYYotH-;gvS}1m(;51y+ptK*Rcwc>eFpv3d5G*|JWe|A9`5 z(r&xX=GOz-t$jkuBb=r*K9v$#V_1vcfpWnUHyhAK@!iw zGrfCeJD{RGe>gnuC>GTJ+Ol&h2|Fgthcs2cVEtVV4zr5S`1#vR>NY2K{Q@379Q&+p zNAyXJrj)zwRGEgq%1_;VT_=d~d!~D^d66$NUM(-EFVBIgyB7+h61D>+P{{h|t=*9P z!yr0fGr`9+^mFQ}tlkuK%mh1%y?a37du2_r;vxF{sk?vQI(@euM;?FrGkTyNI_#6d zLfjc&9KO%GHXQ(!>aX`JYLr8#(GMk~s}@dgADh_0eQEw5_028 zft6rJTU+oKnPCRSM|Cxk<+>;D?0dq$Pw7xd-EBCI`B^(R8TFD+6;Pjnn*iD6y(>J2)<{@8Pgskc$x6Fy#fx>@9B8i~{M z1}Rd02uD{nyqSGX;BTj%)CtOzaFDpetCswn@xX!-e?AL7DIm+-S{S%?19Tl5H4HLg z^9e%phbN8Oq(qS+?MmUeMic$guh1(q1#iDZmr;ZNdPB`59Dp%=nsQC zUe&x`Qinb1p&3GF|3LLf<>hq)Y(5_Bcy#OXkodoSd#>J+)*^~=&jzN}fB1up%feqn zhn53vdW=_7vkjU`CSzh#MS@Pn0 zdQ&TbTO8U^o37AOh<<6JlFTwG>~n45DvWT1a*?&VVHI5((FvK$`7YYp(8jy|t#gMo4zDQ_cw-9)iyXtuNv}-6ZkEM>yyAv+MiNGF~_FY8f4K z58cQZ{CW-AT(kyT)`~-m_|@L*j%lE@FtRk`I-%}KLw>{@(GN3pFYNc_9aNZpI$#&2 zXbfThCu9<;D1?S8MZ=b^_-y$y#!)1o*L~$g42T<1JRy@FgsI zGX>+rUU#<*n?t!^9p(GzP8^zCy{0Sl2Qot5+zMwAeHs}7qsohS4a1CXB5(5sqQ7}s zT2VHCK`B1_>_rMSB=*5vOWu`op&TrXrJ3!J&Vi!TFFhs63>+D1u{;(>;8`Pz$MJU;sB zujwxkE4}J=E}VgOrK;GnJRwY16EDD@?_sWiqvU;l2VoC17Hhio31fLjr;^{8|W|d0dg& z1x-kEe^cV z&M(GrsJ8B}p)H{oSnANFGH@^&Yr7vG4~ZVc7Ii7*w~7DG8{Uh1W+FNXd%0R@iM!GO zuq^ybLN=0-KDzJ^mjh8pph+Bmo0`t%af0&G%&8tdoIO(Z{d!LKoy0NT z2inG3Y+ePRM^E7y+wYkBlDgdOA>mi{Ts%4|n>B)Y33a7RWb@`AeM(1vw>8^-Lyp7j z$3NbNz@oj&d8Yaw@wChG7dNl8VTfS-&l5A!+Hafe1 z1{-@*-(G!Wf+P1`52ar(#||S=Lq4A^|NVQ!I;b>$6S9iHu|@FD-49>1eBSpGmLwb! zp^afx!|Kf&@;$Imq~&_(-}5k7OF915ozO>P4n0$GUpi0k4hLger9-#R!`VP=;GPLv zB0mMK&m^yV8}{PpSB~KjsSChb73^W1X9O*G*8cwZ`7_RRYRyK68DaNsW7@N&79{RR zD>Xw)&%v0kU$5Ph4w%|dIy0A~iW$qRLw_a_dL6?{KO3Fgegw(4&M)}E^8|YGQ=DDw zc!9Yk#KT$cH`F#B+L4($j=g(bm+fD=2AiXITJ7;=_YGq+qipd&O|=kMse>u)QmDfo^DYfY$g4 zCJ*SREdNjso*@+ zK_$syfzYPeHPZi?x6ZBap3TMnY)?al{zF)iv{xa|hRutEL(JONkj$Iddzn60>3Ir! zmp?seth^M5A1yllcw8JaEV%@cx(2$p^v+U}@57Y8W}ltZ3KGY`@H;v0`jGtT;Eb<5 z6+VeArYI%afqGqSEK zg|5Bl1idqfD_$!8g(-z2>03N+!uV>1gJ(y*a5VUuQ(e~)Yzv}wuNH|$3cc-bc=Qg8 z`zDg|be8~74smhrSu;P6-^1-n+0e}9NxH?;y7Uc-mU*VZhJcA0+OJVxmX*CSddd z&-8YWWD>WB_=Rf5U1;Ym@gC9|BXRWpC|P)J3aU27CU3dG%$aa-HTa)-yzUU+tQYTiycGPD1{U4`D~BX^{&n!#a=^ z`r^l7$K}vz-r*HumI8s78P9}e9fC$0KoI%HHPMFnJ6{4~XBcB+pyt zFzs`^@oXjWyc4JUQ$8D4VzP0d*v2@b|D&g_^2-zZpHRNyrS#SuLT9Tz{=L>(rSr&~ z@fsF*_XDyYUgJL^QV7{CCwBedaDwregH0I`(m+4LbXlNy3i~cde=T}VBIe6K%;t{n zFj6w=?J_jXaBy)(!*|O~Nd3Zju5cZ3{?mFb-}f1^>!Z*swQ<~OY#z@GO#L{%rhS1K ziT7vxnSyI`o+~Oma=Rn2Ka=@8dxO>!D-h z17%Ux7>WCIvFd4u6r}u?G}pUE)a|SK2z`Q~k8y5_w|jByY^;^} z;T$BFohn=ZS`Vfd*|h2DdSmU$p@&~gJ&^LDr1kZH7E&Vjngzb}2fOm}{l;PrIC49L zbB&%qj+5;51~~}5qV6~8s;$~YA0<^c@$A|}V;uOapt@({IQ#yLg>Bp?Gui7DDZV;o z+$Oy+>`gYeV*Ww;?b-d`TRni*UUT((dIV5LY9p+lKOx4|)0$K3Whzj4DFG)lEMaQZ zOQSE!iP+zKlt*`BIrOX!dEKgV9D5|Xnob7ud=m)KLASXXf42)0**!K@q& zrliTH>g(A3c33guoV%=sz$MS^*tX!zel{)yDEGx4{5bm`yeMP8?s!YC0+#ao<*9xI z&Yn&+J}>`<{oXKh_1xEm%gm6x`?$|y*GE{JmZ@}>(-|4=JG`x(>!D?a!P5{%@CA%s z+V_L=6k(qjwkD;;Q&KRqD6Z4QOc%S$ix&N2RiblmMsN136l669xVkNUhw=6)dsz4D z@vmv5z{{Bn(0BNi%;|IOI3BZdnHEnrFbe_MyazeyeNB2_3oArB$me^5ikC zOWVx-&M*w#xMCNjIRM2G=L<{gj$`xTB{rFkPY`|iwel9s!%;$EeO&9VTW6riy}$l+ zR0T4|n^pG6ZH5$ykz&Co(#X{0m-*zv=6T1Aq5Ubl+_zw+!yUdifihS*5P^*!sL

    nhd3=%Bzfw8iXTouuNll(KJw$vFGW!cb7{3_jSbnwc(OE^hON& zuCCew47MLl;&@|fo5QspS=+kZCBPYIFK-;E7%9Z4I^U3ec0?U!&>&I3>-iO+9~CNE zy!#In85Qi+R^N{!@)s|geI(+f9~ErEvC*;VhKzp_lC@Tt=P469C;e^r20rJr`;wumxMG>sTLMQlIy`pu(Q09I z+3w(?c*hx7KU&88nd8{U&7oHqqK>88WpBx7&g0RA+$%W|rsAJ*(D8uR--K=0VkU4T zamp9zg$oNRqS$^mmYuHoB>t0$b$`P2JsKFuaLTO|HqJIXKZDTSx{knBz>c~8PZCw+L~1w zVf&{wtoBN_9tDXfB15TLApqMpTQu3e3CD>b7yjHoOR+xpi@9@6*^u;$}9 zd|5UsLY?4E?z%*ZesN_J%&cO)HoLeE3(|eWmBtVMU+>fVKyox_Ig+>@n>=>gM(_$z zMY8nmk1v4%xv>D!CLQbu+v@7zsxtSS;)p!-6nW7C zkAz_;KU~xqR&*W5oO`nP!U()TO|x*0-RLRk_AA@U=VS+sDUx!t8+SlUVETmbh9ltL zqkEpOq!_6-qYo8==HZDUSRlpnz*`RX=Q>lb&r?qcoY5#n&ieOgHq5v#H5z>p34=~C zrAdOjVCIv?CK<2oIQlnl_o8*Dpyp7%($&3cz_{SApxi~&@5$sQsV({!VJ7bTCHp!y z9vX-CFp9*Lm%+g4YTLC>2psY_Pj{hI;=KN2VqMmFgtYE2#;QbP@vYTT5Vc0_Y5Z1I zEV|vxd!*z8@!U7n8tQivdV9?uoj#>J_z2WXO?A0nYIESxnTC6BUlB4ZqhW@kNHO(Jxnk8c$Pup;}+)=9*?s zE;xZLF;|uL6cPLtjgLvf$Nibm^iVs^LfHWu4aD}gv`}$cq4dL}kPsZnJ!tuhLGX@r z9m>`&{`3JUX^#!QD#b#z-o@#9VS?AFH=g6{M#%{*>vdo1cA*K#7G>cSi})>#f3Y^vY<{- z(folFP0De^B8?;R&_vF(4z)k_X?!l?D0;0qyp zfqX5h4!FsR$M{#_6 zNedi+}kmGf0$f; zfcgBl1a?yk&R%f5htw?n%REN!VCrUQlgyRd5LJ4N;_g`hqgONfe{3~_j$ivj-!rRm zFjCl5_Rw1(Klk*Mi6hoYeb;6EeR~{%^8C)<=RlnQxx;I;l%MhfcZLNB;gn zT2<;q zJnC)GCj6Lo^R&Ra9#LeP@+O~%KL6kIg(c}DOHLILc>NB&DhKyx{UpvS!g1AT1+@_) z)KiD&^%0PGJA-l_W^w`j=dHU}x7R{btK{2}cQV*+5zpy$vylD!4f#I~-pSng-`5l4 z`b*E`np`3d9M!tAiRMh=SkZL*7ymJ!yltYZ7|TLW!ukI9C$59T{&PA3fMrAs{6tvRwsWgMfn*7`rsRX`*HpleY+#88Khb50MKbZiNzPBv? z51DPJ9$W2m6-JGQ37JrJgFxn5U7Zc~1mtKJ5IiqR9uQjp7;n&PzMFv%D zq1WfF+wY74)lTv?({;y$x@U|8cWW-)f!ZTabrttZ^)B199R zTt4#oq;C(lTuR*YL-Zh2Z2Gf1l(r75)9WKgtcW^pZ|sG8>%X}n`D<#ICshKwW(Bey zB_G57%{PR_oe8@-GX7-Kv)fGUaErKFn*R|f{3b$O&&{Ci$BIRz4)svS>-#G{f$h&> zR>oGxF_I)S&%De~+Ubw<-=~WQs;^_OU@y%^jmH_l~23y;YcIaWg-oRMJQ?=%rxAABeS3mdGo8elx&2!7F)EnWt$m=FYk-3@ANSl_z7?rc zHYN8@5%EPsP(}UL_ZdL#^}Ceqq>cj<`jkdRH)Nbh{Mf}Ha7xKv>XygJ6MewlH_!35 z%_+fT>t(OeRq{apl)8L@+by8^-79V%ZY0iyg}&qT9UqY8)44Eigt(qn!M%Om`z_e` z(U)7}>-zuRC;#m2uPN_$A;@gd<^4K20<@9Wx&~4V>`46iV9PQUXkxq_lFrWu=GAh$ zk}@U~)JoVLz5Wub$_wo}3^*{@jw>OuSQg7`|B7EXV6!%9}3Y>#4IVw4p_tc<$`|>W9#TYVF)T+;~*aj1_ z)|YSG-UCsBqt4%3&G6m5Et2{LwIr@6U4Dm;A|$Sj9xLM4iv9Qa5b-H1-L*t_aGsqZ zTemLZD`z4#?x)AqpM*ck(ya_)IrzbJb-LD=7QuH|AzWzxV8Ia}e>UBHBc6>DC2*;rYVaQUsgj zbR!#*u1QTKn-F|%m!R^wSm>{nRrb$X{yIk{J&Zv?kuy*$%ySFmX^B`4mv*lyx%Zl5@Ir$S<-P2cX=&P`5 zxWT|s{{xAqX?)h%`W8;ES;t{-whuZrr5pWuRwI3jI9Y#+LE>uaIvO-#i-U4DK5qUt zFtO@f&jSZtoM8(8@OtqV8xCXd*t&MepV8LQdFuqsBjI3UJc(0#7QQreVCNs8uXfs= zXYB!OTZ@~Sqz725A?ft<6$g$bNN}hAB5+PU2RdqR&iUcs`1{!LF#?yt3dpd!tJep) z&CU-+6^`NH(v^p$IWJ@3s`nnJ_h{oZhhL|yS37pTe|{x=7ZrN_nlC1}DzL8~x_zL! z>aql+P@XS;^;;Vl96Pfuy`_N>mb&Y)%xa+B%-mAdbp{f)o!Q+Kv=|5GTCZ=-xrVf` z8&5lFCxIzZS}Jts&VR2Tn3_D5XRA)=%?|C-Uj6CaN9YUJ+wFR%lKuBdF%REki^nl5 zux_ioekn}2u1-jo`GgZ^MwePz5PIKkGlxH{+Or6&GW1WD?;-H)bzU>#-$F{T>fxMq zt$V`1I6jE0vx4nULx1RSpI!*NPKv=HDmVDb4*+G%55@hekkPi*{xfe3SPG>^>F?VP z^$TCQu{b56e_68luSkOTrN{mC?r-*XIJDE-PvjbbX9(D%wCdfg5wy5W$>i)8APJ_F z8OuJ8LrMgSv3989p!Su%w!kA$y|AL{X0$Hk=rnO!ZzA{^Sts7LYjFmUcp}c(UU}32 zRrZo%dCWl|=N+2f`0720Gs;Tf&4Z)ZB^P&l>$Ukj1Cw{g+*J3tBcpVO@WPYXKuK&! zF8+N8$F~-K4&S#H`Z-&F5B+2T^ZXx*r;kaCHr{S4Vd)^#FR(_!#vjMq-FCfC zxs0v%l#5$>u41d?b=|FJe<5o@>}v=5Q4*iwKz{p}8jf=KwLgx1Ps~fxqTFOVc3%_p z)|UMK#{30c9!92WHwk`-se`xrY>o=UpvASwBwjL5@At13`Mekzn+^_gpNz+z$g6_K ze2?MxEa|MR!+ba>CM$mU$}VK~>^-x&CJ?Ei0!NgSld%6&E&uZ`)mWPRO2O&4Dij=E zVp~vJ0`;~IoNHDP{2)}(YI}|D`TCDkCB2dLjp{J4Zj-X8K_!Wk@9w0>^wxRbqrJ%4 z^ShNew>qu5uAZb3c?nK|^HU$}uVahqIw!Lm-(l*x?cvrh3P2fc$Xx6B6dUwEOE*ky zfT;#}DyC2Dr`pa?lBeZi7$Up1JO8&NSDV6%pAGZTd!Y6YKd_ zek-n0b)S?7`t>ds2l0&SHMxGU4YBw_~lN z8{eVE<#^-p+3n_*;aKrpb^SUi!awO`-ETVQU=0leACEBmQ;;l{qG}ay0{M|j&$ZI_ z!E|1h3XL{jx1FW)L- z%$%{C$4kV1@2T0*w*%J~VwRhUjKDjhAG1oUch#0+8)&wFS#Rx}jPxT9 z`SU93e)xMV&VTRCA67R+yv(UZlyeNtPQSWKX z%|#^62jyCOyMDuX@TjEiP(Shi?Y9-eMEvHBJ1yi4caAMlZx~4Ty8UBIFLnio%H{ne z;;0_Zuro5E$D#H%U$ATKIEmXRIO5wYUu0YmJID8c@R$2(E#7Y)N<+=l1eLrWguhbv znpvzE^c$PL%|5Ls&9fgsd$4a`{n}O7Ru|eGc7otX=k3fVm85UPo_@h9?!e{n`_I6F zy6ga$2zU569+wC$_uZDP-YEc)nPq%ThhP%dLdJ`VmSr&hG4$8S@;c5TH2iuxhg zHDF5JaCQ?`j$51dMc>43Qd57yTM~{i3z|CgXORB-Rb|uqnT6V3IL+fAV5Vsa-UP>0{C3Veh;}N#QjQn!Tt&>s zFodfO&d$RzvVUhz1USFxC}V$RkS(Wz0Q@E^UK`U$ga(%`ve%9)SaG5->)>W@q-ke% zWpSwE&_J$5++ua;UVG^4wW^I796@1^&OF?VKPTnXOE(B&r>dv^jj&I^Jaa#UMcU22 z&seL*6G=~Y-xM||1T%O!1)zn$+t;8&4S!>}^2JtBnBu+h&ZV3k7a(KmgYC#V0;kc> znrR;RTLZnEh1O1_MI?^<`pPFhDZ{`a%g%|-$-t~WA{(?)6${RvwRz;v-p|-*u&eCY z=oAc6I+v*IQNf{239Vw!tl7U?cqv5C?{PhLVDZ}})$Hql+Q@CAJ*0><(FVoPly)q< zB;0*6pW~mMxwa*6GodqFwYTqBq~9q1**9mp`WG3RHG7sTSS%%R(*#7v>(1f$*@q*d z%Itn2>?oR(`R#t0{khsl#MW>n(SO&lL`}fUU=r(JPq%G++5yaur_}kszJjiD6JGNb z21s)c{Vr-t;MXfYafmFoR7cvPlo>~rFF3xKX_-3QgR>ttc%9y&2#m$ejCR>dY-f=l z+wWe9)FL0@kP|Dh?dnR0$$%j2HdQEWdZ2<6xuP4l)DU&Tz9R*~8qdOr{ZKE{JAIR| z*F8F(G6!}M^{XEGeoEF0f`7dCGvz#Q`41Rx*z{1ccmZiadXdhjlnE%muz96%*J+?} zuGri7rx6B@7|8A03(ztCLjP!b7Y?bVz2VgsCUO0a4O;ZX7Ap7d-Q8jlj`?LKllHL> zfqHqVi0#gANH0BRAHJ;*$r{Iok1kvR)ATIvVZ)s`^?vyi1$r4yUE5R5XGVu!p1HP{ z{zBNJbaN>C=VzRnUe}wbKMoT+);NS~|Ar=;yWR1>_hRLFjnikoPvWFxW^Q!hR;<}{ zU*lnE9Soe1Ua#^^5<8sYPkBCg43U9ydPVMB(82xu-qG9rh{h}zUW0tBe#*G}^N|!r zr^+Y1wNZtfXHTHkawkwySbh=)?0O?IHC+~0&T^6j(kJCMt`EVS$z<&&uSy`tEfo>E z6@bGfUGk|Kn!vp5xb)m%F`O=A)>x6M;B#hbfMDi7KU@7fzoHD^0B2|LlgG^)8%|U9j3S)Cj6{cd7-@%yC5WtMegu0=L~@I_==s z(2LXur&PEcF90?1(+YH)$D4?qpE6J0zEp~FKJ<>IiV{fnx#ef0AcRdva@U&gI0ti% z^oNgM2*F@mP}=M2ZkRQ;a3k%xhSa$CU;0O3ZLGLBc!i?m^jb~52a6)Op4sRLV zd3LmoLr}YIJCV1aI&d@jRpUH=0%@G$EV`Kv4DO;I%iNm*qn=i36UzE9^|I{5PaOhR zJ9cryC*Pw6FqUQ1R1}g1b*}NJKY48fiaFPnk+(uHb!36S4QXOt$R7er0RYM3FCVTMbDZKSgsc38p}H%q7Z8 za6^XuwJBV;0BZfas_*KjLDe5I(`~QBk>c!gw2xB`=B!4#XUUF)J>R_a*Q)b>aO}X3 zt%>ySIJA0`=Vwh_ppcIZ{ixSM^6&lI@1JOc-oV`*8Xezp>}AH|1qnnQxu4Nq<9KKb zF!$9t?|q^R$)_{j|C%hp%=IA~5?^UzqL;t$0>%lLup!5H^WH^9^3EV7%@ZV!??)XE z#wB2Ki2dpa*?7p2Uj4pP_6N{}^Cw3!zDd3~nXx?khVqDdp9csKfFb+U0_ zB#xKwJvleaB8w`X)mpW5-uf7~y?j%HLDb{BS@x$Sujml-^ke_m(;|C;VRqs54eNjL z_36Fu5<`gdfpO0=G}GW3Hp&Wf$Wcx~M;{*jCK-w&Wn3vL(S7RVl+G85DLp-XSA{I~)e{|tS< z$&ve43d7jeI31_2lGpabx0<_7Xx1#?+Id(MVrL(zI1ys)xw+wJSv^WyKc!}+fmb@#M2 zN4YEx-@n&q`opYejJc;6yLi)o-;n?CVC@iOsJpzuKOlRixb0edwyEi zS0K4yzr;plW~zq1pF9cEE1rvl>}tne;UO;0Y@)Bi$MYlO%l3m1_5R*?5 zHin+8*ohT?7Gn0PbwxYMj{vLatWmqm8mOHK_e{?@LE>qDdH;<_BX%24sW@I-htzee z*IuaEFz-3&56;%x>q_CEMdJFa<^NINQL4(A4QKb4Vwzl;%EY35F#V|8IX8%14+U!R zmA86MYhmjA&E!tqfA-zS?M_$T1(+%{&U{?%3FOo7jtu|)0OWQ1?$ze~gq(1zO^Usr zaj5m0?T;5p*uPI(^3iYyOm26z8Z2VqxS3~)fA2x~W!^F)QgsDs+4tx@3+eOLVaw*9 zswR7%0`qM^zf^KPeo0tucVH>O6Ec_ev2OJ@73>I689)E)81#2%*6g-##u(mg%AO*k zPhvc5tu$q8wLRVd#h*2-s5H)aE1DoJS5<7M;WDr9cdT}FxD57|{ZuN*8tLqZMci9B0Otg9=(uHEz^ zc;VZ4C+zzhVvw=B_{p@~wt3??{d!WU+Q9=EA&MUnu8_Bim>boDLi zd=AtE^~P69Ux8X^<+~=23X}au^>a*Bi2ZdnrFMNKn-?3YVq5P;3j792hGUZHaZ%{^ z{_*~FDzQJOyo(egA6VkpaDV_lD!>j8MX!P`CT6*s=E-Ig{W*2k*3Gf<1i#SGmfyuY zM%Z;6=*sADR(nb04<=8`ikB~ZOXAwav+I+$K2rNzcOUAu!(QXEcrU7**aZi`w+kf$#aKz5bvsh-ux$nix!!UY9Z?~tM5;hfmeBN+@;L)h}F@GoD zL4&vm{f8k-RbbBL?%=L{$&g8@h$++h1@v8-)_kEUlu-pKL5p;}Qv`MNH#uoR| zw>P)+!mzWvi*0urGWSX}iYoR2O-98`>$5LV_x&2oU=sCt+Pb$s$F#VC)sr54wL}8P zEjNAi{&EcHkGD+PEZaficyd$N_Xa}EV)w>0Jqef!nU+xkH<(pzIe6E#2B`-Ydv_hM zCvi=lZ|pPm!u}f%t9yH3;y|^ z^ULA=zB6W*4Hle2?LEzG-9PwSx~!`I;4LCAz`eqXZotmZV!=rcw{P6{VRRd>>z3X1 z^X7|ra@RYBT5X&uT3mMYcQDZTc|JPrDE}FNZzbG@6FC%|Z0=xtBd(mSD}2 zr$eyG9?FyzznZGC`~FDWS-CXGD<|TMuL&R-Qi^T0se*eOV_^3AsVd_Kt4X|@s;=Mb z`41gyE|QwMu@R1qh3}{Mu=~THBRWW&%i0m>-d*rEvIFQLZ@73R$=LV&8`bUkv44Jk zJ*hL|3JiOsHH@1PdO-u{?`W)krwp}{HONT1L*nVw{}H`A0>}fFgyPJ)h(GKs5a<#h?zSBljY7EcpL5mdFyW4C$nZGU*+)k+e*|AnePIu z?v@h$RJ83wo_=W?U?^TkxNr@@BU4sJEw;M;3hFvXmQImHATDymFT6qtM&h!wy@KY~ zai;zM2v#~z;M{Fxp4dPW==)If6PyL1*?c{z{aQ9mj#++bmTiEsa+9a&b6GGrz#Xw^ z;VKgUhD%q=WX!N{?T1yBAy;79U%9JMsvX;x{nkA@Jc~7FN8hGp*Fo>u>|HTh7qL(N zR@Z32A($=NuCcs^m&E%kri_(x0h(+L@{`_qL){XoW`|N^?0WI#{u7DM$Q*nVCOBx0 zmGQ`>q)6zljw?QP^HIr$ugWQ-9`DJ}_4+|{yYwiKu|_ky;|fgX>hBSIbrCu!PVLp= zvB>C)jQ6`w@B?&*+j38>I|q3;AxU05hQt@D5s=*?jfFyCo82Z&a5y%@xhI#3!+u*h zt?lxmF??T2+ILeDuV`SrM{yC>o!#jcdF~^2wN%#~e7OhWE!H0QxFP{8uC7_3u7@yA zzS%84{?C7xJpA*%OgGYZz1UuvnTrD}%2(fZd50sf7pH70`vY|UMLbEBn$WXh(P}k! zUudjO2oO^3m->>-^KdN`ygGkC~#3@J@LyXcp`kvO8T2oC;+F~QTR zRH?^Uk$I<@fAAMfzMV4PVzeE5PQ3c#`ND?8{Yv0Ef5jg768p>4Cwt|*`JY;+y!p2J z0QB;@i2FyTLcoQYg0y-==m_I*TynkwYBfv~diFNrz_7RBHT8e<_$yKN#)DlLw=A*l zkL6ix&o`u>XRLtUqvlIvuH?hCn@0pc>lB8#m#}2)PGkEI4{C|l3TSM(qx<6bGM%y@=|}$3gac8WA-V(CUNU|+KAhGKfvA=buq?X-WLE;^7G>_k;4Wk!mfdWgm12bU# z$*gcTFF4lgcXxZfQo+$+hwpP^`Y^i5%>|A}B5f`0=E({rr1-2dh??OdasK%$Q`>bD z#v2QkZu24f=LhST-Cdz!PZBs%r^8zR0Q-9`r6VulKW41lDWbqR4)m#`UMKv%5a-0u zspHbf3$v%hpV{81gb5DndS#6ytkJh@ec{ONABFMHN%TB{?NCsb=f>aX(# z!f!_}eueP2K;7wNy31r0i6=@{ql|A9;x-s-=F68%zs`#KBV)zT)tCS_t`{bx zdpSnNS7N)U>4v>82)~snok||QOz1QX)XxQM`5=ec*F1D%wi5kJ6^;R7fpP?{hvryn z6HCp6amGGX`TOj6lEmj*on!ev5j*R(-gsYQA#10W{DSA7urBQoZ{Fm8^a%=&71hk6dSE_^|Q@7bq{3RQf7XMXVPm|E^v;A}_}B$vQ3a zYJR-n`RD_m;vr?24X6Kn>0$COp0sjygk3mjl{*#{#MY6A@|AF0)!_~lZ#mldgR>hb z)Muunby?8UlcI^2)9z+LTd`yA#hEkhd!hH;WV~ALHe~*23%#37%->ksr}5&C z>~jXFQeSKBa*6pFv4@CC+gqR%C>RzB+VDt#RA z_3P61RmbL?E3WzW)&LE5y;ZKi4Rw-vyH;C-1EX50>gLBjAd9PSxM+6;Xk{^n9;R$Y z^e(-ZCNVN^d@GgJt))5sgV$&JYQRfl_W6cWbJVu*58q&R(C#HkAe+ReUUnv-#~C}U zyBFHt;`?_lzR35^jmF7Px!zm7XOVi&>DCRqHfUCUC$U(*7Kc;(tky0v##RZzszjO% zGFSeV(4g{RJKP;`=_Po48}~}0m)-(w)_bzy21@~^j_4f7G*N->=AW^OduOoF`{2aW zmSAj<|8)QTuNyGIpRh|N+70LtIR#ca4lo$-a&`Ew%Ot)}CngIy79ce|@6$yWS75Cp zCsoEZV*7XPo1X&ZfuU}nk}&=sd?FotkMT*Jf?sJ*JYSfnB1P(xy32|8K>p2hN)+^#C>52X^jFeL4lS8G0Fs3thMUQ_IPL9Y&nw))Yh75i>XeSPp0sNdRl z9N1xmlozKY|89N(RUa}0^y`RtZhYS*5$%?A7$d1|p+s^*r%@$YVnGH>JxCS~xGjo< z_9a2@GxeZ)>iZYRcp{I;qAzJvF=Y35;)s5mej&LF3(fcWS#fsb@akrbPpv{od8s*G zev+vFQ7x{fb(I~*s;CL)h9W{|iE(=6+9Jygm|nwGwS}(|##Fup{#MI^fiaU<>6JqN zJ?>bq7=Qfl2@7CeuAGBqhCu%<>we%%E3v-MkDNa3z8~q*o0aOf!~iYh=ouRlVHfD% zYgU>lrs8Pg`7>X*rm^2ZI*sMWj%#4HEAXr}SqsZZjkm+*xRJ@aQu2411Jp%-&nz_m z1vBAE+q>R|Vpo&h_r~teIH|R7S6&w(-d$5j7-d#dhUva)Nf>bH=k@KSEFQolD#7-ownbm%ZGXmw=_5 zq+mcBMT*s^Y@cf!Hifc!I6o?oI2dnB?c3FGB3MW#O7-Wzb+SQM>}orSGkDi*?U`>d zxaDH9&VgW*d(Eq)6;Ql{J;>WOSRQ6ej$|OvG4>>|X zLuj8AKi2Z%5HN$kiU$c!VuPU28mBxWj-1dKDrsM}5(c9h6G&Mhkf-&tPtr6U8?I>n z7B!xyKMxbB$6kgNzK01(!F5+Z^&U*9EcS^> zL7S|=X0xGA?A&ae-mQL?{k+t8le%SR*gPIU)_dt~(97mA#rD|n(DIt_fA;cM#-1uw zU^!*kb=?qxj#Z*ZyiVQ1xG!1t)ZBwubXtk$%(-x+{C*-*w8xUfW3$g1SN?z@(u!uo z9l0?1ZWj!{jD#7LG|Lj->rkUc=aqMF1?s6C-_OBOBp>QkEjlPm;;xIOEfg|k*gO~vx#rC)^);Pwa`#5rW4c5? z*i3oZX7AZ{tbQcJz0U3-7HhaN zK1Coyup)^o_93KC_-*QbN%Xys8K?VnE>!`#-9c+bSw3jdi{X|`^o5?)ADt9wU!dxn z<;8VGof~F9{2Y@Vmcx3R=!7li1TWaf&t-hO=d6+5v-KL?c@jq6yVr80vh$xfC7qCa z+PN18_1~6wQ%x@-YnY@>R@pn;F- zVJi4QvW}@OOio>nv3e;@mDOcOb|a<`&4SziCS&ar)g*QQ0{G{X=&pKkp&(C3BUh zNc^ovm{+z{q9^b$Ol2LvEh50qmtsxBA0bv{JIrYv{Y5=cj-$0VNX67SY}dFR;qXV3 z#Npr1Y2xt^;t!WiW|cKV-@`fj?-SWTfA!jO`1fC6P8Mor-wektA-id=hMy!Jm&>bo zhnGP+W8=@ek+s-a9VXzqUJX0fuHyM!OyKo7-bVD>*W|%?-e}aVyFYQ@Zjn#C+$8jV z*suDgaX-+>3I=(ncKx$Mj~|sQPvVH$tX?$VVxalHI$BVD2}Tr;?YZ^k|I~3c-TaO$ z#(|k&O|xrt|K@2OSC+(Hm?Ag0?LPbw=>~~g_g~_`7Ckx2tYq!H^-U>mDa-ve1)Vvu zR^L$ysiQj705Re`C1=}xUXs`XbPdlU(o<6?+1?zIbL%el-@TmA`GlPZhRQW+i%Hs{ zz5tn~=oubkPYJp?0iuX_!`hLGa!erF3~7(EMk&Sh)pXHxRfX8jq5t zxo0a@l#*91Sj>KZU>tUT6(yjGm1iMZO?an2RQg_raM$G@aPuXu0?|#KAzdLrp#FITw2aW+TFPvW=J|`96 zkf!&NTiKp4)FynWn2W##<@Wv%muIYm&IMH}FC>UM0=enF^$(o@pqz0o;Wm>)#`B@h zkbZ(MU{Xbt*ZoE_PFP(4Wkp#m9Q!gD_TVOx|Gv6u5s(F?cla3*>)CxcB%VVnKDLl| zLMP8P^{vOmu_110NADgLVja6x$~?T*fQ++}@*f}W#oD=58^^b>=B=~7b&SsK^7DAB z=J`_tlid*k<1;Y(YbWRN4MZNXGcjmalyw=hoSr)RU2q}d#C1ZW4X4)4yI<|OhL2HP zY(9Pxm*CX-^IjyJvX1=psq7C9MVN4{>m=%6OwwY?h2xKbyeMR!pH~S~4!9iGxlJSS zU#mKwD6|g;3(h|4H(w4z)>SsaQbs@teiH51Bn5x-50*40e1wTR$Llv7Sqe?-ckFv1 z(F#4weh#UK(4hiu)r}4E{W~YtotnN=f}JNf&22g`h2_CtBOBvAfi7}b_~=d7jO;D2{|4bQ}AO3tk>LLf-eS%rlRjog75NvwVOsN3+v3F!H?IM*q5z8*Ct zGVxOA0D)^A7(A|I^;D5qX9;?bM|RB9Sz=$; zXX^KbM&ZNZER!X{bHl`a3T~dt@D4$$ubGa4>Tc-jS?;}8faoKeIPT87=usEkuWmMw zIQb5Wm+#r?`|d6<3R5#~@Ox6qktU%=D6(=iYvvj+Cb{J z^WC37{Z;c#Ugi-_-4N1Ik4}enE8ecw{nw!S>+iw~S300)-IqCm50(Gy*sow2yEd21F2A!B+ZcKZ4h^fL~}I+@T&G(k;4Nks<3S?_gI~5 zCyAp*Ecct=PoSCazbO3RI1JtK7@di8%qnA(CJ@F8Fy?uX)aK|Yk~X`o+N5@oPc z1M17)ywWu=oM#8B-gSKbdy(h^ZrU5j`_(xQn^(o*77kl|ygv1l|&dm-7DIG+e zZ<42>VY_${HY`<58|5bYg2|VH=JvL~f#Dic1>rPXm>9nITCQOdGH3F|(_RyG16q%= zhlm9`UxY*3ejj#|BzQ)t<5wfB7LLJG$e91cvr~}l@kZVKeJ9XFH3HWpvGp}coMQ_k z#_kaL^C@HPp`JP-Uh5Ao=ABsm0Q!Vp-!T>c$i82KYoGw{DlViZgar-!C6l;)H>eJ7 zWcOJ>{Vmc;E)G?s>^UDjXp@P(i{+n$!(pV$TkjkCTnQBy25!X!ts%}0vcHIP))pMA ztC;@bE`S|8(a|RRi!te=iG98rna(A>OOyPx|V ziCfQ{d?#xZ=|ZdPyo`UsOoGtj*&PdzQYxc$C^#Q!TpvHbD9pr(r>(|k)+JDc%rj-IJ;(5(Lz8~%oy4wMoO}g?p9lVKk zORl|qNjVZa-Mgp9X#&W4*m(afmcWQ*=`FL8DPq35Hz_=NAc>9h^Wqe^uvT`=z1EC+qV7V#m-^xXd`{NbWwr74}t?*;s zl5jF~E%h_xC?NFsa!a*M_0}FCu9tO72$-&e%HG~Bmf8i7V{yp#+W*lrSoCM`z^56Q zkPv;R(>aJmnR>fx$wdCN(9-g{CpW>5kkc7s7x~E*$A4Uss@$Qt8 zM|*1b?k6_}#vZCsSAK#$ce&!1djDH*oX1;tKX{KJ9Dfhr_%knW^Y7eRE-B36P3*52 zN>@(D8|c3vG_b?&2Gl=$UE1W7N#a_o(&?#n5vGo+xqoqB*ZE;=^P?+JlMTKR%axX_ zI0duYE}1`W_zxcIQ(uny@&B`@Co7r+w(SFYZ;xbB-4|r6Uv$>K```1`WuM5syAp;= zxzDf6|A8a1*%D%R2tAmZJG(C=pIeA+pdctx@fEukJsdmQ(+fR(OaXn{K%_@rT&I3E z2)ks@3DKYA!=Uy=vLC^>$^M=4iPMTtL;0ZNdW-a@j)hpQ8eo!nc?-p#R$X(LQAnOtkY|q)mwe`Of|`_R(w}0*G$@yxT$9j@a+wDZ^LK5c&KD z6Az87sBh3fS+ms5ViTk=-f4AbviW*Rg5@iH! zaY61t`=p&(8Vq;JjaV&Z^YH-vV)#!D-A(Mze;XWv6E^Z-WAqAty|FvcrC*bjc-$Gt z$@J4}zY=;nO(D0x*=-~A{{H%}-84-W0>*N8vx1mbNRHmxc+{B9mxJv`gT$^LCU_p2 z*Pq>&;y42o<-zs6dT}tAW<2?k{~mT2YS?{S!hR1JUBP#{J+%XB=Uiy%wQPGy;=a7) zP@*oIcMN7t%=-Rju+L>=!SrHMQDVCep^ur{ca#V@Yg&59qM* zdPq6Vw=Z(h091LBg#GmDk#@^x?yBDp7?IfR7`&FKbC3`Ij@bVw0a->|+6P7yvGbIU zSIwiX(8Uwqt14NEElzo#Ha=zNWk`#8M%M+LT8#7?FGk8O6aV{tNE}8PB}LT-;D_tC z4}WxfkZR7&^K(bqy!DaH{L=nri6)81C;9psGbtRS98M{aZGf2vEwzTnHv!9lw3AU*B=g>B7Y~GEF*MP#}+G=?a`tG?Ql}Ac4rk#a2hnYR<8!;hLgHe z`L}^OAds)l{DtKnM>$3JuYrEU_UVZ|SD?z&V^?~oKMs9cQ)PR$8>X$BFYm~D2EE4x zQ%UCWP}yjlC1Lj*TeVz|ctzjFdfjVx4rPlYCCWObc~)lL_z!uxq~DZz`QLwM|9{Vr z{4{@xF=qBsj(mt_?`!BOUeJEOD-5eD#Ro6x?8T`iYQa%I>R>eM?6bV8EF2~;+3R0!yY-^Y@ZCs%5q8UV1>K#3ut( zO@%ol^JN$K!f@o>9g#m=kFkgIOz8%d|M0)=Sn<24Y=f?Db2ov=HBj{9*)h7p6Qnur z-{r$o1B}H|@%4d9P)zFjkkPXTyECnL$M)aG_T){TwGpq7x>Fs!uM+hc3U8`4ICSE; zW#E49w*)Un`xbS_yfZ7IJ;uI`^CufeMdJDr-@njF9L9_f|G98CYTiDj3-vwUB@jhg zp!>EZ+V2pF}QxWO4l8^OpLENc0#I?^VK(yTHS+VOX2qFy11K2cRl-Hc&`t< zRo#27Rmu{ETax7zIc;!Ux5I#%PT)C`jQ=<)``v}XzDDmOx1Zt2XAQXraUXygClpu# z#B&Y4ZZtM5B|cBOSKbd*69oFAb(4#phhW|(H{F?A1YcY2;~a9_0(YDqtf>1mB#+dt zeQRI4dO(%Qg#`b+W*m7Rppg*yFJ5eWZ}clA7|Pl{i1RHX@+s8)5mD*SM_|w*OeX6H zkq;k>3=NZVjl$Mx`4Nu&&tPPAxxlX(q7P|+J7mh>$XBSY(w-?9{tk1AS~03qM7%qF z;cH-uq#H@_*lnw@+KWK_LEUcs*$ApvMv11f2wjWuu+=mk`T}VDIw+edd<>- z$6~c(W_Y`{DUjPGmh$};!|uKNl$!OEV0MLQme3Aa6940a%U;ChV|K&JpKpI2hulut z{7G#$oZLLzqnk+RFx1-csOyZmU|UR=!JIqbaG=;Z5AO4rZ0XxiKe3M3k4|wm7cS)> z?fjCaCi)5X{^!&D^DboYDb$_0Z!bw9_>O3`osYxfRs+R-bkDLJeiN{y+vS@D}j{Yc(Onx+t%)jlI)H9WkUKSCW*Kj=E>7sl4 zPNbFfuhY8AfgLC6UaBu}hMA08zFF_wp%ViX2X6R~cy1ePc`0rItd#=?>zk#Z`Fn+W zCtkx8kNoVxi;Q{ef5akEaYiu!8WfrpCq#0v)96^m>ysK#KP_5GovDPLh1=T(lK>{# z|>bKCY&-H9N#zn6`9@so@VPFVn?5l`h$rb(6w)6IH%MiAV+@<^3}`$ z)?>SnEe&>9ar?vGAI7WZ-G737{YQ80btqDQcfDy%9mZH2Hvc%!#x+65-<#h9B~N3> zcg7)}Om?1$#E~WG!ofR-BONJzVcEZ+=g`iat8?dow#VmV#ODvtx4-1*nAawxqy`tf z)vSQw7vc@M`^s=^qkz5ALAD=?jkdczx6A*9h9!HfH@_fsGHJOvKZ>*+fmN+@yZ419 zG>D^gYT#Dv3EfFs6RQCu_NLdLGqbR*lXrX3TrX1BYs-n9eGi?#Z(VYIst5FY={5u1 z%1GTFbHu1_8T2J99$|_X!_=gI{PNaB?AkCpox~H4J*~yhj}KR4|B8xjuc}9pVN-9k zxJVbvcJBHumbVZl7rt}vX)b}-q#6#$iPIT6_OB8=T&9s%*8G?KB-w*Ky_5kkMW>WjJF0U9@stCUN- zBy&LZ<;}qi9~Lq$QrEsWVB?FRsea`C!Jk^#<6b7z6mbw}vE_le%MWAUXw)J>4_|2W z+`ZGIE&QJ!wblFPKsAzecYG{y*#Pu3@2ie=-K0fd7T<_7AoMK9;&_cpCtpF&(rVjE zS7W5F87+SN=>`Owns+bXB}C%1=q)&XyBJ4c>ANSkJ}^Zt+O4U02RrZFYi&O%f^Fe3 z?M>A}IIR5Z#}NTJXmB%PwyrnC9>XeC*DcA&T>Zw}Q^5!7i`}?6$`z3UmrhHqxdXKu ztqe>5ni1!{&7;BnINgI|U zyAoTW_iXR=^u9wdnxxY|IW7fR)s=^`wfLZOrRE{q89pfSI$N7~rWUEep=&Hg2BF8h z*Z9f~3z%)em+y?&=NeRLR#1~Zb793#22vvbleN@L<{>54C z#|M#Gxpc?=I5G}7MmRhqPXp_zgFT~g8YebxU!CQq4+EJ&0?sm@a87rCj*8cCVBpR5 zfV2Phf!P~>&bs?BC@rHMe)Jc%ui3taRON(|P)alW6_0iA_tgzBvVcBkN*XQAo;QB< z+2$?ow0$_WE;{#_Uo|q8xY@eMR6x~u{1ub?|52ayncKd=y9_AKQ4fDJ_XBlnTFjXl zC+K{luHJKU9=|3Q{psZ`dQ0@BO=yT(9e5UwwK6=77b6IK1;Z%f>>7U_q-bj|{#9d! z{hr2~McU6|XMKFB;2RO7y~;Wqku;Cz7Al);&Ei&`Mar?B=&UpL(9!-;Tm95Y>}zO# z8QEz|*kNsNbBXUnKmLHRVfoA}c76o=aj)(PLqbRCYiX`$z?yy`Yc}C9&=tJ+SVwyf6`O>E+VSA53(PbaF`^ykUW>1GrT6X?BNA-?8eRmKh zlRed693=89!&N4<7#3TH13G!7USvJhKysdANsvF`S29M*8yt6y0VN@!aK#vP-h7jp ze|fV7*|abuXbS-WQTq(DQ6LSJcHNN zO`e}+AT?oXyXIelmu$3Sox`GTcK#fP&9vUAM!&)V_e9@ud*YlKr|xlVP7;EfVWJ!I zo{)gmNe^Cb849hosu|X93t{4BWYkF-yFVMds`qLYoIZt&O+`X-mt`O$bhy&pj|7yd zG70Gyyx1dlR!iWJ1++TJd|H+hdt@frYK{Vaqx(ZlVI;Dq;I`dG2!h3V-n6{rVS^sXkqyAm2J1M zYtccWR}te#_BgTdtW5*fXRP0^l>HJqjm|Fcb$o_&*+UL03m3w`&b^vtd;}k#fZ5k9 zkKGS|9LcM=RC*ZdGNi7DIuQG{swaBOlH_eTHZr_hel@}0)Va%5AK1D^FySN>yE023DRMue>~AkdR=C_?X4`k{;-j{A{fn=&mYclX zBBF6qp0IZP3B$Dw7huDexBu&?~Y;=ilXq4w(j=o|e((0$3uu;y_W zR%EZ|SReBZhCj-ESRqq@UB%O9=f(vHd!pf47yWev3ncpg95vjGgJ0zBwkNUkL`da3 zB^BA(544tk-?nigZlCx@xpnZDCN`C(@1^*3!H{iu_W61dq69u9}drP}lC8spW#>e#`zNOc-|*BSN$#~urD4({5FX}Jj^ zPqzQF&#N{L9a9*_@wF>{b~f|=vjExm!s3a(#g5;Fa?(SD?%I@C%K6$Ci?FgIyOc+m z;E$3&&_T61TJ!34Uc()MA$-PTeQtlw*{P*7VC2GKET6t`=I0`A9 z7ABT)3ivl*GI~5C7N|9U2kjd*NrHT<4~%wwz(Lo$xwmtEA=xr0|JL)_49=qmYaAzk zs%b64#l67t~*;utvIkr^1&)7+WWmd)d*@F$?KK6<01Ckx%;Gzb^&IMOOr7F0+J8@eDnF8G%^O8 z+OD)ulDHRsFzWkIj^j?dicWbOVTah+VM(*qNDN^gj-HlDX5rl$q*oR#+RZtQ zQ!6!pwMaI=fL7&4$*0-acGmH9!Dhm)j)~<2yvW&!)k>5T8hfJt#dS}_MD}RHbhl>m zwLL^%e)b33t#-B3FtU|bRx9K=%6%ZFZcF0miOipVag6 zFq0eplc#J9S|+tjFZw=&0msU<_d<;^qbhKD*|P{JBl(t|Egb~bMXQ-L;j55(a7oIg z&n7UnC^x_&z!u3}n+>$%wgMyDi>YC{6-Reca?H8;{`oVKf>BaLzN_iQMj>hRoLWz}9Vls&}8% z#C9RSmlsr~p=YCeu<((MK&CiM?4ACJ<*&ANfBBgKOw!;$^zsQ3_v!k)H@bn)o~-Lt z@X!(}Uar`*ZS^A{Z*{D8R=NyR49kS+JOWSd_QNH`C$<~Yqo=nu>D?V#%|g1TiD)r+T$R#5#GsHoV1|qfQjnxn3JPwFd_GCCfw*M z4y_atSb2+$4}|{Jfr2`TL|*}Y4Od=_sxbB@o_}NVb>Bb#I!EF2p7X*eB z-U$xdf#juUZO+G-BZbS@Wy=XFw5?fwvw$xO`nHa~-o1dx2hIe(XmL09gJ4ekPwubf zpl#;R>Qe3h(A8CK3N0${#nFIGnf1l3b0Ur^!&{WqR;wN6?fW8>?wSL*#h zU0nY&|A8q?-5t7SI1>q!MP)hu%gac7Zx5BkZtjDbD({dftrtKZtBO5#tpGcM<0P+Z zF-aUA5@$k^%F*!ctt7FR#J*&N?{e#@Ve==FIL{j!nQ#q}1VwClR5Dk>l<@AjQ|Bnq znRn)RrfB9n^OEBsE6FEcYEL}$IlB%$2z&yQ4KepZ-6pUjE6m2s^)A$VIz7D* zv>kirWQNqKWSA2X-FE7e6b?2Ny*?y=4k;C)_q#T>VaMd(zXDBet!&3B=ikxa{Fp#ZTBvLLcNuj55awL__tL*}`I!FMTm{HHT;j2;m5sB* zJgq{x_BCvMW~`x1{k~Mrt}~Lje@}P`R;se+W%`X^aL!ONiKi{ZEp7v0&vF)xyj&Me z=>Ciw#ru3(<}=TJHEL;z1RVE3R^GA04@@JlbJ^lkDsgXcnDp)Kj)-AosJc{EbB18^ zz3%wvcy>P{j@p%+%oM8u=Aw;{HrdXLi`l zUB^yW+%E{iA@P(0ifh?CayY5jID4K?h{RE+SMl16jhhC_mQ$>Anh{77&Y=x66 z%D(>#42W@)YOlH%*a|G#TgA#Z>~(_Q%b&hgbUFx>prP+)I5y*im{IW8XY9H+QWrNX z^)6pQk$Cb4VK4vcI(-!9?0`R)&pe#sNrfLH;$PZVEyrQkz)2xH z!v5DV%Z59C6XV9{>T#>#R|C3{SWL&xYRv1lmpHmV4W|zX%p3?NaMDGFMrkUC3BRJs z`n=D{{_D`hI3La#!%OTpPLIaULE`DlLID?i04e^a$ey29VxeqKd}7Sd07~U*f6Y`_87>=7hTnUO7y*UMQ&^u zFktHk;0)>L`W(p#tj9yu<<%+J=;<8U$}587FOm)S%qfugB%aQy4(a@_b%Vd2k(3mV zt6>%6oHN^nvp7sylHd}YB!+C;=doD=RU7^Gl}3KtWMm&-Mn-S zj)t}-?BykRj0ea0xkzKq|9xK)N8@bj_^op2jDBS)sk$D!OS}{HR0+O#YIe`^$%h1< zp7HV0iMEgHV9qByUc<%^GIU3SpW4pTUxbMzx4K1^hhaOnEWg$MS?CHa&JNwk#7eIz z2fN{wK-1dN94w`TRL{_@Nk-l{n{d{bu5AI`hu5wlkEOLq%Es?dPn^} ziF0;ElBdi)WaL|?9nk*;Gx=?llDcwat^RZHlbZ`naqaAq?CeMSQumC{D+s&+{jyT; zpJPMNbxc7ys9*!6`0E#ntDVR4ywSGnHhmjgH@9u7x-d za?8km;tJ9Zzd!v&)r+tPHNR9JEi1-h&o#`qqV-5Q7prmhi3Kz_3n-j<6@>jhJW^-g zj6rwIrl1?gI3Qyd_o&>W8wL$qUw^!r3FDPsZ9L73akBC($0pHxBn}D76_(uZU~)lbh)DWf zAe-K&3fB{Ld8++B?S{irFikoiFSzIj`|mw|3#1iFcED`jp5Vlbd`M9^LiIi1fwac6 z>*6mc;HaO=3)e&rXcaDX61E`v#_Rn=qRz*PB9&X!oPUW34Cf5Fy!2{7>W%~7mabFA zE|K*u-Vd=>Rxn0_7B zCR?!#I=r;K_3EDDP@hB4$oF~rx=_#RFqZUY`=cMYZgIb9N(+~GZLEsXzRqx&3;`o{rXI9f$ z7&PDZiKny_n~nH*>>`dL^^FL<`qNkNpvug3XiwgT0~aX8+Q*l^>A2 zS&uKFuN*pR^Jl*woW-G+zq~{v|HT2JvhNRA5IFBKlUR!LhK(fN0RQF;r8#VRqP5wq zE)LTra?C{iR{%|h-WXN61)SHl-g#BJ2&y6t4=+u5@NYkS_}+Ja410@v)*Q|3#bMu< zhRe*|fq7NhGWuXSiPLOw!Dy{AOdn}ZH_v)MZ(Y3o^3dh9sw>d*6@N-ft3qpdT`9HO z6ier#cD%hu;K<>k#%uPWeArm&Xs63t40ubz_OH( zWm^}{%bOB$^}b3z@8F34+wViK(C4baCuPvIp`>~D;1g^-D1UJL>woy=p?q&BXXo>O zF@3j(YAvw-|M`7d@2*ohp+e+2M?Uiv|GgvrKcBCU!^|d*z_wW*P_nEpV zQ;}ryKi9$kypE}_WAI(W8yThTkzJ1v`wV+@54yfY+ON!dJ?CD`@PD^5$o$`aCT;gP zFtZ$KmnIZ7+P)*XK@Rh-|IfZD!mk?B{F@*)&u=1ewW+l3t=kCcPF5ywJ4$*oAR<31m!!$f`Ns!BO%3rKh86v2#G@RAjn0wg+tL z$tLtGk>!2wt#T*PFEQOKsQs;v&|QgZU+K`Q&aPA7z{dl*+pZD(F|(RX(Y|YJ{{e?K zG~aV_o~O5g1Cp{k^52;LulI&`o2}&B95Ux!N8eXdMnic zyPZ3nhuCW4iG#xtm7wM zYCYbL?Lq#>DyF<()ObwZyLOhuDc)ureQgy^oz{D9y{QeRRa>QuzMB9wK;y_($CE(M zllGCTamE3tpUumFAS8dW zIbl3T@a4{VA9Ak@z5-L@V%wdCp5U0+#cvmMXgHxEqBn5a7AZ90A5P_LzFz!s@NM(n zyYHZU+wd-1MKTs$TEgP~=}+Rw*%WSVMerc;$}H-TGn9qlTgScknx8}ZnrtPzZ>1!T zmv&brc!sb*w!6jungeuc-QdX7e}WX>p2C7#zmaj5ey8(}CX~J@cR2ZNC)5sbnN${^ z!_LERLQ5I_85{?rH|887_UY7GhmbxCMWilaMu=Tfg4z=)m9-Hj*utw4pY6id-NBiG zzIZNqqW@`ZXHipY!b&XN>Q#Ann9Z99qs#Q9-Evj~eJcd&JmZ2sMw~`q^i`N{`gQ8` z^n6@(N~`f{B?)O{2pl3=y9`tIf=-M6< z*J@FE+%cl=+_lYS+p2@hkVaj6eD=H;^b{V?lYYg1KA76A^h)hDQU7R=p(UDivhx@u zPR7T&CUUDVNXLVU-Hrn3?>7BMp5hnDvt7CwX=nJ>+Z;X!RZG8>J`~Qy z49EU3VZWU)Mo~ZKrI3iDv%X1fb=RPRW7F`R2x7nYY&Cq~QYnQqWskYT0@?ls(&P^= z*vouK;yLYSPipx;x{7hFG(S3tqk0YV$l`zF66XC?O7=Jo1b!0b4fP;#Tw1=uv4a=M zB3skF8sZ_2FCxBO%^lLi^&6y9zhUdCAP3(Sg2Z^;TRa+Al8;SEVp|t43m|cZUgw)o zI|FoOc@51~-Y_XNGCh)JOX7Z9VR!WEEBqj}@=KD#u;4fy}aW?U+7sBu^Hl z^-JYqXV)8(I@kZulg*63)pR=*lg1A(^v*K|`r4$B_+$ciT%yghdjAJ9_72J$%E{ix zGKZx*gBzl8c2oOP1$AL$Y&-P%66+_h`;x&+^P9*Fkhb}yNd=nL&QtH4RIz7WYp16z50LL| znDai@f(16mws1SR5`Lvl-pdcG6rgL6Tft=b@4xeyCT!8`jZLKWL8qRn0Zm&uf~02l z@43t_J?qg%`lDa6*dB?7kV$(q{ou>iY0S`ZO>!kH!(i_9zO2N}m@#V%(W%U=>wY%hcB|{!4YJUaP z2Zw>d!Smo^aS@4g(`yYe0RkV&v^{jL%-{@UZTPnTiLeFG>p%Z-u>6M8pABCe|7SO; zPi?lWnqv3UA=9?aPjL}j_XX+-emH(nGey?EKykBmBS;V72$_?xCGm-l4~gH@#&H|F zL|?`}>TME{XGHVSoWGc=3X zw@5AV9uNun5By%s48K0h0VcF3Pk6Y!!;X+Y0SC*s;7oj0?Z;6i_V;Qtc061ja0dHs zi2Iyvcu(SNxiBbf<_5D1Zu4)^C-^kF52}4m*hS1Y<@qy*>sPlxOHG@78TUFE4Vaj+ zRGxq-hqiE}xR6=P8lwigTEaS_jkA8}ki9v(l0Gb`jt@{%g^p6`C-zB|t&tYGd* zlYD|~GmLLJ=(_)lG*k}Br(0Rgp^??RQ?BiwUy(MlY&)MS z4yaFc9QG0-ac;eEYf0pKqzXJaS)g?fCOG_0#U0oI^o3scT&AOuUV5})6>|<G{BiE<`}p0b$LE~KIe&S$@B4aR@9Vv;*LXgkFCTj~GgC|- zX<``P)2?9-_}DFyGEDDa>5!d;q*5ag?~MiJrmn4Hi4Ly!D)W&4x%D%7Czb~pR;=mI zNyT^v1j>&KOLu7c6KFv>PS12)4|*MYCypNIq8sO>$FW5p)oJ)l=pZo*ASYEA5jbNL zbk_^UKSn+*s=)MU2wcL0ZJGrzxi5T&oFN;U6Eo>I_OB$#OQVDIFUqZ2~Lf6AjYG|14t3FFW6)IGm4E?jO3gWxiH*4tA z_)IWhpwVKHf$(e}vajqF}qI3a`z#jaMZTvIS5(O>yaiv0jGH2kNyUzPTKo&top{eI ze<>9$ZDR}Ku#Upz`QgrRk z-3@uV{un=WupY5Gs~?IzKh=Ei!{S0{(KS@9_{NT7D2XmZJvexI9%Rsh`o|px zx{1&pX=^c@J_|M0d8tXQY9LC@ACNAhom)62L+0O$J8l5M5?}8dsRH9i zMG4Pvx=>$tyMt195*D4ln;%QS|KIV!_UF}K$KPfjK`kR0skLEi|NkKJiU!$0uiDZM zr{d=@soXh-1b-sJZ*xt4zdgd?7M*U-h=vGK_G`ugn+R*3)ohS+)qs|UUmzO{pR_#EBu z7yLcU*6{_PIW5&w!6ce)3{c7kv?g-8(6A5HsORzy==PgAu;oE7)XGF!$v-wjtM^)O zZSUR-gueuLj`f%RIZkuc08c=$zq84YtBs*I=A|6R_Z1jF5}KdT?gnMGV_$g%q7j*| zaU@I77tOSN@#R#b<0+!%@s=6hcSAr3V=6kd?>Y?aC|(e3Ylq%Ull#`EhjFYV9M#RJ z1n{bnpZ4cq{?r1m)I)&*N0>@++wC95g~sk|ckAk;;j2*pi?|HIBj;hIuRCnyOE)4L zW_7yn>BOxI(5yx|E2#OIUTpQp&Ct0ztg>~Lrb~rm*Vun&%j9)5;~}gad-XmHH`jB< zJ;K8*#eJlWmmM`}#MfxZ`JwY{6Z!RRoC zC#jFnR5`zI0*dajxQo>&pxIlu7y`Hb55H6YfZxiovrzZ-g~*nlDbSpCXqR>f1De^+ zN7Of~L3QzW^WOii1!}#XxW}8_sJd=n2g@l@pxE<-DdKAw7f2m`c%JlS?Hj}tWBYksVNYySZR{(Rs${d zN~>r7x&e_v-#_zpBrKkm^+1t(p?CD+70)vII1ygBQqNWc;~@`L+UbQ#Vz|Iz(QA7) z3CTi7#jP`=dTaiBsbKOB4l5x4iuY{#(1KR?TR80#!QyV^!r%Vx+>wN;7Yx7m2FBnR zo_ZH;UBUF<+NdSL6xX%)s%-u1Lg#0I@AG`Wm!AzHWH#5oyolB7iC)I<_&Kuyt;~tN zNj+?enqu~y*uuIGky-7xcOGj;6z!law{}|m52{TUO}KW@bgZF!^O-@;SGUkYJNGn0 zdKC0OKGCOhVH(ZmlJ&ah^I$C3j>pir7Z!HxVSK`L91?~P$LTR|gW0H?b2glquHpFi zvwhF;{ZQv@@#*BcLqN64YM5scL)65ua|z&#{!Z!6N|i{WI+S;ld&eMZk2vflJ#iGx z)cQ3KoT;MQf4y^)g&GU7sA(=eVegkLG;YP*cfgMiy&2o!PoDgF_g z&sJmK;&7~9*m4wmesUon5D(tzI^2oXx9l=CUX08Xgzkc?VND+~9K*OKpCIyGQ zU_t%YMe&;VYte9ul|ynb?YxI+YCx4lB7Gcnw&KQAeh$VH>C|OA!s{$UH&0Dki5E9E zenG3ae1g*9CHiyzb!=hX$6u&7_wy?k!$CALE3P2d&;dmAo-d+;L>O_8osMxCKvUQE z$F^D50YR^xd1c-S$2mQ%I?Lk#Wl1 zQ3I);!cxx4fyjRzUS_r4i>SVc{jyIVEc1P7eDQG;%nTX&-_J6Jk%tnhL7Y;s1ZUp7 ztMW(rri~r}_$J7>$}s&(#0LHao^9H(3qq6 zCm~c7Rxaw;CH}&%`T6=WU#1q(^h@S|6W#YcrOpQ&bB1YNPcRmT8|cWC zQ*v`gP8?otG>~=-M?m+_L?6^VL~$(9WUyhr|Br*7~>SaE6NqXN(!sS96uNI_UFVeISHYm z{DYj52(v$lTdb>i6GVX+cVG`>jkB%INorx{~qsog-+4N%#GtJXWU)fAep+sQE=y zY~K{Pnzao`a&{_T#s8qmXqL`0*Jnt`gWB8}3D-%a5 zW7L70mfue!^&%veTnipJ1WM z<@LM{4fl-t9=~jLpIlr!AMhI-B6eZt*L~%9o3uOFcwY_k_+`O@V`^S)lB~h-=*{*g zylgcvJ;&iMp8NCz*6fwE6SC*quy`41O0@mS$TpZ5?doc`HG$$?q2%yp48KJb)*C2M zxQqH;ys9tN#dHQpq)mHUjLzpVNuAr8y54fl-_P8tCVLs>_3oYDqeejF%i|WNDMZwD z>|=uEVhm8EuQofMv;gY+P*_x-#@1tbwD1wPQAB!s_p2xumPfDIdo-%g_c9`0jNFgg zJP0NEIHtpWb1-u1RQrm~MWK+!7qBAyy$u=~;$$ z^A|uEd__tsas9u~Lu-G9P;!X1{TU)0IQvNYi8PMomHzQ-y%-OU8n3p?eHzo{pN~hU zvtz0NKYvMW^x9*{&%GWPObv#Po$JF__~p^uub<~HHeh()WL;qT!*dJAVj?nJy;A4iAWY{Z>}`ELmN(7d_n0PESBxn<=)has<&QHjwmF<$Qje0 z&aYI^c-KkQvHEi`!+Dcbyq>082DKOXAHC!lf{KV9ZLjZNA#bN;h}KgR)auYj*!KDy zw0{$s!>LdaxghRL;=B{;`z4<7gdfAx%o{!nu7AXfhzskc8`FP5{|7tc*+X=86IAL{ zIjFlGLroV5-*l*#VMJ`4iTNj{OIfB;HNG`%H|pKg@F_J3%SX*VPzu|4Bo@uNG4O}) z2*xoumh*5i3L(NlNtXc=J@|0zfSrHhU%LBI%#L5G7RT~ti}-*)y%w^l<(Grq*RkiQ z${o9@^cwA-nw@qfK40`v{VDcT{IzY!$Zpet-EN-&c|r;x)W%*d4;uB z4cp%>B37;gwiw=+%!%gDYS^QNFNt6si#^wX{;Pd&ZZgAQOgYccEhkv!81vW_qKspB z_0EX1^)WP@TX0i&HHgOCG7l#1FGEBVu~Q{ZHE34ADdcI@1{_PHOqVEE2f#pNtmq#j znAzA|))h2?X6;>xRL&(d%;*_It)S^L!bnCS)hk#RHI(3E6F1<1;CHuG{kAQR-OaWr zp3n)S=|8VKrDNwBQT|<%-?%7_Gce@K))eghJdxjO8)p##l;(m;yMqgeD$gwutEGbG z-ID*9G;Ia4=6Y3GuL;QA!jW`ZY!Di`znkXO^dLg>vz;ON@z6t^WZWW6<3RyQ=E<_? zX)GSzKcJ8kReT3sw0|_y@KO$iUw2YoRBc6rGKxK7Q^siF>Y`oBlQ0-ojl90-gYh;; z-|21585#k?fz&srCu#X0SaQ-{e#K^jW+U!u{#1Po3w%kDF{4UQuz$s4f@96jcoo7( zk;m{ylyAH@BTIErlZ_GMhU_pPO{$0)m11>n$J!FVJ=XsLBb!SnL+{>))jSV_6m%2_ zgpZ?C_vff~ZoA-dsx9Pb_T5inFa>-d>4mxHSD^aGUQP}m!?LnSg?saJ9McgGx6Aoh z{-U~c&q328KOnYyl6zyD8!G!u9+D*lK=UY%{vlB*gwJ8Se5TqE^3U7slx+Qls63~& zcMsFzXZ9X>$(HYh3F}CzNkQWp}KgN zY#MF{4K|dIywb2ion=Cn1QUAQtdY?Re+!00fPAaMK2^^QJI4YAYM!vrbpK&`JV`!j zK>+1>Ewh@M_o95ptly5?uz0~xYQ}hbc@r}mgQn^0D@;s?&!dQDD7~%WoX))r?JjCBJPt{tg{a?-mp{En;{_}aI@V)2 z!}fC~{^7!2Gx^i) zr>ocAOCJ=aH=TqF9F#42(HlZQ{T0jlO4Aju**c z{@0Lt7}L+UhoOk)7=y0xXGAr;vx!x!5Do3B@4C+>f@8F_v|i^K0t*Jr_cy40gM}o+ z9?i!$pi-kG^3Ksb7;2pwuzyKcHw?%X?2dU_No#h0#74u95@;cIllH9K5g6v3=29Io zgjwK8V2Kb#wda-=r9-fK&Qo>VTP{q}c+H4>Y|9z#`$DL5-g;k=z$+lx?(`tlbFA4( zviH0-XV9DwKJ_k+39}FDnF4~9Z$JjeXf-@t53A;u(}uzxXjpm2u{0^>HGlg{IqUKb z9Ea74Hc@N~Dmk-*aNl|vkxeq$0_7yu{GEgh5R`+dO!N58rmJYIxzJWb0E;*Mog$ws zt%!xm2gSKR*zIVa5Bdn&FiH8L%vP=Qx*WTJ#46us^c<@Ty&APYjmbY8@LM+7Yzj01 z>Or59$RX_9CVPc1@F!vQhlx4GWw~jzdK55PQE}F5hzz9H5~8`uS_uDTl$yyYg_bsy z-X=%oqS2XOM+*g=q5t6azh&nK(29Y?4cpK)Kj`?%QR$cjy7`IMJf1prsS2g<^*wmw z9UU$cO$Xfmv9CuKX1W!{7OxG#czMy*r%!L7F;VBMo}F{h7{1AM`_WHm(APxI@=rRD zGMF9@p2zG-ax&im5nCea_j{4i$K!-bwlFCVTJ1t3#M9@cUq_>{_0yyzssF*j95KAL zX?+e*t>R0fM$V!BXSe%4P3(eM>61PNLRkK+GQ5zoHTWYeMRytgJ@gb7`&X2Pp5;KX zGvoI7OYvy5F4exT!x!tv&Qn5Z7R50CR&nDwgB56B5%cH@q~!rn)tAqHQ3tcpRO7vM z%jE@V$Vyo$3^HKSz{Lb*V|*d}#vC7h6$;EBuKVmjpy{9j<+{{d{;E7AdHfP{BfUbz zNA+EhRgK0jU31U$cgHa`gu4D2!R&{wy^%u|`z^5bP1;y6bYKds@_F5HW9xym#j>Vk zu^Y{H-T9hZLDMsZiNt*|+gBT*`OxMX%i9%bvh4GubDJ$tUbDa6`->L0LOI?1GcD|m z(EoL3JxA*yG%9(6=ZFn<{~_ngd1@ToFi<-$5fxU8#yJ%ae|}Jdrh-IEqOSWw-IJ52 zYDzKub*tWI@@y=|Qyx6;F07DcjAIy=2$p+G)AvA&-k#6TEn;z*kt4&=2md@oBw4&% zOlKaBUBeN7d%^*R$ql!ig4f2k?pbfOyezc#W`8=Jst&_v`Ll$yKO-XfY{I+6tSegd+TWP&a0?ckf|b#pM4+Z{mu!;4;s{GQ-BRtA|5F#xMeVOd^I=4s z6en{Zr^O>sYmKt32F@4=SKfpjxfcK*IiH`JUz9{MyrJ7pp8F01vmLFqC+g7biK8dV z*Jq=}M;motyH;b6#T5FQ*Pi{cpPnNG4j!1O98Q$F}zx=Ae=| zw49hAf3e(8_x*^rQ_LU0&dcs&A2{0cgQ54QXG}mj7c@((Pvrb~9Z|;3JoGZNVb*Uz zaf2x*8ij(*?|wx>(-+t72T~8tLI@TV(PlMR2iy?ALWrvKABKgr?|DuZL3s}J(gszKwPxoMBHufWLV z8waNkkPyE6QrS8j&3*zx@n=8v&+l6oH%hW{jrRkGROD zSR;HHG5kfH8su+3>K5{L8qK)<-djI(4z-hlR=U`pp#^OmX|zTNC12_G(=qizqg(eV z58JL{exajDe_S*ps$CGw&UV=fV~mr{V~^Njnm~Nc6fBIU)85xt8lOO9HztE0Pw8}j zvGe#O<*D@pCDhAi=1yOLM=r@$lhHzaq^_>x0?gR6-{HRz3d3$&*i3>W5M^7p zu;8tGxOKPUFU9!0!ZBqG_dLDSjgkYTl)LgWXzeeMEO8~yGH5WtxcKR{E;KDM@by}A zJ&wJsgrQEw73O<5cZ$h&qXo->=_h{$a4ehcQjEu)5JhL-PyAFKnpbwHU|)~T&s335 zbMVe%FlOJYydQ_Hv#EPRbz(Sk)M4|$r0euO=x*~GWXNH|F>Uv#W&1t{13aI87kv~5 zLY~ej1&^K2Qz7*sKh)w-d&kAaNAtJ-@A%+YY z25+!UEO?8!f zp6&hTz99=Kb?NP$#WC5T6SBH=@fw)<{Orj1b{WV$P|)CisB>+9Oqxo)8%LeLqK^fX zrJ<4ET>93pIO*Drf9JOUPbd$L{iN}%v>8GJCE0Noq<*3L$ylw;$JU|V4-viBpK=1h zTkk$vu}AeOrypN1oP*L|Qjg*!GSEU%0iSk~FHEjG0dFoy!ZVv@-^FKsw0@PP{n>2S zeFk+;NKGn^RNIK9VT4&!1Ed6BcR;m!h7-XugdUdHfNMvDC;a6rYO* z4-H4Z81hBaj>*m?zq+t~YTXRyIgZtFZ9VOG_GU;MTDo$2!x?ct2yb9WzPtV|jkiJ(b*eU`^>+>w4N@ESB;7+B9Iah_Wk}Hc|_fGWHU~`3(dHT zhZMXXhe3v2?@Ne_up+g&;pXvIDAi*egmZ+Tco_8V2X4VJ+ilVF zkvFVym|YLl<2*m2Sr~yJhmRarpyS!Vyi8rllO-n@Y}%#q4rh&HoT+!Tj}m}s@2W`0 zOePpt++=&I*#?MiOaot^1*5rxk9HHszQC9xmsY_BZW#7)MB`=^sFQp&L{GN~re^}T zmP@Px^={Y-W#a>wSD)BbS4WH6A>wu&p%Uw_(0{Z}GtKomTJ`uIwvovhHEUg#Nz_OM zim7*xgFdED+BUzd&-L~N-Q6T{nJ z?=|K0IM#mpV5HX|u8W3l=Z|R&3 zgi1D@NB2vu7n9%xaGSx%aM5Vs>4B``-X@rsd{WA^AS9Q&2wshUErZjg*n4>8jnH$wa2U8T@AAF; z0PQ;BpM-<*LwS&phLhRBg-5ZHFF9^dnV{)I`RXqZ6QKAzu_#(70U`LblTYy}nD7f7 z`ME(0QEU%(PAgz``0Ux~4b+7R=nMYy!OIbc#$I&VS6aOV>fgo)iAO{ny8^rD;~ctr zchKHFN$m`?f}vavMeDJRDF1t&nE(AmAm0z~O24>>7REQTUW~jAb8XMt9$JM!L)yk* zcMevdRPanZlDmlp41@Tk7cu*AF4r|+^w&=`a{Oq%R^de;>!fD9h^qP5{9txySnP#M z>uK*{9G8FjVfN2*|8*TUzG3G3Tux$mq4pxn)bB1yXy9nb>&0t@s5|KF%|qv~`;9v3 z_Ac+iaHFH7XveMZAJO#-Z#NkBmhj)a$R*1ck8Cb!qKfCovly3?P|CNfI2lPb6nOJ} z#fE%+)UfYD`t2Ydv}{k>s`Nn>O$mNw3*4TA=5;(Gtce%V82694_VXBz`fFgqjxN!9 zaG=~h*?j#ld^Hu(E8dv_{aZM0UJI=NvTLF(>F**?Kd;6K?>YzV+XdOw7FuyEp1EqI z=Ns1S$Jn9-&j27X=*RdH)yYH7?3LPNt- zgFg}Q&vBlw@>i>IO+*8J)r*(!(CTF&vZxe$t8gH6j=ger!e9O8IL#Vd_fxB#U9;C1 zSbPoX@_yL)SsCN=d*cp7ZhcViVf#=22L{p-=3EaQ1j@Z$Ekf8)y8ce)Br}W&j#%<)+{=; z%L-unNn)_GRTQc{^&Rn*HKV!nYN_nIRq5JmGw=K*%yJA`{0a;5Ea-TxwBJ2RE$}Md zZ4Z4NqXE|LyP#mZ;kBG(H8e=Nq8!U%1hvy5+)Ag4fy`o=w|8#MU)B8#)|!wi$ar63fSxQ$AvT%53#1rT)97Fy2`#z24gjb^lDZGYe?~qPXgERozdh zX;~6dyZjsx_Q;EzeSQ|j))3A8e7en=Afsr#f zu2V^6oA1-=4g*1^=9Fb+CXSi2lKE|d*uTb~ZhY!*C`@@UjUrzy+gxD|EI<7HU@?N46JMENcma*syo>m-yc^96-5laSgyAs0hWk8lxA8*t ziLaIiUev&hOFH+n(ibSu;D}AtSR%~qvhxX!yMkls6?(iPjqyJzJc8g3sZKeoO9 zeQ*)SnyB2zB1y}KpvnWnVOiYnFduyFWTR6S8r3bjr?&k&5X${i)hoAgs|JwHUe%21*K}(lI_$X?>(cI&RhB4iB zP@Q~rwc^V?RQu!h{Ra6nXmw@$tWfE5G`KUaO5(!|jE`*dERm!?=Q@3dTQ^2tfx+KL z_d28;`mg)Lkop-uYPAwd_dk7{+gKGynL)%4vQf0}A1_fyc!?%3(tPw%ZXF)gzZNZT zSbF`>=NSx`R<+2r6j;@qO*i~!|7seBL+ZFbLOzG_ zweQE2VE(-N*CU&wU^QXAOn|o+;C&6+WNj^g+|GDrsB4b?`84ijdgpjT&ISRd*G0X6 zCr+B${6UB~+tzVv9fofr*l?RBWFAJzvOW(YM&_X}=kX))O}C&gzlx_XQW&UZo*AXT zD}m5AtpDbYF!iqbE-=bZ6;w@pEM`M>6y?z#o}+m}ACe1dt->F1hZ z-Z1$*wt~I-DBvv~S`RXM(~qYuXBAQPFZ7ZI?k9v|ah_4t+8a;3aQ|Evly{Vd>Ig5? z^W=~!?+h)Ef(Ro@!P(B@Xl{qr>T`+&Oj_3sNWII3_7(<*=dD_>NWSCjB==7p6u_5A z$wd98=QH(gP`x@miq)r_S93~rUcu_qrq&Et`fc$;(LYqq)dso)sTX(h>a+@7JDser z4OG;p(IoHC&$I`ap4}8*GQ;E!`ntx&sb6BqT-|{0F`qagH~~xMn`@(4-@xpj`Wi-i z7RYlweDY<*9(tpqt~pt8K$m;?KoMIS-M9_?RP<&NJBbz|zE_HNF#$pF zTicaWCt&oM;lY!mU1JoM*4L9YOu+1^O%VE%KR zP$kAnN|+o_-^kDg?U!t*xY(GF#V-PlJtW^MBIqDe4vhBcd12K0379A`mfiR65^A4v`0;wS0bY5Y80yfkWcAIy0`{fHA)4SMyFE5y2zEXZlDo*1xA$EF@CYfc#kG zz5U}w9Lw{+TTA7upuYorBny*JW45o!_LFWnCb6!AQtvOLX+e2*MYQHm#7B-c;v^AS zx;~lzF_tHoX;SUEw*lit{K&{{!oi?@%5Zoc{nMpMkcjk2_$dB45iTRgFP&PDiL z!EWvQ5oL31=+Rv?`v;2jmq%VM&;M)QahwAh2cpC=-W8#E#<%xmKO$W0JiEzS7N{y} zibHb$d=CS@*B&nwnuP`t=&8A&&cK1w zySJ&Lu>+zpag~4m*`JXI9y3p^tKie?`DNUp@c;Wg`s*fr&#i=(7l*FOAD2i8C$Q)L znS75w7aPBZ2*bpP3pBk@G;NtET(Xt!Jw*F{^7pYq)v|e1bBug~!XXXBciST^ZBC*^ z4-YSk_gLMKrVVf9MUT7%yk4BE0!tIZ*My1b?4v@dTzt!-)CzhM!rod(t|?`X0sK7}L0Xs|uCqOYT0+E)PTTSA5&D|HtmW zRsSP|o2Fw9tK!pV*j7wnfD;#%eSqc7<>^uHH zJ-3keR;}{e-k;EX(|s4MU#394Ya=GNc@AbD9JQC)&xoc~cODe}vFrK{R2fS#cCL^GOZ>~bPQt9mXNE9Y+V7yaL>)GLV z|1S^)OH#jPR->UnidkCX4Ve77t7C4E76*W_aI0qv`6t)r`LYVnM{8`HwFIxj{$A+6 zKjEfnuZ&hq#a0gQ#`ye;Z=G>xx`R(-dj!*J-XgH(R$;OcP19#zL9V>5QNfFeQ?&b(R^$n~APUiop z(S>Xm)l+WrCOAgos=$(cI*wiF(@kMRtZwS^*0ZUP%Y>mw?tsRQxxa|os@1HPArIfu zJij+?(?L|cx8;J{CYU>H9L2SdF3y2^nY=$lHu|F8-n@-PR>nAnuCu4O%5I@}A-^lN zM~G-8<3e1*6j6hf)#{9?3-!_o<;?uMJz5_Qb%btG=vb7M-p*-9D&~%H9_0PWQg1-!H5IUtWh@ zM&%O5&f9u0{edd!D7I@jtiBmJqMXkG$M(;5s{K^AUcOQRq`i-iU)ucd^JVpO*sX=r zM^GR_cng(nV57S>ogJ^&SSb6jUNs(UfXPpn&c*Tzfn1k z8eGnSLvx3E0i3Tw#n=HE#wd6u~y+EKeHX(a0KOD=e7i<^#wifX{Ck`hzUPz2=K1=K$G!xyDr}00CEM_ z5$2~_fM@&hc=!p%!zMlO%Ks8GisKZl=~^1kgXNdRU%p&)_zv8EqWJ$meSS1=x@i2o;TaI+W^b?>(D;35g#Y!k-Y`eB z@S@b?FFP-4DL67V6Tbvx?z^fAb9Z2N)HUa&#;$+1+fJC*yJJ}egx-K8ckf9kGm$xb z0{;qlLZ>|Ju{hE{-1h(IJW2KX2Z$TbnQ@oOEGtd3?oW? zr1J|29vu7RS%FL)8ZR8j%$!$nL&*{5&VHA+yqOD--D-9xemBgbV`4ufm(ZN!<;w8* zdqCQmBzH(p7_~S=Pt@}pp#I$@DLlV|P($V0&sx6!Q)h6@)?wzV8}z=PId*{=i+j%A zzKwU5r`NF@GnMx_{u1*ChM(aJ&&|xE@%_P`!r{AMP9|n-Z=NLGbIpHyEFRTp0QFn9 zv1PvY!7&Kbz@04-<-sI?Nxuk1q*+ygMYfv)$OMBo6(3@E}{1V8Ws!U zId@7IQFpPZKbyG&X+@`|D1l`#F8YmWl_dpc2N&xfRsTgR7SE}K6`CGAEUC(KM{r?z z!RFYYuO@9Q(0{ampJlf`s@FNLs&}4_M+$_YcDtoL>@eWS)=n;_={&&9*5^UDBfC-k z*q7jjud@i>x{$%;tp_BB8&;LG7+*MbP?(tfN4 z=64UCb%0TrPc4seK5`vqZPcQQmFW3>)F$>sj>+#p9KE#jI?oRv`#Ll=XEM7poz>pwTPz_)V{glfbVJPRcAW}r9Z*i(HPfQ;h|nC1*agPzXV4$KSkK+3 z|2#La^>Fh%^N+rIG$(aZ4&L2@MV^yxU0Zw+G1laJ*sUcrnVxjYo_7HCJT~LpB(;Qk zq*&5~#LdxQ_NE0xqBqQQOEGU>7(lI6XHrMQ`e3#?y38VyPB#jT7HuA>*4~EtbPwG* zxs(fo%`X$Va!YXR8{e_LSEl(#s0o)Sa_-79s&*k(6GODn%-LM3fDTQc15zgagWjiB zq2<_+N%tyCxoa`#T1sW7_W2emh%72Jaj%uZ&A>~+{oB_SRL3`~C|E6~m@)b>Vm zmvl}&kbb_D@!98yrgoG(&-<;8TlZX{v8?JbkWD2z?iA4CdN{^7zQG{PcZg!`+;P0n z5J=*~ca!v}usD6tkEoD=MxX6pblet;&6CZc^jlXuaGV+}pSQ(#;27V(6dZ5c3B)#K z&D-|Zam=D}Pi=qc0=}K;wfW^QP%G`6e&{U=A`LTKl#KWSOWFCvFyLw^j$yd*q|@L$ z)L%2uo%6{IdYnG~I!PgA>%61s??Us(`8I)dEr`4& z;p;(B>>M0wQ%XN568V3x|EGP_^|tRG$EF}ucS$(7V6_sexXq_J+WBydhEiu|a~aXR z%_m!dmn&#${rAf77q3urhU>yhChWPWhP4(0sk~?ey`bo{x&Ucgl6hnesRMb z=h5O|d5j;@%xIU}zMKU3G}mAD(zN=~h_Xk-glm)!h;d;Ava%f1``mwI8exU?PvqRi z!%-pVSNgMLhtT6_;!r!c#OfGK)IL1tMmreABtrf>yro$(HF32kOin(T7;wNi=1xj9YyXh%(Q7yZE{4 z6*fX)}bH}j<&P%9|{S!w`yDkbsW?1k?P{-G7xK=HM zvV_iRm}xTLSTuJ3wG-7sF=~10tp2opi(@G^^Y!9c#4$b`v~rSKq-(F^#K+eqrc#Jn z*1&#eADuorTId)%`tdYf-X5sZ*-lPTg8J_5wg~e}3yFZ-Q zG#o+8nz+AH-E`{+W=#TDeaapG`J9!D<5~^{<7lBcm#sIDu8uG5ce6jX9nvdKg>Vb% z&A2-@Xq{4KPk=*DgA?ST?1Jsc0SIH0B3+&RAO_Ndm9 zZIQKa1PIMVq8+-L(B`k5A}}D1V-(Z5bmE6NP_N11Hy18Je8avYk=3;@<;Irx=Po-=DSaHHykssn$A88V(0t?}S6|oHX!Vc|huclMyf%(W^nBk^juEQ; z!(eqx+ZScm>e>cg8Ha_chm!7p&Y)K%V;ggW*26%4-wU@DEWbmry~vvX3Bzj=&XMbs zn`wD|s8Bk|PPU$aj!oO1PQ3pKeVWxzN}OqNBB)6pyIMFbk7`S|Qq*P2U}o^{#gr>% zh)A{icI}7*2WNO+2pb^g<=8Z z-B0uz-{%PFVT5fQu1qldKyE|xj3%UU*_4-@wMF&%9VYxkj;NoZ&^OXM3>w{;Y0~ChEJktEuXtj3jOKXGt?aoVhPQZI0CgniUM- zNwu6oLw*867fGXt!YMoL@+X6CUKVcm2;Grz`>*-K?0Q9y`1{ZPiCd-neWT}v9>!na zag3L86>p6l%wyORual^Q;kSC_b5k??7=dz>!a4WoG7O(B4Kv$E^Os@do#H!w8_GZ1 zUmAYE?ER+|GVKq_le=0F>BFU9K_S|Cjqvxh|LkdagXWYgMRWF^0P^S^9fBUsuZDsM z?Yph%boL$nb32E6%cWuSPmH~lu6LaWD62B27E`o%JSvY58``1Li6(lUScv;*&^{j^ zd2NnZ4nQNvE;Kjqz<7Agsv%+8=oiA@PTHV2M~gcF#JQX@Xk|r|zH4{WjDvCP!>4fX zgz0&0_-_x=86P(z5|`>(9)bl7hAKBdcoq!Pb>94ERIvC+{F~heW#AEzw_N$UMGdpF zeX?;izix;?_G8_p?IL*8GZhi|_)8ItGA-tF&$FZ7BH^kI%j)Ro+;PSM{*7qpRP3gh zrkiX2j8dnPV-PI2WoHDeG9toVUBw0_ZnRwJm$v`<6;%ITCfBD>21Xb^1|+;Rh9vfE z`;(K${x!ZZF{!8Zf#VSnw00d}o=-&M*)tkto7Ms0i-MJ9RU_T+iT7in?;cjimukz~ zq12L(as^-5?oz{a2M9ApSKdEjrkhW^^B+NtN{p8|&vh`fZK)Y1t21)5HJ|_IdKds# zm+4D`(ELF?<-+rBAd)mZN4J)uQGqjJt-AYwvc77sy@5CkPCm2RZ8il(9K5qhmpD*s za&yT3J}hrVku}zsSZ9UmD|NpZifSS<>w#xn>OrXgtn@Xu++WaFLW-Z&r$S%a`v;j% zMA0|>H|F+#?gFv@?B4v( z^8}W=*O{=g6``qmCoTQG7#xnE^D+IrGI_;#>O@g z@E={iW_9d^!HG}>S1#K5f`(M=qyn6R=-MOnZFy*V7#qKpe$xGx;Js+U%F z+qCJ(xEAgEIWyK@l0IcnyZU$vztt42G>JE)T3~VJ_8LEqnZ=vXd*JqtUIBF=lw{v@ z^6rBH?($uqvhz^mL4mslNBLkhiX+QCK@^rWjR_kzzQ)E)X|`%lz&kW2m>ILtAM^JX zQc^^qa$H3>>%B;UnmF2W*-ypO{XH_P}$i#GT-RtZ>B*^ zGqmh3bmqR%SA9gJwbOOHtmE&kcUZd~r8^w`X*eAiX)Vl+3R6I8xLuLrXa1SzgSphS z7B4^P|5_iYo7`KXo@z?>If*e93?ylG7BrvB_DpawfT<78>5Nt~s1Fx4!KaD&RlUKc zl+dwnF#jpSe)%dEe__{-3|#yh1Pi!0!RXOtAP(x(O<4Z`QqdXem)&&u1(^75G4<;a zFHk;bWVhZ=!f}>%StXY9z-;W=z4_?yir9Wjzk6w)GKMo<;Cc|K_wWFaT^S9(dp$&x zfzn1L7mNV^DxzNOlnJb~NJv!%5$M_(w0@p-pNU>~VAxl|@{pJnBtt;H@asX^IJLg$ z+qGdG{d16tuX-~i+M(`Ft#{&NFSL3}(kj_58JpLKfjJJZXnsAk#rKG%H_yVt!ZiEC z0RoWEXijJGccEpnSVK<6L0Bo-!zEaH4J~}Uu{F%#C!$zK#<*@>TVEyyuibn6=qgLC}c^u1ep7nPo66wCrA8ccOvxmn2L#t<=ymT&j zkA7}b+3TmG4s+k@qu&JNLUZUAjjT)`G}2FaIORiECl;1kWm$@RXnOE4lXKKoW3x3h zjN_JnMqGzUD?1+w!~Yu|j}71U8AnYhZGWqO=)N-4>kYm9AxVqhq0YTbO9ERl9ByNW zLiKDOH~K5#^Z3`Azd6CuRD5@_8BOSql5F_2{%`v@hLsK#T^h)hc6tE|<8n16uGs#pWc_}{VVe_>RgZcfA5)l)Lv!o$!zO$&y_G51+&EH3 z7cAXA%q7i3h52oF%nHv+0EsIqxHV_ZPY{{5*xA3~pYu2?a<9q%;WkLRjeDNdkLmu- z3%z_UwUzFt4|Pe$t9TKRb>NM+;P*ara^sif?)U>L>q!l{<0<%Zj{r1Wh@S|Sl4UCF8ZP2p67HL z3#QFKG%_-6V+h6YW@AbmL|cvoG<@sw%*sbhSB^4r=tIEQP#`ICJx{I9gB)Y6hf%MT z&|lLL*9z+=s8&=+!(+SxrBBT$mtG<{UR!6fofIW~S2i)lO2=}R!35HZ6Yi{aVG2W2;Ybo&CuCSn_J z-|dI4Q&qQLzQlN6Yq+VE{0=ueW_DlMamAW`KP=FEl*#pM=ioXh5ff-cvgal#})I5-;g&PeKBm* z7P_4}$}UPiL`$Kv@k*Vv@kGxW-_&mYvk%CLTPZD1I$^AuZRX)pAZqHKxbm9l1Ei4S z9`&!k0RHF=!}&H-G{Lp5QCbYs*{c}d=_J%F0{z-M4K~hPgq6tMw%Pm`U!eA`{*W9` zG7P#6yXmT9{Gx%3szui$yHOsKfi~-MGTr!*2F@7BS_q&;$scx>Vc7eieMU(t>&0yx zbF_-v*Dn}fZu|+$wVhv#(XbHr=0DDBe&9pH03sg4DNgJ+F;w1Sx;E~C%&wgKfyl1* zaMQT}m|(xd{N(!`)T4a+nah#AQ0D&dgj&)cG&V8IC-U$kj=hsiwN~5*eIY#S?&V?i zXeqWb9NH0cXrQ+@@4C7$;Dd&YL5LeVRv-GxP&lFeD{s$lst%4J`N6z-kOS0x-9gE% zi$^c^tbaw2!1OLw-&HWZ-W|PmUU>O_B2s`%az0S}#t8#RWr|7Qdih`v2&;dEFhkS9=<@Z%Mbm z!EFw$;$odgy-p)C<7I}&X_c5=QNSvEv-&T#Z=SrgtTX#(-39D-24Am5?|*~{e|1~- zoz?}S?d>hWGlw9iCH>V~b|Mf7*?W&_SOK+aB7eEK3i|itD=B&iqKlw1 z;IH9lpkBBbHMflIQ;O|hXCtl)IOcjmqx0Ko{6aJ^l4;oQmx4OJs&twNz6X4ImmDcm z9wykHRV)c%{hDD8|FY_h=^W(EF!~th(A6i0nZkpIR1`MB#5I4{g`;gS_n|d^*pTLL z!k~X`8Al-Yz8-!nW?KIy9>>sn^iT6b0h)^S7~mSv zdF0p6XV?0%eQZ?lHPp`m27(HeoR=^^ifDPwRhNM6vsq?$qM2g`jwxry=>Xg^BJkI; zvsYQd)N5SsxX=!mxG70}+Kk0(yOp*l?eP|&drtgA;lhY=jJG)z70s}3%m0(#a(0_5 z(?J*!QjD88ip6^dl4RT-@1yBY;W)XAWzXf(>7D?kJKA;SIJQC)ThYyaUL$DJ?s(H}^q;sZ?EN>$;&APcB}}|45=po`2xDT;_HItxgjO}H z_zjq=|9M?sLRpLJS1gXu$0Wn&r*MM4e}^u5<1DcAqv=?^QF#^?Pn(e8uB*Rn3uIm+ z`OCK(fH*mk7gvYrpyOXoJly$d6EyB-d(f7YYdfnO+bF$=WwhT-nFrJ*u3o-~d{!+$)6523&JOP{1^Q2;vJM6$Hv~UkK6S#>w&ypH?P9=KjXFFr+CFA-PskW&qr7q&g@2o zll28^2d|@&&vV}c8pC1fj=lAvQFWM2^%NKFrrE)$S8wIymF8>EW4>oI_v~r9@tZKu zV||eLKj()4C1~+FjlYVz#U9*|SrP)G!g_5_l{vs~Ij}DMJpl;2oH{9c=hULRsTfeK`ie>_0;k%Tt2IiT-!!b6QL(niVi*Hq-(=DR);3ox(^2ofbv!mj16|$aH;txI(NF7!?Z0K9^X)$3##W4XGq#yz64!{; z-<=oiEF2oYkDdQ}&ZXQ8CwKu* zdA*-^To*=NyuDu{TD}2NJk&x4-DP27U1_|q33iUn=bx`?lUVcTf4%IAdbbF*RKz7N zp^4h%{qnq?c|+gE3)(43w0cWWqg#Eh`(+m3Zv+fxzji{i;k7SnAB+QKx}J$_@(#6S z1dmiyy1|lx(HFim&!LU1dDkLN9o0n(hd7MV#ra@9?&whaS2|q{l((0w9Jx;b`Qd1B z3z*!_fSL ztqyeot~lmZ(Nj#^{4o7D@b_t4EE<^d`HG&WK!5-Blgr<*dI<#`C*{9oq@%*iF4|u= zVRdE6?-k;mo<-r-r!t3d^3w6HP}98hrl`A^9uI}{ugBscEFL^#Rb`!Xzm@Je=0eBV z0&}tb)Fd7&KP#aFb5{~Ve7Dl@3FvqHj#rM7hR}OqseN9Iw(rrBzU$eJ%yex1E8FY3 zX+MD}mWeNYqe3v$^Q-)+xF+`gdQ*J8i<6EA2h?J=pEG;S={_fH9B02a_z>k-cYkX? z%m$MJJ4Ibp8i2B0T7y|>2cqPMEbJJ?@&P1Qu007Iv5r0duLeT$cYhIHlndB%G+Ps_Cf#6;B}n0DL_axBJDRB0K)k7Eu#iBeFtcaxpH?~(lC5; z>%6+sX#=x{>?0l4SiU1K_YiMP6C;pL${7cZ9R}h%x2&WH9l%R|Z8m*L&lB#C`r+3j zK}O59c3a(}*ZgHM!=&0n7|v)~EPU6g%AYXpy(xV*0Q1kg!UXX`SFySkGxExtE^O6> zVMS`+#aB&exl~R>3fN#lzj|091>+ahpE)Kwh`Womf0^s)W0yxTVXgIBzeN^}koE}` z3ee$#U`fKvIjWcTeuO#SK+E07p1~YzH#bjEJd9k}I7R$GLd%xZT3brZQH>}An!G#16j9N`@P>!n5{^?Jl&QLb35cURK&-yb z>Nio(b%Zf_Il!X3&`yaTVo-ZW^taL5-9XOk3Q)YL0xO@d+1}nj+i$R#c<@8rf0X8iTDL9HXh*5*>X{Dg zx%OV59AUD7IqhWYAe0Y9x?aZd6*OER;PEfQ1iBYsWUm={e|7-E8%XoA z*BY~Ly~AMYjo~;&bQ&hT{)q1&XQSWW3?$iGG5dn>+%oIe)kKv4(9}4OcNT`;5{HEM zx?=roBfoy)vWQv)Lcfz1@=@CD&w3B`VR5Q#W08#;#J)oDk0Z7>2Yq0A$~2JcHim;K zVp6r1yiJQg;+RX)9~3MWWA9^A0ma(1GpL{AcVVCIf5t!2Li*mQ+s|&H$(o3q(nlw7 z49k~xQ?}FmbTpjdaJtoIFLc!Zm@Kq=hvu`34@>0%Y8-j=Gd{uzn&;kWn@gGD7zK`b zp1;h1x+-6)8h{X5cvdAn#E}a-abeuvIb;tX%dDd%oMw#=BNIa0|4|?5ln*@UJW0_C|tUi0~!7gs|>8sFP zaDaPUAH%PY6m?I>4yMqJQ?uloS%`zvoTa?8%(D@yJ7Yptv^4md%equr1SIt zKmCiEFXm+wGXB&5_~&&o3Z468?$F~YQc_e|Yb7*b{8MJ>)uy3tDj1Tmg?t;#2E z>Zx%k5oTo^^P}-|U@~m-`qK^cb^N;g6#PWQ|A~|O=lG7U*s41hJ;ANB+-b4ZY!h8S ziIe!OnqTLEbk66XKPxTnfQGH2&wY3k55M|6o{G(Urn_$ST+H!-fo4>;Zd7gW_XzZ- z*7*}#!DATf;!fLHmIj0Dmp+J`^FvdIQx>>R>_W>2T2e>k7ErB;?OuW)A8vi`ho6Ek zusFepdBXhNTiQ6rT2IF(ZyFFu*Y0*sbafu%>mvrYo7;dW{4lNc8eQJ^|JT3JS?RJ| zV~W)c!lz`izCDHULR-3-XWs3h^OLdna!kb7xkH^Wc@#e>G&G4;@XNm}W#>_!kHU5F zN4C(X5rdkVh7jR(w+io~E0i&Abn(8M08?b^E6j0cVVq|cpP;=S>RU;#brzPv>n*Ff zLNKk)8IJkM6%&UKpMkpYMm(xF7UtqpzN*$+1Jxua7k?-f2r-{-?ow|6;^4BLKxPG0 zmsJ>h?cRk3+C20vn3|#O%cfMRxn7w1xL)V%&U?tJpXj(4gVk-DIxZ2r!%QEI-*wwJ zG*JiBmv4QD3h9E*KyH2O2giZnOkA!uWJZgg%H5krzoT^RyBav@{V?py*J{Lh2j$)K z<&}TRjB4c$Ug{QCg^}2@xJ3nk(WF|z4Bi7ksBue_)xpMdskYsNcbE?if3%W2?wo=~ z%~E$8Stg)CJ^#rL?l_o#9lQO(qX?M(uA;p^_!LZRiu=5+(G&Wa)L*{X&j=Kk1Mcdk zK4@YeGwZ^)U$DZqE%4LeB^YNO)Xuok4)Z0`A$uc#qdGQk8d>4g&~vP&06IrCPv5AgpDm2@zW;OW zHda?<;rWFq!L#XTbou3vplNNO%>MRJRab|Z;Qdb(kL$veW=A8}cFeCE=<<3o#(?Q! zQq|8^M*0Pyskjc-4Tpa~*9+3a4fcOf*A|nIz;A94Y4h~$dY%~=+BEXw)v^(yK35Vp zI;M|0P8gXwht#f}6G5%9gf~EzESKc8e+-l#_jNsHGZ5L%;*jGP%&)D}e#p&k)B&W& zV-4bgCa695faQ}**ttF+7;=qWRuQ!<@3JZJJObX&-u_kU*$Xkl=H_SA^icup2wzN_ zA&$-b6|c{SJRoO8Gy1h->v(23*)}U*7#4Is?3nsYfU%bT0*l9thr^34DE|96Zd1sSp4-6#oH#%xvhU&^d*YjGXsI`?{F#5d%TAaHx zICgUhQH4{JGjm^~FQsoj3*S2oqtvT=bwe7_?A+mVS)DI|aE>rKs!od|p@E^e(Td`9 zAPWS((5S3~h4AbX>~9sJQ@;7J_BL_U&S^{Fd@qigualV)%rTv^5&zVb-}gg-q}FmV zwCJBYr$BnGmfi7U3lJLE>r)As4jt)%Pw8RCG?<7TXYP-td+)$(ON6I*BE3Eowf@!? zWlIgZ&zCjzd`h#$|Nnh{BZ~EmFP{)rM+?%Kh2o!G1PY^5?Z(cZFu|j9VuAPwl{*xQ z9MroD{a>%J5QoRmxZ}CH5l_sHT#A;N5Nch4<}1=S`*UBwpINb#!7bSQlY2|{<~zQI zdXvnef`BnJoY~cVPNp8EedscA;&q2EZ$G2=?|ES2!j+6mrRivnN%-`Z-NvZ3Pbhc9 zZYRtU_I?*+>xRAyoEzD{%fal1#FKWRqKI@Y_k16N8+`cth9G*t8YZN!+$5hRLiHkrH(CNvYjxJ}d2|o$^B#wtCP({^JeS`9q)zkirkfJ6eoH93_*9g`_=bH?Mqvd;4LJ`?KRX1oi_!in8C_84 z0p3Nn|J2KXc=Z>(k$33y{}ClL_`$_W42ME|d1wDOL(DFWAFJ*!&ZfsnlLM@{lnt=B zaEEs5?LJR8M9qK$W|bHYy}$i5;=N#sXW>Y3!kj!x8C zc+Tt~!wzVX1HQB{te)mrthoy;^H6ZeOhtv9mg<=eRwk}l6f;4jaAaxcR~U!?~NIl zFn$Tt$SAS1Y=Nk4gWAVkM}EP~i@R<)w~e6fU9ILZUKwb3jXPLAx(H)xfl(DN9Z`4U zslosp9}2uKS-kkX9zAg={X_bpf@TM_gm-4TpbjqTy@W6<@fN}A#amQC5%f6HtzZK2F;!fE;eyJg5Fy{oU!;336uJD zncuIZ;8-3;zxjaG4FeJZ|FB(59>`58TpkJqbn`XOZ9e(x0*ya`W4zz~YWHX?ke}L| zwrev)e_Jwx^3%Iun9V5e=7luq-anW1YP$%Ec4}W`+KI<8yeiHv61=+RCmOy!@D_&_ zGd|;mER<2#cyRK%VI`Pw-4|Hq-44WXD_7yig{a}9D* z0>MjN-e1HFCO@!7_ZH^CP*y6hnzae!GI}P|?z;f^!iu~z@2D#`uk<}XmZ2%7~;fx=)Dj-^Y#mdmm(bf z0UF;~V5IO+_y#dvcrn+|@~!bA43tb<5v@E7eJv*$)L;vm-TQu>;U!^Iy7@=nqHQ4b z3(GfmHrzpNF_O_u6_|bt-d}1<(LD<^y0dw5wg9uQNbIbMKS!>iR;G7P9{s%wJrP|O zgkI3+Y095MR`wQl=q#kpO^UvRNl&zCL$xg`3P`^Z;%9_r&m0b2D5b^25ZIg=L_^aA5L zYLsDQ_w{O}+TXA&&qf`U%Yub#xUf7PbyUtFrT60dHJIFTk*n5Y1oa0rN#B(#hUSF_ z5=kZUF#0kvqQN@^2FRy#e6nI;!J=E_CxbGO9;S)1#OT04rwxDIDyGjk)nON;s&*H~ zB9CjwycPh`5Fcpz#lU#?6oP43oL$`xq+K)JG(NhYK%Q;kD=q$!TfvUOa2w70Uwg%(Gz|hh$+6do8g~v z0vw~rM3~j|y)}DN*FFC%7XO}y`K;@NTE7s6?@`-QRW&f)C;3=t_roVdnD~uqJp3}zyf1G-rT;vT z8HBbaPrL$x?WKo8&3l0`oTRwF;xg*Jk!tnvsU?iPz1s1Pg4NGxR=pT`UdI3k=`Opw z2i(!ftMa+Od}`1?M-X^9PQ%5aIcL_Q)(%V$m0+|ilPP5bg9$pk`ah4O2_YjE@-# zlLDusm7r|%gQ*`mj?mG>D_?CKj$>q2a3=~p2gKiIhh6SoOrWduxjr}|r~G)otahlgX- z-E!75n-gZDjD9=~EQaF8k)2s_i)g?(eK%k1PSg_W=tpIz(*uHu1&JLI7>*W*C8w;u z9?C^4g?sMDqU}KRKi97Q4dX+OGMs)(^ee!|jm)UAR(PzpMrSgH89T@Q^wj4D^ z9T-uw`yZYAamg2{9saPWv2nAfDJ@TjdiMOj@6|O5^G`Pqdyp}G&hD!G8^Qx|bp7p< z-1fv;7~=!>pD?)>%BX-weVr>Ryt(8&}gxu z#HuuAuM-U)-VP5e#NH!k8E+BJ4WvzvG87&e zLf^rfh@9POKwRSEQJ0f~&vLtL20NIrcBr}nhbq${50`Myyr&Ee6wanvzWoW)#18_+ zX*8Y^nkw*k$(8X1N?G-r8A32!C&E<~OQNqo;EDGhOzx|PiL|Tvq`=28jxwnC*iQn9 zg{#($RR-0b|8YqBqZt(7LiJKE{D8LS?AMZW`=Rz?1lfy`2lXD>pWwuD55~le9Lhdy zq4jTch1JtIg;qxiCJ!cQA5{^8dB%?>W5On=vV`b8wb>q)-v-QHaLa1izq5e4~gnutN95yMBJ>R;AM}PmZM+0nPLg!d6A;R|kPv6+QhmPwy zzXX?ffO2(8zeZmVrsBexWRF_FOk&%y{(MaDtL>ERg<5Y*AZg<&vhPG8O5Pyzz;XcU z9Fj41Po_N=Dyq}_;U9q26I{Mw^1?rg8?^>krsw^Bgc^3sD5SW!B7DEg)y=p)K(u6D zX=FP9quUiuoAK;H#M1NYb+r*tYiEonwyQ(d+mKt8mWN?#XcJ3`5wkPoVt230Ho~=%Zdy#_b-${Akc;;nt0V ziqQPKy4ta6&7Z3uxwlJwZ9EjJzAx{d2EsNCN=8yNY7{=sswY^B7AS?$?8E@nB~Y^@ zm$3mFnrx4XeC-5EfW+}nybLgY;mB%o>wk$pSoIY~POTU})X>_JVWWHNYFJ zPu7+`3S}p4D{G3cL*K;%_MxTUVYndKLb6&FI-f6WlQqF~e;W(59v`t>h51W&j28TO z(6sP-ah5Bmk?ZI+x8#L);K=WhCG(mt{(vT}_c!NS%>vP6;PHU-1~e+Ukn{3?cqDl4 z*YBn^55ZUKp;`wGnqD9bTwn>+tL%kVf~;J5#sjE5(WbQTRuvjMZfoK^R*qwMa%AXw z<{{MbCiVx1;3c~I4j*P93Y%d3w1{+*O~=F&VZOQM)$!v@FmKP5^~qizhPsOi2MShWKLv6OyU(uqAzJ^Z<=4vV7| zWL$Mhc&H0~2GgHSrTbyll4QQRq5$)rH}Cp*Ux%qn)rRWdyWV{v3i^2)2e(rrT?kNLAx(=WQW_~15bfMNa&2ozDu`$I`87__#Uy~7>rLW z%NgHAV;?m4No?K$C5Mou}~w=pX5_ccVV z-ZQVvJPo6_-^s!jK*F(uGX8pZfS0{#Qd6SNKFcXl@kR4tmk!KC=Us92NI})Xd5(6HDk%Ti;Vn&j!(fzJ!}XxH z`@hB+s1GgGo_Jt<4^oo8r*`MrV;nxMKtkEB8RP-hHFJj zR0m%jek`9cKO`T0ghA%jQsJwUK zo`e{h?=5fg+nxgRr>4D(BYdFu>twzn=b9aswdI(X2d3Xr9cCF(9;=9`?t6tuDx5&M zEKq%t^D=xJl}W1J7zVvFz9TxDGEq~p1$%(cS(IHpszaIAgz4?qY|9lAfV|~SU8{^B zDo?C_5^%2rk=@ybJqIx!=XBz^@)s*@Fnp;?XrUwywMP#n4riW(0WF0cgWWZ#EMmo~ zAZ`{2vaJ~g0zt6whV8?I2M*{csYfp3)ASXRYXeZ!!E{YS10+&XIQn*YSX#8(~w)Vt9VQ_TO=GeObbf4K)9Yoaj*yIrfi9x$P2 zW`$>bj^c39a<9jYpujRULHpnOLTK=?x{!Wa(?93AZrwB!{{b=#p6`=9(DdK$nL!Z? z!f5|47`PNFZu|ZRw8H_0cP7RRtd!@q{w zjrx1UQ^f2yz;d9f+0H$`(Uj2z`#qW!h$!Occ}e3fD$V)ImTt!Y_J;&aa~lDPJtXwvmH8eI7FVBHA)ic zT(kS?pDD^Fp(XL2U&@1g&?WL^qPlw+Ee>o7rX0G7M$cx6a9Uw?7O3GOrALg?U^utk zQ`p}Tk!p4O%vJuvM{Wtj%gIG(cIE-cwwzd)Rr)YCluWayQC&)$;TEk57@luC(OWEl znu45l2JTkTjc564cE{Mh7L<9SRBS$I5&HUL7eq+D&~MZD>}LBcP(NiQMOTfY0)D9w z+g}nu34V8d*`2m;fatUGQM5G}W>q(&e`%^g4H8=7hEF)*x5n`;e5)8vrtXjRH}4=U zf49tct2xq%rq6?#m~&Y-#a@9~nQZjA=Lsx0S@Rqw`a+wq-Q#x3FBF=3g^_bz9Q3Mq zecNWV8*(31+E|z9!er;3sRmDFx^>#XRW@2adL4dyq}@^+>;{Uvjd6YdHW>P=C~Xk! zj(P+*26)qgq4WM*$KJOG(1O3|IrlaK^mYsi|?mz@Z+_;CW@zi~Y#Fbp8Fy|i%D zULxkF?7uJZeD@m|*1j^<;8TNo-TUPq`Z}Y<05<{a*L&gC={NT;gcqT7DZ5AZ$1H$& zZ_{y3lUf*VF%SEe+KB%A74SdhqzCbYdi6$|yRh(F-H~uF2$6b!@9mik0!s3r2PJv{ z@aDxF%u3f``r(x`lcqdS=OmTS91763Qdp(!<^dFi%lo-@et^R8tP@-QtRl*drx6X_ zjW{;H5e17=$xwb^AgOhA?e~1D9D%_P(CndONheV~#*!d~8{>9>bAS$AYe#Q~6O(w9lZz zg?q3_IoxaH@9GFm*#>AX3MUMF#;G}8Td(Jr(YXj&Jr&jbc9Oo-O` zU^xEN=O!jLOk3!#!>>NQoa6oS|6@ChJnHy3?^ZKRIEoJpe_h+>h3C-E(q6RKWgGfE zMjR%@bOt(;+ktew&R6y{9S#>22eBXcC0I}UpAhmlw(fuQhsnEIm2Mpmrn{~`O>m_2 zPdFN_IGIb?jp-%KI`{O?Jp2p;(e`Gn+Lv%l{M4*>1>8Vbxv;(L^9b~|cS$fSVegsV z^OC1drU=2P2$%24$j>l~Gnh7FxP&PEyIqoXq+l!}cI5D1J1B}{vYg$ehy3q)XlNcu zMTyUg^K|$3Lwe(e4`Hb@Xyoyp+Z&bZQQjQWcN6g;l(e9#=pAqh`XtAzt$kBqc#MNn z=GY_n>-i#D`SlA#IPmoV`=BroN0wC-{~UqY7taz&y?H=Bw)asSF%DG*vA^H%Pz(KU zUMVQuIE3acO!f$HZ$nKh1Bt9(>fz`32E`9A84;FGR;H#IhPh8V z7wuJPI40E4!9d1mV|9%C^zSKsGxdP+ttlsyKFPo=XKkRpF;*|D-}Y49k4{?L8M%Jn zy)^PY9VU8|x6~dKK%@f?pI9e;gW(su4oB#6L9cf9b(0`T$O%$Ab2;OmI;FJs(hgM0 zUc7h(7E2DEzW(%|dTTh&bw-wc=dn6nr0bvT{?;CY#U_I=P1h{C`CY6#GNpEormuvG z{>J#T@X+anp#_bv#_it!Q!m0i?wWw7Doo#e?y{p{5m7%JsXcS^36O4o*l=Yprk6~t z{PmJ0fyQ3}%Bcdu6#IiPSNrqfp&YBVdGug>ey<13t<1hwyCwwl=e8bT*W8BYJX4tX zPb?t3yid36?Td(-asAPATMabH-V;{x8q=-lmtMrNy^2H>-)5ma^3(qu$7(lJIlk10 z@VtJD`1`bViaOIb|LUov!{N}rA7cF==$Q5m%04}ObB|+;ZX7e}x{A1YULaMkskdf4 zB8aeIQ9pKR5LHN(#~j-52#vpK>#zK=_P+I28`+U~3U&WI6BeGm4-p^F`j=g;KxIcS zZ}jnd2#Fs;6%`H!qy9Xu>(gqysP;#uV$Gr3sQ=QW#rs2vs7Jy)HizRiloBGj51?2y zI5_dHpzI+kKN!r44ECemf&D?E&390R!OF!nFLMc>0BszDz56tn+wBarx&dU`BBoPrgo0TV=(!aO^Nx+B_MI$ z3Aubm9PqWu*HtUTfI3PwT||!&`Axx3<(6w`X+mWv)~go?TrPOy7PI+MDk_-LLL z)A5rqB^`2f{$LDHzb}s(2yaEbB){xu?k(t>_6-~D@(qAbSpTZ%I4e+ngoR6ZU!q)7 z@&e1YFofSBy;C#=!wq-tAuHs3x`2jf)F1e5T=R=Go|)_#l!XCL!#)XPcWnQOk33vZ z>WA@Z-PUXIqig5!}!Qq*3T*@P~L%U0>>x1p+Ed| zfUC@1G;a2{UDNs!Oa+yRCp@C*h(YItT#opMWl&(c`FX?_5x|qWzuvv^78;YJufH8u zfyM!@1MhXx(Ac>u?~Le4MEYxb*;M^An(-aAt9QKxRDQ)aKH@a=-BDi&sz`tZF4&p9 z`6lYFe!2c=#9K6ZF~iBHa~6GZFWFeWqoM_^zlhM967Rn}3NxCxYwzB^K*eMe=`JffnBH5zEv5zI zH_kCxpAw6(2GW-1Zwv2s0#%LiR_o_#818CM5ecqEV=Q__FSLi~)^YLW7N-N_hhX-x zuS)$#Ulhl_b$(wVrpLG<%4r{dMFx$Z_t+TB@BCP)`SoTh$^m~+HBRwJ5fE3F>QXi6^U<>}jFo*#D^cD# z-yq}gZGbQCjvQj!j|fuT7hC0+QG0xTvrh4U)(eBFXV#@(iS)Q)aLWIV(3 z>yRF5)K$N3>_Nxh2I67D2ks4Z=qJ0P6kpJ3)b;j*l%$0Pro+oaA z$)`^_I_(O9I9Z=|)oUl<<3-pb)fizcqu|(d&}JAHs1#u?nL%|LpVNgN*8#!BbVEn| zy0vlbo$lqL@s80%^19+tFFmLVSxj2*aRv=36kHvOh(guHe@45A*)UY5SXt^H45Tw# z?)2A~pfPQ0k;3va{l4FW_a^iLrQm*`6-zPn#89}47CnF}v!$*ju>kT9JNsQV zp`D*-RrF@5QSBM%ek*Cds9gdD`<16AwEMAm0{-1mbzclOLl88}jPIevD-d;|?-hju z)4%A+6x6vZ8w-T%PYycQW4zlzagi$_D}B(ekaTWAABzu8#=g0q9H0%gc1==j^)qN- zim)+99@0X!JHEHPsBE{LdcJ{1Mma*2~xfRS`mO9}L>7{}v6w^!V^EMF05f zu)?HrCUs9GeI4n!J=dQwS=;{SdR{zvPzV)&K(CxJ(w>A}LEa?Nv{#4JuEgxVwU9*!dX*(=?CnX#z zOF*63`WKqA5lo+Y5ZlFyLxg;;niiFxX#AC8+>=^{Jchpu5<8+Wy@}Z_#M8k40%ln8 z`;EUk0G^_v)inGJ29k2VENIZhyHNX{Y97axMu1;6&hB*#lp1sMtfHEfJl%;>v62Kc7}E9rK5ZUp#S5APPeYz=;H#*tG9(efY7>e ziswrxOiz6?KeYv`H`UN&a+2j7rqerrNn+jeNN$)-d-8p311;W)W4X|qpU#Kbi){wl zHC4|U&~M+WcA@npXAEecvfLfai{Sj$8zlYY~gZr&R^ ze&#y0Tt-e3ucdz|7Ob^v6NU?_gL$W$=GPzJ#@6-wr_pu%6f`$P$mF|q^`Gl#_VmpN zJ^SN04%I^CGQDl6QpX`4%9r?M?HqUT?K&xm^1k85c=v1pYTCuK)_?v- zFR(~CAz9q|;A~$9BTN7%osk6{~++D`{$ZPS^?t&UQZyMW2Dd zPw+j@cm~bv*_0#SL{}#o7G22E+JbwaV&;-()+d^-ISd~=!fE$H3QZ5z{5@N?X7`DA z5(eEhVN7y-_)*VRXj9_e%6ZWeDkf8xPhZCP61@V`tai-zu>GaXZtrU?g<~{xx-W6M z2_m*t&u0wSeZ~vqr84rYBU#AOxmrzOX)|K6=B0z14+ZuiI0N_nC z@^&+rp{bzvmY4E9fm(3#QNy7HM9R3OIoV_ai>YysKaXnC#-+lOHQo4=KXl-aRuqQQ z>L|hxiT7KmmMbjz6ldH|UqJoB;Xh8>alq)Ae6q9B8JL&xf36}(d%vLxZrhP2*AO&$ zID7hdI1b2r3(V75t5HSagVP;FXVIMchC6!S6tMlY*+fF#hXVKPL(Nh2Nn(6(=DlvdN4 zJvo&7SS$loiw?RwzDeRp0s8HSvl^?&0qTwxyEjjSDG&@jUPO_`=twv zi+iNjOKCvu;KcT5n^73QqNCEnl>*Zz|6F@>iB1;|W{p+?BKDcUWNPV&a$nkeAK^J1 zW4Wx0fGly4H+Ygx*A6BJSXl)2Vz|VrL{ik@O+-ZX@o*|*rqjuXg}~e9H_gpq22?KJ zWc_!Y7`(jFx}{8OnDA-38MKv#XT))Q%I}ZoJcCAM0)I#@KZc0|pBz6B3+b+ZdeMGU ze*P$$V#rvr?&L%*YH9rQebLZ!V0xGI1bzLSVKrkXW~Feb>wOs5nn)MVMI)>?7!GZS zhq+(%%#PQGo}yK|JEP&*fpdfbA_hw zj>hD86)2LcsEXrfsmJ(cMJ&is2xgil4&On+rj$-LaIeC~LX7D55RLh= z=Uhn*0gCsDf#pvaev!KS@%y6rWthnxL&Mc2Flj3^e6ZUFyB@{AyY<$DuD|%b&*z@M z*nwkeW#~Bg2E#{F&N99?ZWRIihw8s~)YkT)xM7g7Sr`n|_CHP6q@B}1>In}j-0Y5m zH!oVYhW(GOqwXc!C*&H)7%8kzVm(Rg$K;JJ<9(|afXtbYe@rhOX6wQ}H%ZdfQ->u* z^VcVRFx_Ax?^3qry-h%r%DGK8&H?>a^ErNE5G$YMEL*ndS=!9 z&tfHQ91&Sq><5PtEnf~9;}Ee!h=#90E2fV3`s7Za`n1Gy3wZ)iQf}m5^3_8mZHv^C zX$No&`+W|j+i!7qUlQoi~uhHFle()7KeECkbxnyg1XwgCA+ z_x&A#6+lTH+iZA}o~Pq@GfBVR?iOmUZRK$NISwT0Zbe16LYOE!!k%a7hbk+})TeG? zdb!k*3m$yp*t+TV_NlGaXaT~vkx=Ite~6oT=BYKvhq7(oFx@>oh-x^O;kyK&<{$qL z>i#^Os_1_ohs{$Fr6fh7q^*Dy1S#l4L3=AxV^kP$ZO)u|b(-U!HjcVW8wH?7%{euQzE~BX8hM1zN2^0 zBvNX1mK=ont+Pu)UjksUDW3=Mwyt7PQAX!f&QR9F!8I;$vag5M#nNA zZu0!{U+rmv#kEYiAgj8cZ{d!0@&Gapzt-Wci{ApOQvf$uyYCi5IXjeGGJ|ZsU&&eF$%)Gi31L z0eso@aq6@U6OSY^&N{hn(1s4frcUb=Qq7<)LjUoBkn;cbUWo2;k-cN%QM9~uJ?p{Q zT9mT>DsOjuAX<76Y?V%^K+E4(Z080Fv`F%A+;Z?DWUVznUSOUE`1Mh9QWwe?&tD#R zwcVhCjz){}?^g=_TSsYuGdyvx6$8hH%C#=BN%J$;6Zc*{b?(DP3=h_-*~F>x{WdIY zZgjM9r$FzcR6oBDA5fdcpW@%N3Ya~7>;5wf6_~RZ(Tr8Fnt@zd5@5!bMC6p7Qz{ z3-`S*C}rBUxU>)tFsSKlsG^7nz7k24vqyktI_q~eALEUpzBbqu!}|jYuOGg+-y8=s zUk|fZ7%lrVHGAG0>*uNj87l;$iVQ+dE4yFB-b(PfV67$HfKLPf`;UEd^Fx* z@wSG$h3^yN_5qPYKRC2!#n1G8(_|iw;YOAN%XVhTPQeWQX1@u47W}15C3?#Lfj{dd z)CatbfM&bbkX{@E{f9R{SpUNXCT?!@%l&Ky1URblRL~C=WYn%#5%!~s0;2eV(5mnJ=3>NQQQkJeZq8Y8=#Aw&)={McKRDy}ZBc86xlgO;SJEbdE_Te# za6h&lc1>@-lg9T0@_qxw7Z%oiAT7JS(~ z89p#^6spE;&S{liMf0a#_XKXG2Pfj-H+aK z_Y|1UpO^i*dZ+0Zn4L|$b_@E8Szg@_p{nxO2Z=1ofQStZ`7p0M$@05Df#vQ1cAe$mZDuY~DwBJiajd7UusP zv<*7F6O#NtU2wX=(5b+&HU=fRm(@ef)-~tV1_fbYb56KOB2zws@%*!*$wz;G4PqW2 zy6W|NzDH3-tF1rb?~lgD10L(o8KOVA*PSrp`3H*$7LEKogZc)-kf4;Z)VeXk;?GxP}Xd=o#vhe z>F3h3_?8$vKsYwSy79jc41pL}czm}LW@mLODP3;3q6CxXpR`5ePUBdlE1zYT`XlO^ zFwT8WD|ScgD%WRT2B^cmn{O@o2~gv&yB;WXfbk=bJs)qdhhg(+TgSTka0U7vj1R5_*b*X`*axobS{g|u~wcA27VjxTm0p=z08CGi*pJ2LYRL1 z(o`xmIKEc-e#!rCmojC0 zzYMT?tiN$Ivc#R2*3w2ji>hv!r4wYEr1J*cC`XtuChA=k0z$5Bt&edX#&p z6HgnFwUAh`k9l#=nv0?B2m7=usR0Ir`oFdxy8+W2d-}%x82cgmX!Iaro98ktirA~X z`tSwn_Q5QT@Iw|`qK%yPi-y~IEN(D)mb*Mm(-=s-F?lnp6relq{B=KEX`h5r0EQPIp7F5EYGuOr0;Rb6c9Z+R zamDnm4Yx+MWwH16%po0goJFmfY{Vmr*gPIw)ctK*W5{%Ui{xgVyyd`Uw?YrQhr% zc@ZhOVSRinBd&>>h@_Co2blg+jJwt1=;SQa|GL^etc>aGm9?qIa_igz5`W6C;1m-2 z95xm0WTp$mow?b)e&3+cNxy$hbOXtVu$3T6Mj?vYb!xmu4i`hR6T3V7mGWC*h<9+RAVusi}W)90tbmd8xj&zeX4 z`X0c9rv#ePAvMc!YnZ(km*<}G1)80r9*{RH0eQlmo##;tgoz^%%Wp>{F;sLD!;g{ZBNbCL)`6}_9l34&m7Nxw4=Q4mMpc1rS1goo=y?Mnl?XWJY-@u)a z%x?)p%6hXIZ(LyZdu1Wt8b-Y@j20fBm1e`@2(wM~&AAsRfl7=IdGJ^V$Ww`bHVTac zi&*2R&{AUaVsKaJMVWpNMuA%_Ja7^?aVx z#to1nS}OhCxf|h4N9ue9*1}xA%E7@8=0JMqmQrk^j0V{pmF^KDfxvm8O6AmTpwhM3 ztvBQXoqzL_;`HkX?`iaOA3r1Bl*=08Ih;_0#c4aQf4lVLaTU;mY{v`N$3jDQ;AcJ= zMxG0XkG%Xkb(Se!#MHlYWj9iiZ??jK%~r8;cP70GW_wF8DZuz4#>bv#uj60+oq7Es zd{wp0bl43;1@S%Wi_?L8@b<&)yM}@Mx|c1(uLmei9-^uV|E$-6q1^C?Z|*W5FIW4* z^Q@W%Gc(m$7t43R7=>r|%ejBoPmt5gy8bNFpu(?k=Fwk_?`7h(Mf3Li*gF5+yCnAZ zDa=osGcTLnZWIc>2Obu7){qhT7NVK?_+=?WrbA z`6cMSFS75w(FD_RQV*}luet~reXNkYt;z&w5j|!1Yq0NTB&p}N!&x@Sk*XN}wXqUr zel;s-=yw2_=dZSI{As54X3rET-@{b{nKPB+Fujf5rzSglM{R*xEhl_ms19ade0vq&z63Kx zFX*cUu=Q)A{CB}XGZx35ll-~O>G{roj`w(JFh};36zY5waKqx(Wgz?Umxw*Ya4xgo zWvyl%#Qt?XQ+rg=2gJ}%oY*)XIZy2Tycc?XA6IPHaS4hGqmvuyPC!$P%r@5agsRA6 zc%I!x(C@xmdTeIpd-pz7y)g;v?~;;}sY`GYHOmLCF12?6;_|t|tNht0ofJD0>+l)9 zkw1gJoi2rhCxrzALiH&BWkYXi+=?G^;IyNcG#MrR@vrl`zzP%8xNrLDYfx)bKFM;@ z9p)w2qAqkfAUCz>1l}|HFuF6dDEaC8PHIy*r+xdWc zcJOh`Jd{NB+pp+7uQNcCPZXzqT42vjJsBiWpLi8WxQc6Qn=n41>@^v-iQbqFCB<$g zdEj0G)AQ9gdA~-@5U zDi3`ZI{U{ANJbl^+&^LW9pzuX`6TBEjMccB;;L5Gfn-a{mD0ypKeUo~^{--e{-ZIs z11w8~V2Yqp5K*v->Ai^~;rUxlH?Qm$PGz4o=3u~DasB&ahG-zkG;RFi6BrVBy@tmh z^WR&Jl%Fv_h>iEaSjX?Jy;U$0MQcA5i{*X0G=J^(vBL6%gNX%sTb|5d>t@}+)Y^j> zZ)$^Rn7aBpRY+7T@=?L?4u~Rc@ndLJBjE284^Hnt4+V#Q;v*aq5uSfgfp^y`2zE~$ zE7#2c$_Jy5SKpTd?dMErdan*pwfNW%vlpUC`442MIt}E0&hy&k=BQV5r<#rh38+cW zbyb`gaU?Xsbtj-qk`KrkBN^p7FM!B*@3xl~8xS-@UZP{7Kntn4nejXcI_ebUlNuv{ zc=MXjp%x0H&GZOj$tbBLB6!^s70kMwwAtJjIYPIh@Dz)W7^m}_O zxmnMlUjH@wMIC;k=4O;hPs)eku-PkB?Rr2y{p@ld?mTK}7*sQR<_wgb`jzx)=J?Nu z*+66j{{X5LtIf`Fb3~+buSYH3l|yAd$owQ=V+#12moxanw?g0kcD#4aE11yQes$YBcGTA)ym9;atw57`uhjn{ z5r}IyZ_8R6jharr|M~JU1K)-M1`1v*Z;J+eMh;gHZ7q;)5pMnpu7UureKwpA4WVaq z!}x%J+lrrWsO1rY`8$JFmpHvlaWHgIfy>nq%WsgU?k~g=u0#E?d6_ z?tTCB{TRQopO*%yW4-i+pdy24~9{;${Fv5)J{cI`F9v9G$lIcpiyYpP8cx}n+5 zTqi&0)MiosX%h_jP*eg}?FFKO+6A4mAOmSgoTU*N-PVNHRuLrnMW zR2wi-JNySNKe(DTQf`ZyJzPUJyc@&j#qhv6JJDhw#=W0MifA z<>+O?9sXClW5ixbJDY`}V}}+qxp*T6f1)}2Pib~6j66CT&HFs7{$T?m+`Bf>wTcIt ztoWC&5wg&Hzsz)ooC^$}X1(ylIt|B#+xDLAC+M{B*{mU-EphIE#kKLm%YIxP+_v$p0c)}aNY|im zLqd}6fltt1>ef|8UJuhYeOAA`grJ4+Znw3yDb$-)n^=~1!|Vo|VGGHZ(7C5OA~C9m zsh`RtMSn~_wFO#(OU_vg2Lv-G_MATVVG8KVv66$zyeQt~eRt@EbwIqzS?U-TD>32pjnAG~x|IIywZ>djmt!G0+t6BdT zweEy|`UAVg!;LV)l@=7Qkc9>cV$FG$xnS(e&Pzvw_M%4bp?f()jJO_{rKC=wf}l4J{l-ep#mW9n;xtgg~N$ny=F?K6*_ zKl2($om+&8^2UK0XP@%a7UNwZ{Nea;KH@4O>Yd5^Ok}cqfU0=K?n;d?40{ef>0pahPLe5u_FGgvkRxo)_C={xh-eiB_emBak=E zq@Ld#48-l>)18_W=yu}|jl8CfNa8*(?3)YG4BPyn^);DL`SR7Ftku0hvL_j$hBh?n zlo53_T^jYdo?V}QK^xIV6zLp#4LH_(uH^gvN71N4%)@*i=Dbw7i`%0Ng}wh8KOBo# zFnvuc2O3M(XEEF+f&L`U*|=mTqS*qxy_w`%G|3(yC)M)>YTm{Ra%O!&Q5SEJcRxA- zDLt~)#o7^wx~8S##z}eT|E%!k%@aBrwU9j&sJRJJvYIGXPvfCaqP#75KQ_N7o_gr> zC(WRaB&0j@`tZtkYL$`Gjrl8mukUt+UU~^L4_LN_Kh^;1P0~T{Xhoo+X-_MvB+SbH zaFkgS12m4N#@dalOyfn4LGLyvSi;nfecGw@Bp_v|WM0q& z=waW#iAAykF?f~Se4a8aCVD>HfABSGH&eVqC|8AcouT-if_oPhPtt1n`FbO~Okjt7l@MYZ#x z;SwR;&`+y>UpcoI^@Jn^$+h*ubWVtw{C9q!;Q9N#7p+n8wb=;sOZ7m_dNX+Q^LD@| zRLGa_nn3Lb9|hR%%mI2$^p)QHHE2QrPVlahK%?{CTKJusXxPd@t{#u!vd7406zj-*f=kw)EnzvJ{|m&_Yxm^})EtJ%u#}n$X7`zg+HNjN?3Lb&%)YC<;h0 zx?bUg;lt=vmEt{(VL-`mmQZdZ&MM;S~j5q}oO*P!-?{1TEu`%wOBrE1pAYoJc%pi}2h87O#5dK4gq z)ot}3G9Hm_hy=n*gTK=sj-cLO=CWfYE9;^`_~cB|iXCs_F}JZe94+qOkr^~9jAOqt z7H5C|Jh%&*uOTAjCfyzjD+BMr=0(t>4JT^c>Sk zR9fB-g&%*mp^B+1Ti_Ob5c} zqJ!!cMnFx8uFKaFg9WKA-A6P=b6Gh4zG|W}@xfyImZ!&P@#YffmQc63#A%A?n+nQB z%uQfuHS)%p_F(fU)7rJdItcJh?OcB=ZUeDVY5HO|=9lM%3=NmwV#L>=ll#1g&v8ef z;}kWjZ|DG7&GKmI3x@p$vx?7KkE$Gl$=Xg{H$f(Q5oUul$a+DTei`}6Gg2E4%R`Rc z%s)`Ie;6iLj7Ybk7}QizO)3r(M^vxC*7-pOE)fW!duC5$1_M<`@L|jz6GT|^F6f1i zFKYauKFrZ^8io55jo(;jieoi!ng3SK0|b9L-kSY}h_LZ?i`$8h2p{~5!(;br^v2!x zNp#UoROWN7wn%{o&CZ2w)O&CQ$kdCXN@@{7~5xK$1Vkr^t9m6-oMGm?B`=TQUHcTdEt zSgRFHh;SU`{Bs4z*1nasj%5iZNk8$kYML-<wDJbC)I)QTg<6%tpF_CNDbdn;{n8+o;>Gy zHJDi5D0ZOG8Bv6OWt=S1Lip}c*?Y+{D|WefWLe|~$h+QnXE^}dSBllYu3LQr!wCFM$&>%h>FK{GP~2U=ytrFT^>e6M z`Q1I=*XC$Ek^AU-v>BC}R{oZ7%fWGMgWi(Unm{mL{5o-(S%3D2?~Ck`O-yz5IPOg@ zJ*7`I(ZK9S5f^U(9Lr-Fh37ZlqgF2$vo5(5DAZ5SXcKpW_VSJ^>P2f%+nN!5A$HF)ee^lN-v|Duo&29f^t^>eF;C9DHJz9>R5#q_UrOpOz{J1)Y6s8&Tr6ATGsHa zRuY;Xoql_FwJ;E;|=VR^=pJ_DVyO}}7Vzum>JX%1%k$2bq>e+5#? zTM5-&Gl*&u96+ysgCJD7F{^=XXKP+LruO@j4DIrW7hr~M%huM9Cs5qlVRdzZlPmosA~M<{fQU;4 zL%rMe(SUlx#y4VHfUE`fH=qF8_mGctT_ZvNcG@MrxolJ?a{5q!oi&>NA-~l+oF9?# z9Lc-lFuY$DtBB18F=ZI|rX3ynY(0>t&bg=6VR6~Aea?64XWt>>sx=w=_|1XVn(kG- zRtW2dbDxJCa;9MB6-%{JDxGn>@#W`__E!JKXZ3bAvY64fLGE*r$9u{&fvh?A_~H(x zd=PWo!`nT1mE=LF){5AfDA+h?n@>MXd)qe9<7Z#Tx zelB>y&v^lQB9D1kGuMG5ddl_U7hcSI_w_e!-hR6A)`~yw%eQC8UKmL^pkI-)#B|<3 z;E#kIlFZ{cw>F!7)r=Al`sN$nj64F8U7W|h4NUcdXv~9LAh~D)Q*0VeRUaB5vq@R> zht675Am~W<;=ynagda!3PS5+IfxB5z`@N=M!c9l7k9`0wvPXfXf9|JsOS*SM2rN z1;pb<+x~39^s%dA&$d`KVg1k$kZ7CW&nko^ATA;Nd{h8md9jMxJXWK^yT{Odm z?-O+lXIdvXc3!G{cmRvfO&ZXz-7yHn#y4@$z@e2D^=}P%tWpyX#H;Ex(k&zZvmK`S zIB2vgldxcpwc}m%(NzhHFSl~t3sK$l&voy=w%`BGy{q24A6t)`ui(tCV0fUEZ`;gT z@fBEqDofe@y>to2o^fvGDj$M|eAP^$jz(0es40m@$DnU&R&@ST0}Q^xU(uGIMs#+f zMWAgrR7&dNcWhgLF{^i7eKiNrp!3yhChIrAd{ZKBf2%Y-`_CeII`u(^lluRP)#o(c>?I|AK!PlWWdCc`tzq~v(SC=Z|r#d z7@FPR;_wZfM!jQddBn3uFg)Jr#P;b66wmr|hnwz(u9I9BtPagXyO7EK2jP1VIebT_ zjkll$vQe<5j0~Dtvys+JlEJb1ilyd?t%s~%C;YFUScLT3 ze^ruaMQ~ik!8}I$me8DFveBQ{B|zfcoo2bU81)_t^c243g=%}#YBq^6@@QzfJ^pXh zNHEZ<6%|SpKO)M!n|kt+94cXT^S ziijIbAnMkIO1@q!KRGh{p!eQt#=eU3#lmmr$mhb6g;K`Dbx}a&e|kn&;uo5F5%D~Y zTMq_TD@hId1)#2C>t^@FooG@+^Wb|uZZzils-Y|j<5BI~a3x?;hAAJ#_?=X9{*~Ml zKcLjRjQVkHfXcYn!tbKXu<_pafK$>z8Vb#uTVAkZy6hxvVcqGBYf$^c_o1E+hHLF> z{VI2NH4ezHy0-9kg+a%*Y2Wvj%8cK?9_kEs9#TT1b&EgAGF$$+zVzC5%`NP&U^<>+ zLUtc;#p;3UZ!PpYaANCrLuNC9eAxr0mwr-+!MV^-J&!!Z$UvGZQi&D_f&4W3*Dp>_ zQAx>`jH-uDFnEjjQuWLSpxxN=cIRL&T0E)M80a+zsrlbKR{do7^QeUU{LMwd-MK6$ z9=+0XSc}@rmiH%>V15`y;ezx>bL)S;PuCB@X+Nsjik)!D&HMqAF7ZF-A?KI(K|a4Q zJSk~;w_|!F322%T={G|cvH5e)azC;38Bk4(e(T@Bc!=w|H2(Zp^#o`Gk`q#P{}(*` zpPKCF?_=>G^0j8y8plI0!|hjbj|1bYB|Z46nDAf+OjSNkUw;7Oz35$IC;HOU8wM{u ziyc|A1iYH8;BP?;$4Gs!z+L(L4>S>fNq-fa0-8fXqKDr9S^w5Jc@G?2>F>ig%LTNd za&y7pp*4Xkel_o}jYp(`_D48Me6I+i(B*~`pO6rV_?@hOIUdOCKN-zjiGMkJiNKMHe#PC#; ziw}ckxVFN=^Pi&p%H23FxV-2kOh?UiE?kcVbYRA=ovV6>8xWk%5}bH0!bmXhI!y{4 zCLOCo(r<~tIE~+j-Ov-|q~Sdh4J4VyZ{clYG4E=3)D`Ky-7qBzrV7*rEQ)&2f=)Y} zdFunMEx$MmhknCgb9T(JF&&hVUW`lG6A6=V-Q?Xw>Y$*N{w!R@2ED6Nn(fc{3{!Eb zr+-ODqXwau<}Wplpw7q#{sITC!BFsdq0t`<{1IyX_2q_Dgc6#d@=4|^l3{i~uH)ie z4MdQScg_A=i~5RZC>MFyf#AD5+ViG`>A4HG>Fc-&8bAmB*U6A|RG8YVuK7A65)H4d zrtTi10@*U+$n(RP-pu%VpO`)WE>gjN&JiGxzh!| z#IgDgl7sE@ds102yYHCLp~JpF5xA=wugg?-$vA(4!{@`F(wAtuWcRtVe@IO4JGzXs ze&cfn{2w$HPa8SG$n;vB504o*9T+_LaPNgT7h$gUK(7C955S+8@+hkq17a@W*%Qu9 zFp^axtLaGx@>u=GL@^g2lkJIY(K}%NQqaCFcy?$mZKo>oD*yBOrba3XUpSAzxb@98 zVeMFg%0wyz8@LqcYgsws8l^r_Fv!97LUbe3AiOz@5rkt zZd0>iZ95lGLn~KZI>ruz8(wos#4*{WfVVmyeQ1dT$~SI^zHQyHGEchFgmR7m>8KRF z^fbnw*VTE;uH4iK#;xQWJJJ@RclWsj+kz*kr6JfGui%W?bA_ae6jc!zlTd5R+03ka z-DkP&wb#bgxvcuzcWkg@=(C}bcfUPL&SUeJ7}LkAeexF2$M@)`(dytYMbb^e<`G){ z9NRZ5*b1{xqfWU$UW+CKK0WTavj)dPxQv=N48vRsuS~4kRT$P>ecjo19;k7iswt9J zU}9tOqd58(AW@QcM&uJvcZ;Ly7*z*NmS+!q@nk~@W6it0ahU1Onmw0=T{(v{nm2g-H2mNP(QOFFRE>8IS=$c z!^L7JLFnCniEX365)iWP#qC?m3KKb^&G{a`(2NF+OaJg!9QQ=Sccn2|RCUYfR(`J{ zih6K3U%-e4b$=qVESb%_95_+Src)!~_pmy?5i9kw-FFgU>^HG`t_ssB{^+zy zK(F#NB1^8;{B&+D<9Kg)jbiN&M>O?bSNgZ&f8fZu4lH@LoPv?#ieve-XK0k`xroni zPAEGe`rg|}3-FtHswJzp0!6&ZHoSKi^p7o7b}l>tvZJ2^$sW_yonC+Z=1t-~)GEVk zzFcaFI(N!zC)OlE7rCwS;>#`|UG{BB$g+l-unk!q?rmsvsl|PD_Z8H=#wA#Mzzmv_ zNmZ|S>rmOc7}>78eyG*lbA5}q(OIOocq@5HoH!vxj@0(oSQ5#yTWc#F z)9Wx5pWCTcm<)tuja0LDE9+SRfQVrLEZbeWt@@r7W?aY%bc15(+3**?o+J*vf3J6G z_A~MeFlyb%CEk|@g?H^`*~b@wsN`vKO`{#CqR(6GOxghNQB+~z+>EI8y>||0e1!Ii zRJxBrA*!Nt&VM3EX2xjlVvfZ=BbHGgk5;i5vP0N$^{^8rxKD&|k~r2)}gjL&^A z5@xJ#W|GP_!&J!k%x?a*i0CN2Eb^rWMitzFm2eRTXDT%c3$LK%)0Y&-{(fS#Tg!HT zL-wuTFw7x(e!uX_{{F*Kk^jVsUl>+9DZ;M>(*hklf2Z&$t6}ZF=jj|UC*o5U5s(cO ze@(gmJ9<@Zoet?$S6?r_c7H8;n=Kg$1mfXVCmGZH7@-f zsEfT<+sV2BCVP2rY}`A7h9vsLJoRH?;#A%VSuqvVZlADq$V40PLE4)~jB1(YH-&5H z1Fpyt%0}OV3MV&OG@Pv1JMoOlh>37(0azU64yCBjGWQ#Ycuw0wS*>x#yw z7Q@UoE+Lv=+Sz3XhF^r^ksWdU5O*CBI43h&UD6SL-N}KHvA1X{X0UdxQ8^k0)w18W zD`7f|Bol;Q!bH<<^Lr|d(64EF_hW@LOl-Yk=F>$)HS9kHtnPKd-{taY2X`w99pO}&qPvIdf^=zIxUk7NDOX{)3p3F9#j z?%Qr(1!NZMD<_}C=CV>-qlR7Apq8Ldy=&fIMTA2SZZ}x3f$VVoLmv7ZIQEK%Eb9_C z0O4Bf8B3lq7{$@W*n>);@pDYab!8@gW@rj-OzJ~t(de1Of@hxmlSk;E_DMgtuuswh zjF*XcAn>lG#T}R!DeP6QkpfD4Q;*SJ0d0p? z*ReHaFiuoHO_UmkvC~&F3{@WjnO`g`R9p^-8?~g4#nIr5jTLkPAet7 zqZ1mR?d37tM}Z0Zd0F$SE}*QMIoXnN493u1L9YXh`UXV4{HEJh&lPBn;&JCHusWSO z&227a=M;e872YsQorqJ~engx$*Gk@i;qB;JPMO{!z(k0dmtOq^K*v)_A1SniAgP5IKJW=UaGF|tD}<2Dh| z_;~;Lp4tK(((6we2`WG@uh)fhf13t~Oi7n+YT52V`dG;Gn%9q|e zJekfoZ~khUu|L}*G>Mmcb3Qp1g^7*2WjzT+G!Nyixqo*7#rdhpO9_TP7#iQ8eet*x zW*-tPZae5v{ZaB=ZgS;EIT+uqKirxgftpdzI=S`x{@G8XDkn&k{Zt67)9~FibONS# zoa2(~F@b@S11Y$e)i{>0ez7Nb8Z0DfyjG!&0zLQ2sg@9iKa6?;UMBLKaD#^Xa$7J+(hW|4i=Ws5w1(=HA-KxTa@pvss$qvg|ra`^){Xi);DYTfo{QQCRbr>=ZT)KVA8D{*I zHecV{0n==81Y^IoXyo&b`;%)kq2uDC%>~zaU`#dY`TCR|6jZq}l)j+|G8Ga`h?g-Q zCbHVKJKK~0;@EkLTYgL1{j;B&uRh!|y_KO0wsQa5qxgmW(04-TLA-Dys#_a2wCwg9 zvUNSx69aNlN?v}T?|BZCax>iIY=k&Wd+eUma2bK7q%OMg)EH{56&V%dX0jug*PBR zD*(xua^y&k7}R;y@4S4A!J~%g-1}uKl=9Jt@LRF{8?m}F!ty02(v%8}f6)F#6=Bws zrn3rmamq6MOO!DCcuuIonCU#)sR*;e&?@MM%KL&Ry`e*&CUGL1pXt1Xe(#HxJ2)WI ze(^r1b^;pGWe;~X@`Q?FG!AJag$s_KcmFcjC>r*H%VI?g@$n)+{LHAzs5r?m(a72S`4pG&=lUkf!_h> z&$U3@Lkb3V#|LJ8U4XP7Q*GR@8Sx(+-??4y1+!G4?oAot+FC(Gjnw87{L6{DjW1nX zY{lYrGq2K4;EPVeh@0|w6#fV+ci(kv*b=;wXKBFwXiX5n9@nGXo3-VYtz2 z{PhPr`04OFIbl`>wLBw9{WT>a!sGK=Px1m`RA$!!vzzg#)}Ji({yi7!j|kBf(Uk+j zUYD33FRKx`$D;itT^~)qOE%^W6N5gV_dYCg7=F4VUpzyv_99GDkKu`XeX*Jec8d6Wl863TTKSf>&F z&xyRZY6fUtyqwqTy(z-W3_Kq6V1>!q)8N#hdn_*eIvRg`#_7f* zBWT}gw8`rE2ujrZVPxa=9LGvIeElV-AdchB8UCFkU5M()h7<~~;Mkh0CPIC-LV=cv zN>FqQqViqS2j5+Y^tSAWmAwMghmoa!^Phrt9h)C7!q>ta?^%jq&nc*p%bB%oEkyls zlV^0!Tx43OrkBe+iNLbFbP4DGWC?(ibseC0Ggsb?cr$Q)a!) z;SW*g*GBTdXu07*6ZJ&MqCYydP*N+7>p_~gxi_E8LVED62 z@@dXcH0xITT0l?~#vkG1+%9(ESXPNW(72A-O(loF?0lzp7ttJ~_5J63&=~2H)em(c z)E2MUEq)>kRrYVs89s&8=?(50csg+V3>qH}l&R?&K}*|gPZAq{z|UW$W67Z@KvoUP zPt(4RMz%`mPcEvUp$k0Aw!&dh$zj}hM?wkFrEASn%f10^13~n$_))0N#aRbN(}7&L z>deE&^9aA~(P(ym5>Q!*VMU$kEB5WV$?|XUKrzf%E>B%VebSv;N=-|s=J9yi;t34z z*txcN&14@dbl3m(WmyUZlEV6*TY?*aa)+gK__j70TetU7>re-fB(=PZM=zt~@A{KZ z56PjFRd-JvzdjCl^)uvskuoqPCes*jcM%Ov?J?K%3PuC+!mq!*TRsLDwuv95d)(?ajeJUnD~ktNtBO@&}zD-dBQ9J$bAN?>if1M`W#m# z+xd$y^yPi$#NInVym;2FSpwrbB*mNShKHR+vTgS?!bYJlw zMHJe1{kM_-y)F}U07g*KC2Nn;VJz;@>T_em@14~!VcPM@c3J{` z=-!LJn&tuo>alKu@Ov24n+qjK*PxF6r`$H)>(K;U&cE@J4jqc$@XKmMG*!fD_xv5E zgFYUp#cRLnJ>-F$t3}=ql*429eBg~P;QOnn?h)M*I>@N5n-LRm#DP z(wpCbF8~dX#x{(MCZhr2I1c$8zA&6>E%SC7(>o`hRIrHj{f|7!+;U!R|6`yMybjI} z7@{JRKp*4h1wi~zaYHJ)3ubK2HVF&0A&PZsO2R`-kExezp;NWvBFxzp?>{Ce0)<85VDqHPQK&?%38TA61S63!tJr3552ZHmOcC*Hm-XPQ-s-6vc(P5 zeBj8V2J&}Qa#Qer3~wop?VeQfuKH;2`$QX1}?Ox;ofO3u`-+pHMQ1aIw2R`Na#R5jI8cUuZ!WVk}qCd&=cjlW2l z2+?3XCHhU9Is<2m{yO|pZbkj5(T&CVS=?v99~=|k|7*o=ib$({c*Ph^dIZ=vsRTjI z@%T&fnqkmc!{$?#e@4WW1xdfg-*g_Te;3;ZyuP7YySPh zm5Fx(DEkBdrWc$+i}n+vToK+`%~F-pqK*3C!>orulH>B_8vHWk!6zdf$g=OoO%-u>#^IY#~& z4FwLTaC&3k!Mr5j(D`CpG^hFVWN8I64`}gJYPrkJSE!trX||D80OUi0RDbSqAod0| z^oronXyUJy6UHJi#mlAY=Zyn`WyApyl~p)4Tjj0B0>;qvsq}JXMG&f)2nY@0{*1+u zL=~ho%mFR!i?51*c?Jm5GSrlq*Dzr9@tM&NLELH|af^*Roq^bZ6U*PTgcfM7y1#|R zP@g1M>VQr^6gi!{amUdK6&aldK`T8pYj--%TcZ{5-*WJ{;4dh-HTFIPEx_nluTI_t z2SomU*DG}@17x8d_PyykD2H|ay@(&CgGpOFskUoXEL!k1R`obn4@8C6 z*|A$rFCj8S~{3@b8~7lG?>>sP_tZVm-{2@Bi?01G+zDP!T@lra3fqkK$NK; z)Arh|=?!MofdYZ{XE;Nr9LB2T%#DM%mOy?N|(>o)}UssZQwSb5`h6*QM zVR(T-f0_Mmf(%|hwDh~DYmDy-P#!z2X3bm+6Qdqk89F?uQTxWP>*1KbcakH{YVU7p zAg%i8kofc=DrnEzWW;BU63);a)LI=-$%c_%(mwf6bTY%H#T<_Y|3)6QyNl_*R6hGk zy-lOQ++6zX(iR2}Dvaa27QF7H!*tU*6MKa50xgVqvMW8h0`%UmbA-LAF!(U*mb&E1 z{$RBE_B6VQijH2y8E5F?SZf~#UODmhfA$aKzB$b)()?OXem9yziH)z$GU9uP{!DD| zF7iKlwHVLUYV@)G+$TSnURV4EXD83}`+y^BZ_FnQ7t&#Pucvc!C&GuG&h|aJ8^(HP zbj{SBqkgH@{12m%K>s-F7)>ie#PkD`Zv3f0AZDMU9Kz~y27C(Rb~rks>gPV$ZCAQ&s(ZK7U7xhnI6)KhaRkP>Jb#k_v(26EkeQDu-Hw*>wE<2np>!7ybX`AZISG3?YAi*yh0;OGJ?YUcNXhMA} zEpOZ&I%k4D&&%7u(4i0CCY^5~O2_#vqRnxrHz#(w>E<2O8C?5z*QH#@7m;)K-i6`0 zYF+ex`sxKBGW$`FXL9FYz-Vss>zxPDsK~qHwE9;l{+Pw~H)|}>P-3il+tx}P_gRj( zC+<#AW@yN+Pme>BeMc_5i>*dW-!pwRXE|WlI)amrdm1f{^$yRe52DY5k&;{4F<#8k zG}d$4PG=CFr?OB)k&(Yg4b}#Mwhk+PAJ5fFfgKS*+{3pXf8jKmv?`mY3Nzu5vF`^0 zU2e(hqDjqQ^V-&@K(|rQ${&4>YCC=n?~D46_~u60k*Z}*=oJ)Tv;Sxby=Uy-8n5w# z`dc(k;ju@E#PYs8mAf1DvED7`I=UT|8*SuVUSPxl5t(mCxpllhP!lh$;U8!4PD0-U zoqNfKkD+sKV~9mV{K~$%bb9A^0*2M`_3jOtxAKqFyO*`Y3sdOslkskn4}t`O=@X1+GLAB1|xPwx1UUj!wk zW``Areh{Y={py|kQw~~b8FsBo;JdU1XaYk|mcXlS%iXGKZIw{Zu z6J^{7mB%xIVlKX8SJFE~5WeaDWd9vhubr8yuYLkGT;sT6xvdrYq!%7W{Pss}3Qp5b zI4VS}iap#kfwh}_wQO8>p9Q2J!t{+p|_)U|3hQRdm0wt*rk5NZx zSzDlxF0^O<2v+6GN8@I(FIocv&_uGqbk;#^-z(X8n`B#$#XqROTg$FX5)k#}kjLwK zEDzhtnenwz(i!0?@iN_$acCsjWAhf?SXA<1!&c(WgJ_JR7Q_498_jWSe|m+i4}Xqo z?m8>310>?PARgHWH0vKpHSX_+#^@H7vcpI8ElXZ7ARfEA8 zdNUi#9T?c;J>5N`2^9-mJKDl71HGa#{Jk;>HPpNn(6HVNq=D?7wXuwN0!q~NnX*-r z0gCJDnOAieP$egO-CQh&=dV4^>#MZ?G9n(hYgL7BK}ed1#-sq#VN0W zUut_VvnCp$nWDATenVfNMJv=gTpi0-5?E|C7S7~BpV)Rqa6~bEENtI;WhCSM1cncXSagD`?-Q za2glKs1HE7)25$4FWlM)U=i+TSy9k0?pP)#6PxKcd*} zy;}QD4ydh$ZSi$qfUvy5UA6i&kY^7B)sH-a(TAnMgspyP@J!Z&M>=eb>*v<_=(+5l zK@Go8-k?V_a858Sd*g%Fs3lC5{MJiYc87j#vD~lZ4j>-QR4!WW3B!~Si@|_HXnD28 zv8EG0(GSstg^y3PVC?&o&u%igXwg-)%gn?N)xPJ%ckzCOzJi$TMxz*iIgu{+yTM2W zvNJ0z`t2N{GLy8e>D?}63jZFZ@5(0W40DF_QM<*%}-N z$1b@r&L=SP>+pj8ZYI7IM4A55e(9?&qi=)U-eoue28pcC1-ZbC6WA!y4&2FmuvONI}nV={3oq0gHsoV9OLW9Ya z={LRu6=*z1%Q&fA6jc=aWZk#%hk~xEtA8&o!7RYg?rj+Ei6Zm;utVM>7>chwd!B9t zquai^Umbh`C5q`c`+|HCK{~|kwPhuY?-x$wOR)lCuFEab(Zeue)u9&NLxu-e|8$aX zchqUyu%RL2EUNqV_N>+(Ip}J6UaMNij~XpsDT-U2fH?(e zf|lb)81@LXv2y7=XunBdJi3c`wmKYY9r(45C-<$r)k;O*FeJ+ z6u0W43K~+m_Sf_Z~5t zhSj^wnD@0v>BgY0qpWOUm0B>xdSx;o7|ybH(9U!~Qwn`jzpymr zHGnAmJ8#)hTw$`ppt&prhZ^2aSjx0NL8MFj8$z0~yk>IC!&mWVn_wU|kuK$P0|qKn z&y&>3fVfe_yl~?-)R0tj{JGmAkX0ntksmEV?WzpUqID-w$5=tqD$9@1(v<(AFxL-y z+O&^s_=Ja{oM`&@Dm7Fqm$_SLyE0G;n|eO&NP(o5BTKv+xB zYKA=tgJ16ocy3pNp~|6d;q^omdGBn~#W-0Y>RSXQhA{HyXei&_XU7t@f4q9_K$-Jr z)S08j(Cde`B%eX_rPq4F>ndSDJ*Hh)fx$nFV{!k^$uhA6X39?<4lZTjp~23tQ#7gE z2c=Nfi4O)pL&N51nEZ4AQnqH;8pv)!(^rpj>wh)|QVn}?Sw=aGOxWJr8@>(5H+q$K zY(cUpN450rJK?0>rusB zuO~@+j8MwgfvTRYT3Fb_TJ!9^GfasbaE-Wo4HX{~-OpEb3<_fsMJo?@0d%8MeDnWve!?j0&82hhuBb&Nmaf}+4w@=<(`LC3 zq1M#pSMNp`^%T%$#78y=JAh^~g*M$6l0=h{`RDgtd50Q=!o2tmoKQmsMIvlfCX5~B zXpV4bgVIjrwEMMX2&9x3{}KmKj)C}@e7$kh`cU5Ml;BYy%j@(NC1N~_#1kJ|*b}ih zSgnZ!3*Ye_s6WFHuN?LPsH2UBB1d|l!}mbLc@7(xO4wp?^Th@j-1zGSw>LM$CnRWu zt;cjJYko-|z8o!yy{GFjzRe%`Q0x6I`4tl5XrQmv>|nuTAZ~f-kR|yPQ3(25ZPg9{ z;eL^rN(cdniA|So{8a_YL!mN7l?^a@kB{xrCSK?k>NhI*ya)AlpUseabP?Kas@U!N zlno;i^%B~5?xVckfcbDQEYI(M^7DGSjX7$%(7Jy3`7vyKR!Mw2cPtf#`af)?^sGfA znF+2Rtm6P*_vp{&9oo=$u)KA~s^ou;7Zx{Rm)lmIy%i`wcJ${gU^?IntHWI_{$l(n zMBKMXXZJ}MJ>37T)DQdaC@tyo2Q3>=iRX^HWk0Yw&#~FdhIhBdzy#MWYE#I6@Vi`> zlMa^pH*X02SHcmB`XeC4wTBC=H-`GJsn%h47~_d*bxZhWFGT}6njrs8SRa}n9=o5v z)g4Il^>v-${V@3cwb^h?HstH_hAn>GhZa!g=Dn*#AxFoRWqSu!mpG6)s%Mxj4Rss6 zt(9e0!&2j!&(CyJ5p9Gm)!%}FBY?1jwS;}?ZaDVUd!n|}Z$syS9T$t=GvZ)~8aN{y z_<1{OiFL_Y_ap`Io_`ESwHSP^C_S@Wcds4~Hh*f}9urTu0x7Ln#4CppH-i;Bki82p ziaM@w>@_bmgx*_wk7h=cp$e@go>vhSKn=TcTQ)adh{=jkW^ z)xR)yLSblw?>21w1W&jq{*?!c#TF}%Pf1V|G5>ffb{D2P|RbP+(nGT|A)R=%U2=4ql4uz z`HXvCqeubdivBf3$4j8xGg)DS_;#R^7yb6R37`nCXG-gyQ=wPov6Xgq4l4QXhj$6= zhvcg*=eYmW0`>N%frl%0=v-r${x1~_7qX;&3D@F?)w>hjLlbPA&q37LD}Qz^PeaeG z?X#V70ch}YibC~)GiWSl|DXaJAYuer7`K)RQ+Xl_*Cq#`!cg#!58W9RpYm|N=e!HY zt|3eEzJ4B!2#8<&D~3>0i{;MpOUGaU4hV{b>_wHvsSU;^?vOaMbKW}XJy2TD9b0?k z8sj<2_ZJ5i`nsU8`>wBJZex1X!|@jk_G#^ghEkDE+T%*bd-T|y5vxM~!GpYTYwfq# z7NA}z-Y)y`5XyR_HrD8{4Go8us&qulL-!3Sv-h))fmC7cx^WWY32lx{J9wlc2U_;* zE!q*M1ijmi`Vs$ZM{^cwrVXyj zA|iV`RuVY=sp}0DdKyRSzaN6Jl=&SzN54W>)n+{J7#6RekofWEW;LeALg?#t6t9bi z@*(M^rBW4W-Ttt*dnOo&T=_364zEYV61(<6O-z>bC{wQDIVT#SfDEgTw zdOD7aVL#@w__9gIeEtX{@6YzJNO0|s*picz9U+4G*)Wbg!y!IQ@yoNeHvDUi_ zO!c#foF@Nhe$^nfa^$>!lXC*P_P%~Hqfv*Vg7ly4<~#^PN*kS$_M1V!YSM)eP8}d2 zNvo{)^FVI3dF`6Vin`s79DFX$353($_it(AQ7@15jtkn;KuUFdBb%_Y?%UmOWp7@w z53BFhe~@M9hoM>jfq=y31Vq0(lxJc64amdnmf8YhK+;sGGv>JrosasZzZ}cJv8~!H z+!B8iP09yEEf^fZ{D2QR>t6FO!k}k{PqB9{)Trfz7P`WsqcX$we*gW)HE1SOnfFHzwy%+k>~ksKEK#X4Uark$7A8gPzC;A9 z>@zmL$8+1Vpg7PS@Oyti1FP_@4INxSeC(r`?uq5?`?OP{T(7OH3$MJNsh@X1Q+trE zW@8joKN9)k7TFBsf?M$?waj4p@`w;j^nISMGBewoD$}CV)a(vzDE^J|ke*zp z@F1F2F}!qv`U(29I1~20#Oj0bW^zf-ab0M7I{Cq$s(v)NI;}bLy*rxfGs_p#dWPnf z_1o8r3H@_@nhM`ldSP86ke}OatYSTjt^aQ3y&8#EAX2{W7u_xhYACMClb`pZ%>Flb z!xIG&Vb9r`p20`Zd@)h+gq#D=%X~aydCGwLJ#7BT%LHhuX&p+QV$wGODu1eKnHrOy zg4s_VZ)C-SJb-qbS2ZbZE2Q|0aZ2f}tV@y>^Xo_N<2bwA{ibaOw;Y7k zFHF8XbNiw5HW>QCnQwY=6%f`OFy(X5g|SQ2f^gYI)Pq+d`@Cb~^MirS7el<*wn2})S0mbe6R54%z6XTdM6|b_?!WxqfFkfN>6Vx8ie1cI7B<)iwcGu|G~*7y zaLs{F)w$C^8QJlAhqx_TusV4m=h_=I@!jKordAXp4Hu4oo4ExYuPolnZoG+Vs{4CW zqT~_LDZ@T^s0+p}9*xgRegU-Xpgd_VHlUsDVv)>aK_i{}Es}kmfpG3gvMxylroQ#R zaah_2)s3`H)o()RgOMgrEa!SO6I^R?X;modo0~m)w0b8RE%r5x{jLiWy8_Q&zLyHDCVt-|H3);d!XJ*(K118ofsHZ-6qtzc z2-(7h)pu5V*q^~Q$innoIA46Y6zZn=Ircvf0BRNI+7}-&eX`j&{#e^flV}DV@Wb0Z zf}Y42wZ4xZpszHR!?e&0$j47!itUXRxq7RDcBUE!%ygI;Pfea(k`Ff#IIEV(WX#?))IxhD?5{K6Km%G*?^ z%`|#;MlA&m$d8Z@X}&?dyF^tl>D&b3o9zc)J`BOFB1it-srnb@tSassG3%lrI!hIc z`%M?rRcc&Pdv_WQO0HWM8Z3chN&EY0M?4k9{MFvlNeD&bzvA!cZ?p$OFt3`#RR&HH zHJy6fJ^si9I<|(6fCSSx{?mTWGFQLz!VHU7b~Js;t{Alkq6i81$-)6@uWbIlSF;+1 zaa(0NMfSl!nPEcqHl3CCe>ul4MupsnppuD~9q@g({Ni7saG=@gzq%&z18QzRX7SR0 zjCy0g>v0W+piu>7ob7eQr7lU+W)mb3d(fu>utxOz7y!fX{_EA{Jv5? zn}i>L@j_OO&r+?>`o}7^ksXHyrrv%u(uu{f>IZEoSMi5on)(%6(Kh7&AvX1)hagk? z1QCa{Po1w2`S>xmhq`}YKu*QOA&6N|nke(o4xhcU|D4+UJ!3!-hEK46*unpY(N4U@ zsV`w~U!dd`m5=_LF#eAu@%uI}e`+DR(8U0)Du#Um{cjI<79NzvoitAj3vA8Lj{()lvsx3(~#&vr{Y%^A99 zuCZn|?n9%s8W%>r`(d!*(jFHx1Vk3~w2f0^XzXU%og3Gq5!EbOM2vn4ksPmHqcmW7 z+QAf16j_D4!_MetKUY-Q? zv1K21%@Uw{e0F~UYbPSuu6~`qDHzQ+ZTiw|g4v%VvK$6GRb7A-Hpd<569nY9Ce4(& z&p=rBVCiI-BpP*<>Tl$p#<4Wf-yQzefhaDZOr7^ozS_Y5cKu>JUuBxabwz78w6HB< zD!jo8(JqNQ`NpLql6_a$lG(<8U5Dd7etK@h^(i!Xcu?=tm5)sAE|U(2$$kHeC_$Fy z<_0!s4!DC7XPN9t)P)ycrK(ephQh?BBn*?8`oG`Ly7G|7Hq;jk@-u-OAah;}Zr|lY z{dn7=eV)%zciMKB``@`^@brsydP`p7`SpM7~e{X-T^Vgs0ca;Q}MBf`*fm(-O_@GXC#< zLC~#qmHyw36O3>FbENx97Sr(#UAIl2rT>F>if5xKe|I(NLr(=Ps5$7bXKUjh_J8hc zXyVLT!*jDte78*Jjf9MgjYj4&{+~O=O1Sdc70q&NX7Ae<&-7eWiR1%d+Z4hJS_z-s*fFx!{lJ@?+F@i%y7?R;&L~ z@GPpd(0_9XuZrr%m2bCYVL0{a$vMrDmQiL2R<#z*am9 zBJ&8jv6x4o;rRh6+6FPy?__Xu!_EmrP}vao^lvI6-BC`LY5v9Z-m3#IzVkPa_^DUjcMCzusnU^hI_In zyhRZ~z;npD)fi1V4F_bdn?eMK4V-(`c0gu0X=Vc}w!ZWip4x9Sbsx=psCLxve2Hey zb4t&|?Ls}zo&181Fz`EQQ7LF@uRc>fD|D}zcNy+S5!qY$N9}oRe=8zv@2{fq03m~T zXjYNvO{`A{0s*}U5fANlvh}IN=z3cy}kBO}lU78I_m;C?r zJ5cs+DZMB7is?9hvL>ed{af_Wu+1~%Y9Y)?*Gy4+pQ2wow;tO%jNuSxdz>6j%zXsH z*E@rOj;;T^&-58jyI`;5sJTt$)D$Ha5q2##i-;$~WM10H$t`$9`PspGHGKjaX|5MX z_{*VmW+ApDb^+0!HD4q*GwWi^nvCj?vN7@xI2Jv$;pw&gOy^D7yj9v@T#6>XWczK~ zx3d07o#duD*1>SbPR*dbUvQl4CwjKBeuZ|tee|O^8=$=AvvP0y2=oU&rtCU}K+nIN zAYxI37L4!Ln;1)>@oOCK4xDGIM?u7G0ZqOEM`7e_L-MajR%phRzP;%3GUGW$j;zkI z=)KQ$-MD3ghigayj2Km|Ha^4$Y1Uzjhb67ip!n_X7c!F&kzKs@yxSk3$9}t}G;PRe zukpcN8Atgnpawp?lV_L$A500HFFlxeWti^QcZVy(s{8`eaoUfIpM=alA=0X?;-6$R zP;Z#i?k$oG{4<&~7#0ffX43maW4Mee&0?%Rf_(O^&5w68%=7HGTn`~C5RvQJt9iz+GHi71(k1ac3;N>8K>i@DDsr)zP&dL z=Bp}(5tY7N5+{?7<69NRbB5Ox#rMf7M35PN3Cw(RYOP&o0#jPH&j!#F$j;ozr>r^* zV~qpITh|}Q*}$bxC*{J_KB@2fpH)`NMw{EPuxSj|lq(uioi$g5LZ+<=T>IXw1aB#x9e=qlq%5 zQ(vul(F7exNF^-v8&K{T`U`LIN2!kuhV0#a(3JP7%1dW=z|6^s&y5Wt%=70%h=5xL z^Y>lVhgZ+ujnzSs_H1Mv_Q72Am6?uX5*A3FHCpdP6Rl~*Iq|GW+k57DM2oS2T!@_tb5 zziNis=DDc;+Q8Hgg}?vA{U~74*+p}Nea`HVc1z z=N`i3!!Z%(mpo|9Sn=UUzguW}2alZhNoAD$G03&xU=3QN`oCnmPy&-44aMc1Nr<9) zc{SewP9RoQdBn0~dX}{l&V!0;Z=>IU-SAi^=FsP^QD$_E~ew8 z=&rg7Gp4#(h+92DdSuFE*Z#Bpp^)=^_+AEYI2wJAM_-zDB9c+w>$k@R(B$p!9XeT< zeLx;D?0VeI(7%SBDJ|mHJ&gP_nq8CXTl~HS^7f1iM``Z)=Y416?_IawKm+=jm|d|G zDbO$Z*y!D*S(xez34GDa$RD8Mmo$T@L5AIoV}*qO!mxdmk-cYuU{Pzk#y9?2u^gdLX@Qe3j<; zPh7%^AE6YuK1~yb?pqvtf9ok~(#{O9wlzX^8p|pciT}aFqwdhqRL!(5!or2yR^R_( z@$+eMs^->^KQJsEW8z@)0%(w`)W7R9jFxy;d2jmk*DKiDgWdI)kR957e+cS!QaNHFE#DcC}zsa(K$& zw)_j24){PWPiaCFEj5`D!xM_)$lwbZPTfTjh-w1Xj{z3`|3Du;mAx8l|~$U&vkX1JU>L-5zhT!icv?1 z#suWfO>q}7&5NX|QNc4jaySYdh_wAU?DWWN<<|?J|D-nnE8|Z zz!k?*BNMwP-U-n!Z=mc+ibm7qdJc2mBs4_!dQ?%wk7IfEHG_o=I5roD-D62i@d-5C zi(ADn_vrs_zoVtR(>1$4J0c36_1GtSOGL!;{IPFOMZ?wMO-~CLJZfmh?Yy?|(btG< z*`{V*P{=eN2<|Q0v}Zb?w`7B>!>&iDVngn6heuUVRl9~g!u}mL&I;e@_XPeTvhjq* zY6Tyt-W&SBMAjLRE}yt@wR9MfpWY1}dV2^FSMNA}rFQ-g6RICh)T+pD^=jw8Irc5yTyUQ-fMUX3O{DOuUP)kK7)^73zY=OAJv^eJ)Y zbF?HcssB#VACX?H*C~!+L-aZCfh~WsQ2Ic_Yfbe(sJBzWOb+jiDAe=w-(IAlJW;P) z`=4)7AM1WrQ<;5e2p6 z(fmS$3>U)H;9N@_Tioyw7V0osif8-n`cMpYiAC)UxD*Jq7)wJwE;IP$vdyaKXDMo2 zwe5uYlZ#OI*0SK7>EJ)h`JS@%9rAtKqyOHI z8xf_>>oK0*(5}czN~}>!mk~o3Rt4%O_tiP1DjwFPFae+68OLS9sTqB znmlwQEKCW*A=Z9a ztAjFvnEZQ0iZ3*-xMB@Gt=>*yu^8WA!8uLMgNG7OoBxT1n}dH)65S}hG?}q4pa%Md zrWc#!QJ0;*#F8n-S7Tm(X;m0&9BSP-vF7@K0?G;vyE>PB1$8{wo}Y2&Bbq*S+cw(s z8=Bk4!gA^*I~on$UsUg=i2746wr6?hqnSDCNJ{-h)ZzBX@2AryD9*8vriyx^lE)W= ztbby0+L`?`@5*1zA=z{w}0<}rp9P}*VBX0?1rcPyEkDt)5ZYF;MJbeXc{+izI>JmSB)At zQd=oDKM}PkU(h3M4*Ewk`=tLQqZ!V{4h5@!;`wkK>vwC|PKu!MiZ6~CWJg$vEMr&p zS+T<%62Ar;SEFg^M@|{$|J3Eh=3&Ppo1IH{QO&l#p(`UydQsqOLe$QTW~@VK+Vxwzr)94*!}PQnbLFp zf&6d|QWw@R*=4BvW^d?T=Uz10R+}7~eGk#T9+Ojjt=#`kQsD(Y84!DdxT@{_O^+BHG`H5CnxT2J%eLe z&5|w@E`Vl54tloXs?n0-?y*(J|A}`+l()}shNyeu*s`LM-xmD`|KFvk#4pbo^BdvU zqrf&=;s<2(@_-nqkxYAPiiVGzGIbUFCr<94_TeD!2>Yy_EYtB}_M=uKQ;awXOn#`H zt2WvSea(VTvg0uw(U7*)!%q%l^{3;3AzzvX&oK21>2R&2ldLJs4r~e?T3FBYd#7T_ z9PN$^8u@s`k4j=el(RQwPS~X5*fQ3d5ZKp&FM6&v6q|&mRdU^?#dHw0Os}q763Z)2 zu=ig+(w^{t_%3DdwrxVhPQCiOAx!o=_(2P(H>*$ZijQ?9TN!`tUSRMS#hS9;v-OTwxe6C4R^E%G3^C5Tlyr;|7QAC=5|NhB~N<`mmu!1zf2{+9Cg|&AHuRe#V$a!Tp>RxRYi7n+%8=B8h#f)Pu|5H4P{vvz zxh@KY?(lc8sCtFtRNl#RDNqIE>P2ouskra5 z`MN*q2sznw50p)*15!KLfYfMcTKsAfW>VKr>yk2{g>WDTZsEd6Jy2;p=(bPt>@c7Sf(U>bc z_f?!88ub_*6jpnHXhzRioVXAok@Na5pC5sdU3QKEC;e~&cejdm$KGaKUm<(x)ENpE zmzo!z=9} zAnMjAo+H7hQTyk8Dc8R<)?pYM6S{2T;f$tT?;8<08F4%4EH6Q~W*ZlU4hPObp# zp=l*9ooMd&<87SP;y@~P2|dAn21p){w->%vL=B&m35!q05jBu&JuZ`BPXO)sHJ*#^ zdZ=|Zl#i5Fqp7w2{x&pqNWJ^_Ta7ah>R(MuaFa+z^v3~j40RYhR|wxHdfRKL3+C>V zT{b(Z{cq>h|LyO!UwO&l&jCTyGkwt^=3xNz6n6^^}Y{i3AX1m zwmw0m!56t&8~Mr%|-zUD?2D@L%5xP5+MDl-=D69ryT+Gk7;c z;k}qyYQ<}Kq_vOUEq5CYbb8gje8kX~M%1IYC&>$f2*32VfBn)ugtr$j*1UL`sa@j2 zC4b@P4gb9Fg59zy9rx4G_>FIW)o-2pU$>j!{auB2;J@Y@5I)pSeZtM?U%x|#({&ZKt}qzL zaaMh4$yE0PmGP}1gIik{4ON#a|Ne~CvCKJ&7Rn61ht?lg z%N~q&Lt9GxZJ*PtP+gR;x5gK@T-L!|&MG|&`v4ri3uIl6=#a1*g z+Pky~r-JbO4S`y47ACt)e@_JNK=~AEQ;F zWuoe?Z8joL+@WP&!gS{sj}4uUH((lHL>N{1M7cvl)8(y0XlWPfUHoblD%XH&4xM`Z z4r$Jqs-YaN&i8uUPP!gG&oOn(?Pzj&Y|OX5Lzzwb}+N1rMDFoNO29hL_--$(6@o+|63vvBN$NtP2eSX`!l z)2VrxYdVPdMe{;S_+}g@DdE>m-!DK`Yntu-bOjN9*3S=4IRgEpSlOxw1w`@)y8X2~ z9LIj)DgDib1T-S>U=>S!6l!A$@|w301KOsy9!`;MXhbBsle7B+Q$O_HJtN;K{0K;v z9pm%kjQBl{z3uX$27g8z7w~KWwnDl3P`WAa_!X^VFkc+s#xBL==Ku*`qu|&X476jV z9!OCewP+{p-6J%ODC?iUIN-tXs}S*cwd3Qoel&CF35ecM`Je6J*xB6N?!USUy|U!d zXAMiJ&7=KOe*G*e-RwOc%U%jdjJZ)z-Nzrs-dCV;2xv zVMj@H786en4Bwq}6z#|KUaD=@#2DRdXZ&BuHGD#%4WG;2)%067>j#><9ruB_I}T=o zObcgzdc%-*SQJ+PLl+3gLd(|(ogYNYPu1RDsxO4WR9PkVz5h{PJ1TSjgIpP+WvJI3 zwOV9)9@_g>Hlykp0ffcrW>yh7z zUr?JsZ$E+!H(B_ji5O_{sO+AR=v&xdjF6|O8LgnY}%#=4_I8Mr>*1k{fc+}BX-b~yW zjvCT&aC4G5Pu2?)Lt7L%eNpRrFf$gar!O-Itux|xFYwBuvW*H3+kP?lN>S0# z?W&Jxu|PMFh`!%_9-HTNnj(jsrgJ$OFXvy3#=g6SPTOm{ukC^<`ZI#aH#|K*8!fU{p9$c&yU%+Rbl*g`ND{HrlR}o zZA@=_NH%a=n_CKMecFF_|GIDgx}MR_$9KQixQOxq@o4;{%{ZD6Y&;M z{%$T5fC?N9?Q0{sQE$ofE4bWr9GiRR@kh>~Ozq6fX$qITHbjGA(wznkDcJD=&w8Uv zt*B*Ws;lEGL#G`LoLSFp^Mzrrp}B<7DYp|?{X0H^TGI9EEVRsi{ly{0$p1m3-_qXo zp311Y=I2QZjoUEo=GXF<#)MbJ?29)2{6vWy6v7_e_?Gq&!q)y!%d0qn=H_#2=yHs8 zCzs>fUu!$vL5L}ueB2)P_+Rbe*lY+VTU>Nj#)-|RyPOp$iox~|&jX^?Om8?z$Kr(K z;+Ic+PBD0Ap=amUwpz1Sh;;uff#6h;|*dk0%ovZZXwWApA~m zB5NfL`oHPCZZ6w~t;a5pA4bJ;px*xHQmgw75SeIvclFr|IF{A#uvy{)0BUB)Q9&OjYHS+;Gi zdD?^s`+s|tJNxSxhMkMP@6cQA(>{o5?ON^F59OfTyT#WYiMT`0gNA`qt<{Ls8O4=c zj`=Nv1>#97clT5g09-vBH>R`3Z? zVAd6w@~$fIs$jyo{_{JkKU>|gxSB~v8&P_rrFk`(>$j+Pe>kYEW8g#4=(UUPXS}g} za%RbK%lb{Gx$M%D{?|sYqTvA2@q2D}nclBqj>RnD%oohQGAopF5lq15QSPQytnb_r zQQv0Edi$*~F?{Lx%R!6>zR+)E)#2X5s6m(#Uzf^_W4CC!^j;mS?;PBc_G@wbYNq=Z z6?^uw8#8pGVBuQMC;94Gs4P}s%S~FvINsW(D$nNPgeXr+BHfgkbkh;-W5;Y(Q6doM z-QI2)Dgw%);#P_Jey00XAK-a-POt#xLONQ;xt9<*+H5cRzAsbzM3)!a@^+s_c?Tn&R;gp_ z!2pNEm+G%nsQYc5YM}8MEv%8#VL9>$jUUTy&92i$l}QdKAHKly8+{`h7akPKqw)s_ z3`T;Fd$-Mq07|n?^uT;(3%8WPQE8V zLhLWuKNqmNcyH0W+SeR$@P*yvQ0i%EXbTNJKpu2QV`4{XtKBeseUVi}5^u;c^yQ`S zq1wVwn7MDS5+K5u&nSBT%n{l_22UbXo^HEWa-sHC$1Cvm(5mDhd{ z<=Kj@(~DYqbq47ezfk06*E?+1GAnlYaa;Qkc36B=dg`jIH9T{X@JTy_@z5n-+T}e} zRSk_9pYLUU#Bhe)ovwZ}Il@pIt*4dz`WF=FChA)gZoo{4ZdFy(I@E3FHKaHo4wMwP zokFh}ystoUp{J{D|t#OU~u-U`xEn7KJK=$Xlx3Gx`s@xHg-nC1;^tfei%yku_-ex=7nGdk$3I zH*l5cg6WtvUL0vP&}HByP{T9Nu%^qebGeI(`6s`B!?CmtL_1$Uf#Wn4jE_843WH0^ zpReB?V|wmLIS83ooHGWvg^vkV)?)b0K2kU~cxnyN`+#^ybpDo%8zQS;QumNy_=`BU zqf<4%QaupKS>iFJp?UwZqQI@dbS#ih!%8O^RJ&PHYY0Ylf`thjf5FmQ2` z+;WB+p7q8+5#zmW(7$Vs<WaIE)I z>S|{{!pz3l@sHwvpz6=Ueh<}Vatz0Ec$kx7 zKL3CBH@0t2=i27Jeei!jhsG2RH}iM&dm7ImR~6h8ei&9&rYJ$J^m=R{o{1Ra6rrDA zm}c{ux;xy8=HlZe20GX45iNZ~ZZCI#|Ihasjvcp4G^6LM(IU&aA9lU}jR#L(Z@5xk z9=z`J#dc@;>cvYi?o2wm{rCUq&T)6Vs6KNDNXlX&?}z`#k6gJY&L~v^=}f;96Q_5< za-xc)==o?g$-u?F{on!AJ$uWTsQutyytYRnnLy{6pn+8N4J1|{AjUB7=edaR49}zt zMFj2x_1oSyHb({2t@lB5Gs1Y?G~=@e`_tKg!l<(KFy$lk?=1cBvHBj6)%<&_vq-3~ z^GrxPOk>i;Cl|AD!oP4>F7SpUAhuJjsf)~7%iY5D$Zcq{b(cC>KOqxTI1aytKE zNsdYM*XyFj`-S(g7I1q-(R~FCwou?h5uKk8y-~5%lT7RQ4LAnA@2O%;^tvcCUh>1P zhC2?=a#ej{LAD>g+Z1}g&DRLec+|E2{bVa~)-Xry3o3t`Yje@UxyzT-M(+Tr zq(NimC>{U8vou?XoceJU<`u75YkD)STd%GzZuYzS>-xPx=8DOf*F~%!Atho{ldr%> zSl*|g_38P#zOIODdHB2&`#Qc5=BGtZ9Di|C@4tSJXU@Lh82eiX$IL5d`|S`nw3J@z z3fS-o&$MtOZLEwwe|V-TEe6ZwyLgtBf+*=Y0cc1ojeMElg(zm}jB>g{XzXnEpO4pE zfU4m=uOdk2LE;%YZf>|Ky%FZ#*~uTQki+~>i5@ynkyl!LlhL!tVM zH&D+g&gGyn#>XiV3d{d9$pqcQw~g=k`oMsDIT&n7S~p+RC@b3J(t6$*IcF^XExrgw zSy|Y_{pQh-%yQk{3LY56J=V{YtuS@YxCm|pca_m$)3=R6_1pFy=Uf(*v zatC!+-G*U!b<|bLryLEG$REPL8bzzjbE+3~>AWB`wLRhDxy=+b{Q5~%eWxAJE@*GM zrHbk94ce{o#5-EEFNojbwY8K-I7hJAwPweq|LS~4 z)NYyUX(#_ezL~T4flac2it5~?cV9y(@-F@Hx6hyzT($0B5s_mqcW+%AgqEUxkeKuv**J(P|Gah`QKlJ6 zmrMTHDI4zb0xNes%j9gOTK^o(yxGunaH|MVD}R|zg;mhUH>5QrDX)`>mh>aKO5b7n z;uDsZnN#2BaT&_rbD`8KQU-N9xMy^OusOh+2I!%QyU}fQ1qD z^Q>C*bA)F6S$FZzeF54+n+lIKwmx{NpO!r5LeQ{VQw!xX)?aVTbYS3I;S<#Cr}UXa zr+oERH&DR|4IBGu9a#Ydem`HStyz zcY}oG(eGlN`F-36fy7iBW4j>^jb6MvgKMfl9hF(tZ%Z}<&HwAc*H-J!KTvyfj&Jn* z7x%WcS-FW}IAaGPWXj=(5zN&GHoB&};FQD&t?>niO4j`5fAIlrPNME3J# z^topZqhI#f2UwiMF};YXX|x|i{hoe^mS6dc$2k8FF2fWV6Z7X05Njt|t4IF%*E)W_3LyiK;}eMs z%-@GdkA1k5Ea>ZmWB4vR={P9_q!2CM{aqc9_U^ZJB3b4CjvEy|?cux3fz<(z9PT}Q z#&-zj>r6h`s2p22PvZk4HjmA)xHI}Kq5JDNhGVX7*0(zUJPj6QoO_MKouJjI=+WEn zJg{0Z7WiUV8kP!f;M6L%p#;IZvxoh={`m*jW*d1HfU5WS&(S@6K;m{v*2$#D$!M`8 zw>)F?6*TTI{gty~3d9{>W;f)SquI+GGbd?|FwCiTH#`U9Gp-z+vFaLg)FtKrfB%7OmRX^ta$XK~CIHH>u$Se@tR zU=Gp%jR1Y+jN|NKG&~ctK?Q;R8%(H$HO0l;$1^tH*Bv;vj;BH2_pI?5MgF1+XkfE_ z>gA|X7`hN7v?je1&&qV?*!UfP=-jI5&d;Pl|GvQ9pJy&C(ESJ;Yje~7>#xe7IhC36 zr1K<frDP z#(IVVHXOQ$x?LG{g&__lo6a~zecy@ZOW4?Z3hDEM=B^S6{ye%kX8w<#CJx)cVB}*I zXsUx_QGB#`JVOV~rQB)Os!9c#HOG!vRZjT(bI<)cH?01wa4N=PgIPK@4?`y9`5Wo= zRzTxACG%W671gOfw*j5=(AMH#9d&_TSC3=%tFwL3{~Y>Fu75)PID~#(aq5tz>(9cZ z8yd|d(RrkZa>o3w9TPjslrU&C+;|sA18$pc$ETtkCKXM;Z&zS4p#6-8-aSOy%pdxx z>J&7b5p)r-z{X2Cp}CFX_zKaqR=9>vVSP_%68W=Xy;CqI6?dIeZv|0kLH<|x>VQwB z_0_!?%x({i@lB+1PXMvX!Y}XmCY1f~PrSDjAM_0-YTHl$Cl3&)!e8Okco~e5ct{Tv z*YW)DjGHWZAN|1m?A}-Qi)pUw`!lpz7R&8iz~*DQCHb@B8z(e>Sfz<9co?OWzl%`3 zh2^cfzSYih#r_0B61C>+r9IFtda%dj`ug)v@YprApS%wDLrbCZ6B`0Cz34yLSsz~S zrQ=G7s^t+x2*h+k2>#KMe)?=cu8HZ~p1Drf2l{{I*O^aDV}4`3jiQ(;4(j!G%^G3SotNi#BMsB75r3mmV~Bx1KeiF&ILU=H!Ho44^D($fM9i z1qgkyU#|S{!t8a!LADoX5h6GoXxVB<1j>f%p1cB_XttQsc+2oSs_Wfx=0w99%#|n= zd+=!jfoITz#TBccnR{n*J%0Np)GBpnOKZitap0H?4Bz_cKLqN8ar6FMCbV#^Pvd=v zG|by2N2j{1)7OUuMV&qA?v{8)!Kztzkr|+wH%$BQmxM-k?$skd>3w4HOk1*jB6F`I z>e|XRZj%hC+OKR@_mLlspV3!ik{(9`9mSu+l)nRko9$m1c>|~=V#Rx2tiYGhtyM+w z(a`MuUY}&N0OL7V^#VR(_}*+`g(~Na8+QLkOYBP}`U6#B>dej)l4$(;DZe6*cTne4 zXV7DP4vin-5z|-0bnT-M`NYXKeH^2KvC0MYMxdc{v=cV^K&%*O(E3b8Hr}p%_+h%= z0`t+jybt=**3AQDB6+{-COMc+jNtv;k_`QNHNV|B3J`h9JWK6OJ6iNTJHJnEJs+K< z^3rnhoiuWLo4CL4%@?4u)-QE&XCu<2hkCX_H7vJ>-pf$B3aiRLmF^wJ^yx+~zqG#= z8~_u`*&jy?tbp*8)jCgn0DB+F#5cTG==eSwIrQuel~oL8f5j$*l&|wsu<;;;J?#^^XDA}5^t z;2^y2zM}tqiR9vIA&y!!+^FeyaHtry@HkjVp2zwRu6mNBtg2q)nBIZ)Gc8l(=HO9G zbTowd?46>qqmnQ$H^ix}$P6oE+n?-{-iC4|UMu?Vqx)NU7TJ8#V|gA#i!wARn&3n! z#G6I;cIv^nd)TV>v8QNdGlkIR7%IW`2JgOeOB^dqp_kEbhFCCYojaBGWd!3d-ND8o)*6^=br_`#m`#4Oy0-# zo6_E5|DUH}>4scuF~ZLIszao>&L(=l2-Nd)_1iwD4xrw8GHftTj}!5%jp~KcrR(}x zqOmQS&B@Pwp$(^}c4iFY!`JX|Fc>Zw0rgAQ)*THW{&O7UwEMh@#i>AO%VbI0G>evb zNO2B*7jR6E4m{DPPN0_kQgZv=?n3yxNGSUO)<{9+GR2Ak^DLP|VD5Oz8o4&#L|g(&G=~J~yyD>bQ~a z#Uhm^?7i-(zd2`0KYyrUr9p1nYdU@i^=_HLT&h@ICMkmU-r?3>L^FFKAf1E75zIof zA(Tp0h%_80PX(;wui=?A=9M1)s{J>9uZ+`k`$0Cw7E+`!zlOlv+F(b;;-k?~V-c$^ ztlqt^vQOjm$%}ZVKrQvT=C?nrRa1=T&0DbtRPM(lAhL==SatS4d%20m)x zmjE7%bFQh+zsz+Q=xNim) zxP{K6ADABEAoynL2kn6Q{iafX>M^_CU#N6YQH%wKZe8a#JsN?%r{xdvIRlM#^U82# zOUPyqdcG2fbDs~e1T(;NntL^m>m8U{V=&0l%|^?IuX}8r|AObRZoI}@hvi-RK6?uc zH@E^N1;;3Mn6Pf&*Sd+N?eDO`Gkj=a8M)}SZeNdYd@&mQofk?pAE_Vs7KKPhZocTR z!usnB@b6~0_NN7r)zX}=eEJC_7fol~`?c`WU(tuxOdZCA`Xyfu3gKC_`JcG)Il#cU z^Rviz9x$rnJXlv7i3$dld=(thknPyU#~jRlIHo(}q4#&L(;bJI6Y8Jd_7)?`@63=+ zSw9rxM>rP7l!6ErS6$Lvb|D&jKkA9!P*pHK!Y#rCtPhj^?*{xf`T8U6N8CUyQL?2pOPWc~H`cIuNpcN$9XN+U# ztn;9enLeJ3V@YApU3`!6N~l+Qyg3FaK)qoeQtgq1yt5+C$aO{nX{*V(gYtB}F&L22 zQa)Zv|2`a3`<6{Nr91!g`zhO~m7azBuzlxv?*QVEfUN7kK41T|PKSg(&gvT;4h8(B zXldkU(Oicynio+1QD{WRRZwg2(07Mdq;>yS@kP6h+6z3lELqH)#myg^^gct9v zyQ0GI|M`8i$A9>^c80<5UMZL1V*%KCqiP9nPcB25TMAlkm}#Q%5`(E5w~jzVz`)w; z&{~+ihYL1~5<>j~S0gfuEAcG9GzXnO{)Ix7Q)lH4iJ&^87>m%Q99W2%ibsn?pmx8U z)Y2{hd|=h9DFGwMTz1LQf71yQt&hG)D%0^lnC?wZxU_l@s612lms#j_#i%|ax=gZ3 z4lVxib9;N^8jv;=J)1OBfVRh1ius`OIp(L%6JBauGdloDflnwl!EYdM&>HH!wSXki%DY1; z92(ll!#hR>q3Fe5P5m!@KLfc#Lnq^cKBn`rp;n)$S(qT06ie>yqHXo2HLnyqwN1$HV+*u)N5e;n} z9(a~l15~f)nyg!W5GA=Ml)Y*vO1IFzs5Wi@qt~Rn-HVRGWapR1VG`@~4AF{yQ>EN7 zD>Nln*A(@P8A|ldGKA_mBkGZ=0}E{H@)!R(??tQInrAp^K)4uBIbOLRWvd-?_f4St z^U!rFNg%?E3S`aJJmCvi-NaB$R;z7xEG*m+t5BcKg^tNupJ8EX9P?T3eV2pQ)ky;R zxbMY=vg_!B&{LcbBj(=}AIaZfyM*;4rcmNH>9i?8J-Ls6?JQQ;MwFj8yo+51dIM~I z&Yz>-YbdKFI5A*HJ?ikSNZBdv%p2ir~FVP7`-N+`p+JIWHgaD zi`mt-x8j-c1r?AJXgc=AkqKtYyBn@%eMHljf7m#ut3i&_oOY$jQyB1b$+*7Mfczds z(Keo--)}hPT`xa|nJdF!DcsVxC&IT#`Tk(s6&Tn$8vfSR4Ja#4B71adu<;^-`j=%K zB4`#VY!dm5Xtz~ri%hY8aCMn;TJJ&$PEI$40PC*NI zSJ|^_SfEi|L$lJa6)jBc-VpC_0DYmgwQo%?MrmBQ10Np$yVo6il@;;>aSY;0E*e`% zXwfVoE!>R_sKY5^wtHlO7`dh<%&dc`euOujopgJQmOJzhZgBBHf2tPs$9H4?&SKg{ zKB-1Fpp`lOT5Obp>5Lpz*epUw~-PHHH?@vPiaH}io$wfUK`SHnp0niuz#AsLL z2b34mq-b(=7f?(VF9>U%fj;45c8}ZGfM9vH%1JmA{_NM7xWn-fbviVinsmf;r&_n?=UqN>w8K1oK-7(T^Cwv?0Ibb>Gl_>&)3Wv z>PlfmXH@Jo3FBc9>Sa$oF~V@5&kXmkaBimSX`#`u-_0I3uy~sG$WX9#120V5@a8V0 z)kB?)(@Q(DFX~U>Kb&GBh`#IeN|)fdVPVg<2hp!E9E|pBes6U62k1GyqDRt{gh7in z>CRXi)GDmtr=0i#+h>dywwvVWp}8-NB|1+{VZr3d3Ef+Cz8DOyN(0Sf3~;aQP+p_OW3(D z*s#I!E0+Ij>h@n0|AC$JdYMUE_gbtDckRnln?Pnd-VVzXpL;G`8N|kGF>`R`DyH+r z@_N$~VxSrHpRp~J@LS&(sxv)cbSB#yMelg=D0v&bJ`&oNuHkG_D0rqLSv^nJ*BN2{ zlz=sIDSl4>d1|b6+~LYkG?KNCGq#c*ZzFu5617(h!;cp9t+wx%+5;r@kYxuHjA!(4 zt1tQ$3$&1Jo3{tez!>-E49>m{K*@Rhm+!s)|BrWyC;`5w`6;n5>{B#+P7dq;N4-$P zbK2Hea3hi`7-OUp35>E>tqSKkfU)TfDh?M-uQw^;CWTc<@+(w@X zSeyHrqkZ8IjNK|H5mT{#F|(|1}Rd7LRcyVl5txyD*D%`SGG* zw!*zjf3~CU?19EiFD7Uv_Zai7euci^qKh6M`k_(JSUa;Avs0;0>N-x!d;@BRE@kJk z7MideG~8Jr467f4EAJCYKxH^vE)`RUX4LiBGd9n|U)MF0Zw!kt?Q4D^CxsD_{r>hT z96bXw{1!@G{q*&~GlmVt1UXWG5|qa1<~)R=j5waWp5BH^&0Uzca_6AXbonDnZ8y=p zxoL$_vpib8^m4oJuX|8%;}rkG;5_8lcVtK@vA~$d-*hK69#HAXS>4Fu0$mQIr|P21 zsNG7g+j1)oWu;VVwA`hjajQRS*9bjm%y6u;C5?IT-$F8#sINr>3RYdbc;=yto_J-@Mf`j)_|4~O!?fatpUqU_ss`lm1yHkf(A z+z0jHclkM;!~9SxPE&t3!+Ku9u$G^R%K#1LZ;4$V-0ccY1rK&8oSg%I16s2qaz~`= z`64njK(n>V#CO}QVBoU0O5%B}4t$pPcUAYw6dJF2@>iCv5SyPTvk#OLCh-iXMFz{t zE~DN*C*s{XVi3|ld*4+N%fHtr7Il{BB*D^gVzYV!8=g(dsX$E}(+jOxU7~ubU4XTy z2wv5otLw!t5dcZi}HCr;nq z>+5tzVCs1539Zs8L{am*=v>nc)KThX-$H+&)x>g`^2MOVgjJVIhsuE%==R9?{0!>f zi|(DxwFZ*tcbaeHzkBJuE9cuVJw#?tuq~=U(;l0*13sKY7r6>T)Sc zNLaZUSeu069_pQB+EfXo*8)HEb*=zmYd&h1eg-SU+{aEPY9UHQ*=VIM7mz;;$a>8M zL4TCqxuE%5Xvs_7wTveL5hJNzyPFOnf(Y&LLk(S6yCyNE+w&L4a#&2W&!-ytKJ9Ku z?x_OmtjSL4FYE1m-@W?J12JT1xaEL;H6)_O+MgojD0uVh=&iPOg_-WC3`?n}& zhem|TdY_Qa0Quh7mT)a?=J8|ZM9+&F~7LE^MV+&`CCMqH_6|;;)BMw^IY%T zvj;`3D)8H%|B1d`i<-$&eS>GcR8{qSqcQ69b1zbrQpU5H_VC%+TSJ(-v5B)oDCDEP zOxY>>Vd&*nNv+4CP;h>Lr8sI45%u_J-Ip+)!Cd4=Nl)=`m|hAmJu5qnCVRPOx4uY( z`J^;X#=FC4PI+InT^v?t;ooTVBUl>4y;~O#e{8erMg?P9UlR|m=be$c{15ETjRf+Y z>wnq~z5Ktw$Kve2N#Ce^SUp&~Xx6WDxGSiQ+hDSB3e#zBeQHTEobo_Jp~iQUCvU=J z&775UY6}o9g_JP6|3MLpU3=hy3u^lN%aiPn=@#~O@2+|`y9mU?cf^a@*Y#2UPk*Jo zDL0T5vySJAmTYToywRiUh``Y1+Dw}UtRF0arBBH5!#(t6qAk*o^%*Q(f2L$@$qcEP z2A$t_w;;03>fwi7Bvep&{!`gYx_^Oa`MoxtXQRO~wn>yFLI!m!r%1`i20_(_KBZIl z5)pOcN8N&q9ke~Fp4nca39(7Oj} zx22%bA99Cv7WY62IYFm8)d02Zcym!lhlXQPl^Jb4u#S&SALkgW>?Yy&biWi$FYMZC z{z4O#{JNi$5TFc{$mo$o?E#pk+P`k$UB^2CQrE^*=EmPJX10_6<2NBRS1?&3u!iXk zu4QgtzL#+qh;4s2iOBfF&u2U?Xucep|B4hIey@qzJ}OmDZB2(dmmMW1dtF#QGPGStr5_-;{jD=N#L)D!PM z4ixS`nZBC3Xk?Fk2Zuv5kXerJk1eL_^Pph^VV)Aj9cb>2Y@QF>4?Npo;IP^uNvPW{ zCE09Z`+vuaW8U^&q7>D45$|nB9ra!IZ?`x@k5Kro;M;saWinC7ecgs8`I)HS z4%%YxeSeSe3FJZC$uYDBHzK^=rd;ry$sNdTl>{ppC{02V?Z`Ubint+ ztsYf881FEz^xTzL?RP*J*19=7S`EY>Ayk?D7$2@6jqoge;u*p>2}au_VSOFQSvty+ zb)(P~PThWx>WO2v?&ADlHiBA@#vlBwvImIIPM_4PU%)dRav$P6yA!yWcMqgqy7kXsuA6}Fl0jlq1w>wxK9reDg-nMkp8mi5LYG#>yuz9+%WX-)J z0s0g0!30}N6n$sHhlM=^h?R*kb0%-#s|kb1mzFsmv!WK;2D<}Zyj>#f&n~=f3kA`5gNOdY_FRk z|54TZTjoAt@tvq=uzlA|?;PrbzA<9>3WUSYp9$wZtW+>@} zW-BK{cG_)XJ>=L`>!=AL<(N}oaX9y-q?nU|ocB|f~mSPPmpu1r36 zJAq@kD7Hg%Rv(CltbNL@bbc%lZ{3$Nb+`&Ed%hokl;Q(L`(HlIJUIolWwxWjJM~fj zw!5}JKlY$9u9@aP`{iIYbiAyxSP2b;gvq?JSf|qn)GNr=n*0pMU{m2zuuBh>dM<6U zuN8qF&nt(SxF|5|R1;m1_yfl4N?HevMuB)zD*E%FHsqN9$Y*#v1&c)++V|zw;22Xo z-TlrMqW+21Q+$#*RED=MxuqxugqhIhqv7g+k9c<)6Y(D(7y^RlSYDXw8{RA{# zTqjfIV1=watU@{BFnlCA=IEC-JPu8r9iQ7$br-tF)m9Y@9z(xgAye@^te$Q7pv|#L zS0c2?jIl8_+MyXo4_$!;PwYO?50EQ)S^}dge}3;Bw1@flCvwdR>u_5fQ^lI$XMuJ0 z8>S6SySEi|B2teD)!@JgN)xh{@|K^3!9#d1jucs-w3`t$%<`e{XMXC#yc;k=zM#4i zV1|~m+Y5c(pGBQlo2-Njhhbre`ORZI8BJ~3A$~(L4f_5%5Epz6pwZg-zQZ^#%!*#` zg`-|jxtIB;!h9f(L1COE#^0^79O$G^(JD+XZ0OR?OC$iw%tr z#f1cUaic7kBL4sa{rypW2y=h_5Z32+q^j*{q;&(1>8&DP=bY3KCu!^vKF@7J_y=r-UUSR#bMf!F9r3ugfA6{>1xUbV+ewYw-5npkp z>#ZSz75bpLrxhjdADH@{_Rl_b7BK&k#_p-Sxq*A%iVO!v zhQGZf3XR@*n6%d#^Fzmtq=cf5bfGaJ+df=%9_mY5Ij$gu<=3XJh(6MMz6iBkk7RFN z{ZD-kJj;y>FFEd$Ve~Th9vkC(Kn}c{HOHUs3Y9+&Wz7s4L&@J@ zM8)U%6;VN0Vfy@2DWn2>&a@j@Gl2Hb4?QZAl~;nwL^r2;Rk7lp0&$=@{9nP zonZA)U#CY2L*J_L%AC84Md1Wi4?P6 zi8}l7%o!?n!x!m&!w_{xy(Ig{VW@O;ROjL*pixzv#l6}h$i1+@X}(Jq1~175J-)*T zecQUvEDOB{k{wgDJKi4BNh~9znRWOnwr^GUe$?(zhpgPMKV9uxP;G`<#pV@yJ{)?o zT%Mf$5QC;(=L>UcAB7%=^NgH3>AZOuDNYeypngPsZeHG2B2Un;Amgw8$j^vUI=SV& z2Znon-Ca0>dyd8XV~GWw=fr*jwJUoQ&&Kuo1~KH*Vq{wdG}h=cdTVGJM&z$wzq);0 z{uCD5ilWlo!eL~1DAVS`2lRKPoaOFWY~6->y3R5@+m2`dXxJk7i3Rc{8>d3>*09ES zH{ZpD2161d#gF}K(46;NN^ejW`r*iM_Er}^8an#ViI1hPC#qXLH8KP|fLClj<>X6& zp_2Y1g7{x(mZ8M6rlJq(V~^U{(~cwZp^h;Z3wk{ST5xR3e<(mkjk!#Xw~o;Jbi-=g zk?2L&ign`|ge~>Kzt`Uzg9TSnt(h8HJ^GfTZLAS;Mdxz2bI|cgH2cDw61yPqpZl|8 zf8)2Ng>`)eVCt5qLCTx0i2Ulq*Agav=UT56M~ao4$Zbkqtmgc zzGRHjzjsY0qm->OPUJsLjQa+Wari~VY>amv+<^D4cbb>98qrHtSX(4humL> zyZ%kBOcR@d(3Tq1TEAl{Diq%-0iyW#y`sNBHA`56Gp6}a*vAfANH|Polx;fA@_}?h(yYG+G;(o6T&2V{(q`wcPwa`(c(@ zTB*M#BrV?!DZA%1@eYJY-Uc7T4|{8?0o5@@E5?Y05l^_~et*&!hxA z1L?(2i*OUv*^ur&`j;OKh)P_SKG}#y8VWxzAGi!H`rHMkiK#&NQginf)dS{kD1_o1 zj-hr(rI_TrYp7njMqc8mKb~Rxi1MwyP5-rS^!NM`LBZdd-h*hHJ>;&Xe!(++gekcy zwf~8ettYaku)4eDN|Q+WC~Z8$h3$8q*nNN@w_9?7MgwU2uEBMk%7%6Oo4OLs_rm2M z>e_pZF;mGAi`!(6w9bSSpla)Mb@3D%7)#*IBKEAa?`Za*_@L$93K0|n=1 zm3V0yhMe+_XKdDm1*N(}TbXuY`;pz~-t?Ds=Md4lF78$!aRFtg@(a|qPLB^L4~8Y4 z2lTD`oOsfm!5P-oWuYFDIPZ;nW9#FqJKB#*wXZ?uqaAS@DzUm2s^OnO@k6mdnX=|4 zrILVTS)M!HjNz;)EPX!+`*I%fz*#Lo@FY>elL8;uic-{^FUD_@1E_5Y69BVfbyW~9O&!! za6hoO0!XX(v$(FD1FG7Kq0KyfXf_}w`$Xezl)RQ279rVyXZVd~*n0on!+lIiVq59& zhhw~Wq4#6y|o9Uc@CKRm2#kUP#$FtdVkj8B%(U7zF6T)$N zyb9@05dQlPJj{GLu-EdC9T05e3~b+OK}kmbMXn7=P@H_+S?=4qK0|1R;bG$jvAuZK zPvvLj|2m`S#RvOr=iQ)`5aGo4QH1_?qgt=uoOi@@@@VQKCd=ydJROi_+}J-RpTjYt zAF|RXyI{_hYiRt_L+Dmc9z)+SzhWGBzSa2iCitq$psuv%pI^<`IqWsJ1IArfH5M-( zgV~0IpTr#t(d5{g>h3qfsNaHPAn)^gpk%*q+Wt}kJ8v#h>Yvs$VVuem^Kvu;)*>HA zs1IQLr4i-KoBsPxLn2ca+Z8Je*Ci`d-{rGE4y3cu596>pD{S2^O&OHE!=bS)qUwR- za?t7h(e}&4`ntYE*RrPBkvLeJ_-*v?^eCd7ll-$qFd4klnNK`Tk_5s|IpNF@EKfMT z`R?jQ>ju=rk)r3j!43MZ-uivE`k$RWNE_jM6Nbh+t@(~HZbVdXyMyoaUI8U8QS!VZ z9bZOxvCagO5-nK1al>{?`Uz<0-pYQyD+Y$hPmkQ~dj+|MY)nTdF#hcTkkcYQrP6<% z@46~DDe%u<6d!-D_LvjOi{HOGF6ssSt2Gt3jv^oxT;C-Was^Q}Chb4fY6I!so-2b# z8iBk`%sge*5hw|7R3F5Yz+^%m`7OsRP!@an#fy!B>}xK>Yr+N;1zwX)PM1+#oZ?sK zXIOkW;t{}9^Jf>Le1Guz3TYi5ACcR0x`g`8Q1zE!zMu>oS}o&MxMcRJ|iLqs82o) znSyCsW91rWI*$zc76c!vWKp4Pqh9;t_=m8R;A5EB?+-&0z9tMZ`9S=ruslS(4pisx z^51t3!Cx!#fKMw{C)54FtRpuG^9QNXQ)EYQgvlMs5xkeNe9h>I4=+Ep3__8-#dRxt zC$#dRyl69S946J0e%g9bVeGuZ=^ws*Xt6P|D(%(@NKV+o^>>yZ&#=*NtDV}qdNcZS zt1JyE96X$;NpgA!|SCdp#ww?DqvkvV%3hzVf{EJ~}0kny1W zr|=Fm;1#qzbXx~18@x5g7cqkU|1E`lhE_~>v)nnv;o4J9=r#5~=E{PDEa4;VZ+`v6 z?%kT?PdiJ;QKXgS#n3Djv|RGLCD?2T$9!qizB9eLsQpgsiSEo}ke^|6{Ne!|qD0i! z>vKb(_cN_%%vtOu0sUN$)cb^;+uEXdA%v`_pHKJ?v4x zMyK!%k^&lR%>T^umWXl&pRG;o`2-ZFi-cT<29*1KMq2z95f)tARdWd)>)vB{|DoCU zsv1Ce$Y&o^E{6J&tdmZc2f;5su}IG~^8fvN9P_RrOKJZqRNf&Nyi&CX=K8Pb=Pu`< z!R^z&=LJoGIG9|3SL(#Em`NH={m6o~PSMAVn#^bp|6|ie7mSa~Ue%*^$nO}Qy|P^~ ziNSH*d$mysQ@jsYVd!8)9KY-wn$2I^|1=KEGmpsGTbf^^`>$w(*v%F8b;G*nP<0iZ zqO`F6HfEGjE7*wnYm=Sgp@>Bck;0c447^6rP*!f3Qnx%1Sr_^=^fy4?iJT?AFE4@C zBWa+hFaV@|d*X;^{+)|eH6EK!m|o4;?RS6WEZBiM&iz}^`YMibWyI%ju``-G5zE4( zhUqqqvu)Vet3QT%vdjHHL=k{ot=M6x>V%dPKZWUZJ3+5~{-IY7ZlcKx2l8s|M1lHC zV=VvgAR5^5?5?-t25f$ojVqStSzz9!B=_<`S=7A9P(Iu`4qy9vpKP|n@(rYvxLS)3 zm|Y>sF<6KQV)#!{yZzUYlGu5pd{pffVT8v@ZvfFGDHIQWc%$q|| z&3W$6Rw?#~^fYVw_9+AGIp@S&oGTyWnTzrn#Ts)Vt>^NxI>&jS7TL5Xi(CVm^6%Xe zi@lJr<)MFMa20Bg`s{gRvJ1z0uJ{JaJ84LpbjdY(C<8>2iuwLAVySc`5W2ol z_7^e0!dBjB)2*1k-q_Z-9?9%{G@N}Yv}?~b98Z|NPd5)5v4> zxI*l4tgd(Ncny|C#3yiqdae zkPCm(-xtvq4smGO9>+6gYn@R{vVip4jzacV_M^Gr_&YhNGdPCJ4yw1#IN}+e1`Fu< z5>RE$5n?*ph`x0w8yP1Tf^ETsXRTf+4B#!;x5_M{=`AhhqzJ5DVYQC%n_mjE8)drq zA4M#VP~^E)OTdkB-F)_6L#Kt#CL!{ckc;oH(s^hQ>)VuY$p0A-T!+h9=9Peu*ykcx zkMT{3XYHHpc8;J?FG^`;GPcfR=kC&q91DTqJTvQ^j@5UNKI8qr0AE0$zYPPlTzfFR zaB>7_yA%8E7U=O95PVcz4suTdL2I~a_W1-X%sxzf5d9qn-})bs=|2HI`!~0r;m|`9 z$11LNOs?xck7Is5x2tNyOUO890x!>OKpnO@lULy~8cCCVMdG^!%N}0)lx|@8j@dmF-+`3=-3N=e3i7e4MD6k+2_V@y~qC z#<0EugJbT&hlp<=H*KWACh0pG&3Ytw{{mgd1P%GI(Il?q!^{!yEm1$Gkd~FXZF`_K zD)tU^HF&ImsC(2NUg>@Y6ycEzPYxbJ!x1*Y%DG+8*B?CbUH<@@$a=hG(}gymW{)2_ z%d{JYx?VK(J*M+N5aB4PGUIpxH0ctLxw>mZ$oLU`M)zgZ>VMScukA0G*B7#^-}>+U zw;VCXZzw^Pv-U+6nKV?e)myO3+7Cvm$Ggwk-h-Sw2IEFjE{Hn0PjBc(2#&dX+qKE< z$MCEhH^Bl0)4`n>zFV^0=ORq~-nD%rOAAyf9TyoJnt*X{2R?JtIkceHV2h8}f|2ye z&tZ5B*P$dVJ7!%yhGSJbcq{)krc+C5)sGuDNkYk|hTlC zb2Lq|;xxVy5Azb9c*j4X^#7|_5KZ9t`|rFwv8!BUXo2Bp*P>&`DL|raeko6^1Zw%9 zp=IQIEUv1a=QNLZL_vpdoH~(u8&R@ezF>7Py-(}9v-0^7$--=Hr&$s#)uuQ#Y^%XB zY99GhAhJGwCD(2emKT;mGg2+4!-be0^_1+ED5}8$7|yb#+~=kD%R|(th8PnrM?^cC zB48eP6WWs=g`EFqFKeInKYh0J8G>G=o%q^@)#ZPE9X6)RoQ;yRl71dyrq`X~87lX) z<%M{_*ZHHO@ZdC*NY!QPGtuuA9D~o2V`jp^b?@63r;bBv>-Zj+AD8089!E9D=4U>S zck-JFYPd=;k@-aTcX2Fld&}&0V{vh9&Epr<>6kv-w8xq8-sf5W`8`7}T@pp6Sp63D z)y%;3qZq&|J&AL_Nsp^x{PsrI`~Yo49m$v(S1KEYXjAeelI5uP+s?fz@nI`WI$+Y!%!NU;ouSoU%wYG#nU zpG_;8qY`2x_J#uO!>9c(uc)Dwed5-v(v2`&^jjFui0K{7-&g#WwPJ!6GG20hRWpIX zj8k8;Djm=mL0Wyrryb9DZ$Yas2J_218hE<4v^k(bJXINgtq9K&Ju6T*BMUV0gkhd0 z5sf?;76@hN2g*3ad|ccLAsQ7v@jK~x9hA!9SbjSayZ^~~R#!V3{-6@7h&srdRCMB*6y##yihMxFooN;lB^`JiQ)$aAU<8F;qGE02SysxWw|{_yjN z4QRw8&Oeg035X0sg7bI{)b};TVYlpV^lddQ!fl)z5$kRgjXQ3HtSm*T%zAO4gyBpx zZ_2=E@!#fW*AM*j*B+h@tC4__U7kudS8IUE?-R$|wXQz~p2>jAW>;((%H>%U(b@6} zg}>jJ=pYdew6|B?gfi{WQd@pz+Ko^|KI-haf9F}~=UY3gbh!srGQQyHo0UfW7i6zp znLY`G>}kc-zzisM*}COU6vpcr?bvySb30wn3&&u5Zg@oeGa{eBxfU;B^%nEO=T#zH zc~L`k03XTuB@C)+9f;=ngJ(Y^zmP*ZgeKnXp{BNCy3HA~{4YzFQV>yWx^7qE2xJ(x zZ)S+0`)AOs90-45Oxa8kQ7%hlQGYx@ zB(ayhvo8CA?D=JOtD`WEiBnGH>PK%#IjqNxpY{dP<1V?o#Z(+??m*4uaUGayII=H& zJH~&R6XNBwaP0-+BafKijvA=^oWsAj$^&xQ1BM;g6k$Quz>%RNAEq>~o!%8k@9P5H zMnfTlvbvMe_ZcWJ&)`|o|HW(Nr3wOu>ZtGb_v$kRqGJK0%84MWS#Nwaf-o-|>V>+09imx$^c!k-O1x?qZL+_)2 z%x(T8k>~|q&aINmJDOoJ`N)eB4L9g(yg;Q!?1U7p09D3OI$s=3ZhC3)-c)l-X0{pHzfCN(opr#o2HNdL-1BIl zukbGy2c2gFV^4c$tbZ0lcNOjMowF)vUN%SQTIo9&PRidddNc(FR+1~3)G&OMba`*5 z(<>)59rxQF0%GwD8?Qd;mpuyv>q8s)E@nYd&L+3_+s+|MDq+Yt%N5O~G*EQ@-+nF; zg(3ScssZg}#gvoT36xkcY(cwk4}B|~F41@Bb)zu1`b$~wojVNOZawDFYXD7QO0+7M zQ+QTo?^w_3U(k|&u0Bs+40Sg>CM`WSgaPJHfjyd-zet*>V>;(7i3YnLRoqp(j%Xbj zZ=9t*1A*lgC(o0IsHFeQblR6es5P~Gp0kGWoXf52{k_#h(5P$BpK|_nePM7+u{=(i z9tF^FQzLbw7d!M3cwC~|IH7w{7ve8qJlGO@hFiNg((~nL+R$Z=z?6(gRlFG`92oy5 zZPvb4%OwC&U+iFF+>G^^D|uznL=8CyeLjr^$4ZazOE9rhPyiq=w!YV3=roS}ypHzAVb3IPqt-nEEF{?^> z{lWUYeYPlfW~ID=tUT`ziDBU|6&rhKZ=nR9In#PP_>3tUyLoVYEzuhdIBio&ec1y; zy7dNyujus)PDEV<%_Rcj3e?{LT`-JUixzjSsi9Ga1v<5;97s!$WhduVxQqPa|jzy+B$KFukYm(Zf=h?~sN1JqvZe)Vw?CkzIkT$c0VN7Si`+UCQ@QP0yK z1j?x4CkS7+qr?mVYe}V;Jhov zkohc7w}s^GIEQ^cZl1HxAixz-ep}C4m2jg4wOw0S-0PwER6%Qr$X_5jY+T$PZGj4? zH!KRH;(#c#ry~7x22jh?7gdd31LZ^Hm;1#hfJz*>Za!eh%b4i==vaNXwy$wxqJK3>UQszQ}3J5h}p%>h(ilouZT~lQu?Z(zmJ&h zvz?AhLyNPRtl}$AG?*N`Vn8GOwBs4mjijNasNkEy5EgxIkH>> z-vE8h^f`E6<2HoN82|X}=ZmIYrf9&3`5)w3hZyenR8*UPN%y3}8W0CEM;*2o110#5 z7jEnNeEo8KeS6p)S-?Nx6gp{u?GNJlXMvK7SfAf!#>+{8;aWh=+x7nHNp>J#{hElVuG3nM3E_yGa;m{@ z8-^zhx%JBhXRp`Uo_JqY8R@gWK4;`?LiveRtd9-SJU@}N+yu0J0uh>0Du~j=N&U^t zjON|!Bo>riA^p174z5)%7|}AHKFs?W2FLmL-aN>PXg*#C?6MkAjm6ZhlNSq7Phh=9 zDq#JVSGnI4)o(^a&Op*xqooDZC3D?h!t)2p%#MD}pyvuy>G$V9?oGtAWDlwY3e@0O zSg#gN(_UfkpBOz9nUId^HTXpAeC`98R=9M`fPUYglD_l7ie2&UF^AT zUgv9fezQ385zUtiSYA}gK&qf* zdVLE3ughz=rjTkxWq(F0%E0g+;^s%UZ+Jh3-g@WCL1lVf2A+WuS?{w`2F4G#H#v#? zJ1_gsyVDjhe`=|_Kul-Q7RU#L67jqK`D>~SJN9R_16fkaM{4_NSfv;@g$y|ZrRVD? zPdX#&cZ)&yLh=y)ilhd}xg4 zoeN_sy$%3qSEJfT9n64qX+K__<0p{KqaL1IbOXu^O=ZP!2^Grb3{E}%w}1DzgnHG! z#7lVOjS9UC`guiD z+pCVt0S6-g+IjtC$sK5?7z?Id$LiVMs?{P(@#2c!GEFrFIA z$8z0<`zB~ug7ZYiek@_x9qaZ9P*{rGi$%DB zB61#gH9{N_RU?09$YFRMxuU1HK=~U`y&!F=wu-O)+> zKvwe;yWm}d+PGeQ$kuv}{LXHc);hnw&pF8$U01wH$1$L?>#v&{-vkQSMZL#xs}0*X zFKL@fPkW;l-5GwR9ISq9?g$H$XIL?e#B1)^dh<0}y8HVeO^=wi^qsw zdpU*nVe>wg`HQtAeLX*IB6f3p1(zMF-|NGvejCgGtvFk3Nqv6`x^*9%j0nPT_HRe_ z-481f0ehJoPI-e59ye;pC<`b-5x6Q>vP&~I0?8$+< zht(DMsWH09@!y8Au^RjasYyKZp5VQ=|2Uz!ukmx+Ch7TgSP?ibf3PGUJ0B125}SiD zo(+ld)>o@ytUjUp7@Bt%i$U|R$Bg;auzeTGvUo;_k*)`ZC|Yl`O6z~)Sqk6QGsf>kn;t4Cd25 z<{o4Wg@JFYA9VPIVQTaIM4x69ntOXJQrPz)G~D3uj`-}2iuah<9n=fKu^GL*mgvxp z2HmL_)Y?9xJW)$8t!rOl&5uV#cK2Ucs4~GBDoOzr5 z>7*2*wC`?9F37{)OJ*eGyC}U*4^1EFu6&-t2lXFY`?ZAy&|q`I9m8ak;#u7HHoTHx%j4qY8uj?4ef9D4TC$cXn3@8hs-W zQm8<;2Pn;3V6vjy3MzE=bg*iaqt1sBVb5CX@N9ir`_hwttb6{jRAJ}+2l6-;L0+-h z7osq4H~XeA`hV`pNIz-)PwVP6>EAb6PMP0?>q4cL`sQC)+_Ce!hAdrmgJE>j%?rO? z7oz5iLwUmYr-2Y~yl)`{)A>!;i#@tk^BkU4{?j4uE>&1k?Cml#HAibjO0G*=u(-DO zDsT2}qCHyMI<$Sa=_DFD5bn>?xNg2s%f90*=J{BDs!!(1Zly4P7^&T;nlcm*eOLMW zZ@!}U!-G=$ch^f*F#XclZHpfQt1-RU7{AkY-n?X3IJva3q1zHDTg=aG%J)UR?RsT+ zBMltule;)g{`Gz{O|F6_lWq@lr_D`I8!y7>mRX^aTg_RWp4p%-@ZQ-^ zK;pjnAlddIo^k(Fc*O;K9D_R-qjGFF>LJNLo%oL7kTvOgEuPFwF#Oc(sKkkVuzXVY z_NR+=h&=ja?eW3yFk^bkxMlnT>M==*bz{K#z%|{j=({XLzmH)Rw%CH&bLc;j*a4lP zsH<~Z4Q-?YpbjeP_G_Xg$XsZ!2Pp2CkC_AKWN5%j$7JrEVA95OegR_TwCfNd$||Le7V)7 z>N@|mKiBPd62tk_Z^9O+{?N&9e6KMb!L=vxl~(HGXk_X$>&py!oiid{lo8G{n?y~~ zOOpwD2xxeV9yyD3{TJ7*KRG?}@72P4i2MOD8lRxo{lJQ2%@@9qO=$GK)Ryh%mC;bz zLfw_3moUAF~uHaAoHOq~cV-o1xRK5|0SOsyfS zIL3d^8PSqE8DWG*SiEdr3+LmQ?=E|@21)`kcDt452LU{z7mHAF>jIG9M$ZPQt>bYZ zyormQ;vKBdEOQLUV8;Qv?ip&;VS4Q(&xELj;i29f9eAcMZJ)*&0-=HbSfZc0C8F*m zfAp;@fT+xa2mG&8qBoCtTPfFU=F|pOsi?%P)9*8>q(s~Y<*Su zwnlPa!7(4+z03Y4#w%W}I5QQgyFL!BmVBYE68#R5O3XIxG^Ec5YWyV9I58;;OSD^} z$ILKYu%^NxCy&?tFy-_-UO?ag>i^*N`tgOsc=j&#(%)ZBLivPg*pER>cWBaK&#aO7 zap-HkbAwd-1}2LeJ#7cC!E%3?%>2s|pg3*0v5cqlc+lw0kS_-hv7wGL+wCP8T!Hqx zZo{%7-7i5cU&DMGzZU>0Q&H)4vpt$Ozy6iyjsfaDQGMC5JqDCcPlN6ec=p9re8L?db;+is?X!w|>HuQdu*-;(L`@uqibEb3D| zxl#L;7#e!)Y18v(c5|u>wqt205Ph+YbHHP9+9ebp9TS-mJ*X+t7*fi7ulS zPcYoG%=Ce#vjP^kb>9m9;eE;!NNrn#Lo1iiXtAH}%H76B>E7`mIM@IUNFR*q+H(=lR3j4|eJ31Mn>a7#c;(|+ z3@5m5Wn%v0nv!g%WEc%dJZZP~nSFz<@G0X9cn?$@Bf+0xJcMWTIaaifX%|o`hk)`DZBrjvA{^^SQH6Vv{c%JM zWi{00HkT{?-~A3jR+$si>KHEGEA_tdpj0JvJo9p@k7jpcW)mJwedXNTu7^W? zj}GL&F8k*Pr*erY?y`re#kXw(k^(S$yEusZ6=tW%5kiq-#h0LOHxI7ogcpvXxWO`<8Sp1Fq9h=_78XN`s&&}~(88CDl-wI4C~5QR z*(;(+XzGw{pyE;n45jdMZ;-Qs!P^sOMPJxLxAc&{7wa1!dmE2ey?uq!%`LWl#GS)4 zP9A+*^6D{^%$fzpq@>^(v^Sm4TrCE&$_JUXBUce^TCTR_?>=b$J+@8E8RNCh;a7eZ zW>mn$(nM!Sh32~Tnp=DJ#)g8~OQL?wQ<-j6JkzBg)?CU&)Nq`Xvngj?{VJLh&*GCl z@djukIir@(Y$}A`E(LY?QmTTA??Q2!}Cza19?}}r@ZOS%X$tV_@DJ_ z`Cg33PmUcw6sL{mGJK9(^u2_IH>c}1#GHVd3lsWVn2k`6wJFQ};@i+-Jv1J6<|3l9 zjXwxWehxWCsz5vX&n{KICyXg!aZSsQUwp);n4S!=oKbJ|pah!6B}Md%PQ!H&DJAad zLv=egzc?ABj_P=&Z*|RGfdx)R3vLDun3MS0r=xHi`ZhdLQV{r`{z!(bd#;{o#xp#+ zbY?C5KAttoA|n1-78-Oo!@{vy92GC*tLSwFAkxcfju)3nFvq3Ddd$ZWh_!FXcSOvf zP{a8B-0mbqdrc{SZ-ME7ca>Ni5q}Z_eU@Uk_p5zG)OUX#Hjci5p^^T{6%S7|y!ttt-@($Y8YZhH1loZx<< z0x92bw@YOCqN$%v#~y5RfYHd9urNI>7_*hJe8wUQ6Y8;o**dmRzcst^N%bgXi$9DW zJRSpme+0+9cJd+G?Es0^UzonyguJAu{XnkQsqS^SGa)komJEjOP&Wm}nVhD1c{X_ct0VaEGDf-sSmM1*m;5!EQp$3{8{p{|G#tJE?Vd^1t||SnoP^Nf%ZmCznk+EeY~Av zd0{zH9oT*@={RL<<$|UqPoc($S&1bF(Jqi>=lb1Z@tGJ0c_UnBC76&&0$zw~(yM_WZ&nvyD%_;@H{}{um z+&d55FRet%+5q!oKjDg~{w{dN{tuBM+iZb!f48Axe$4w z%NIsRFL>V{EdgrHM}zGTqku9}IPzu3M*81H$+(?lJcQLH=39j>-r5}rL)AC;#50ya zgIhLBWuX%qXml*eeM84Lpsj?^Sf~;JQ(V<=olar8`lR^BfpKT$}-${Fa_)awNR!O*0=LwKX1$w%l zJcH5noAE-)SUljRh5nf-eP8WK92j;39D}_^ZVcK23B$0caPAYcK6K20F zIToPYX4XswA$pz%&y;_LvD3&M>TWiQ@wL%$9UODjVr{n(##`&iN8ismvH|(|%^lgn z82&@s_NUBc8H+EX#KdLhx(Wo13kZ9f}-AOlYH(dM7m#fY2*|ys!UgwQ6ots+MuTJiu5R=e7TnX ztaTd@LUM0enN1-)9BBFOqw#P4***+i{R;EEU&4bu6M*PEkz>EF8i>8u9PN8{qCSSX zCF<8B(5LopS51Zh5IcL|RGB$i++r#7nbQbONr!QK)@6o-FB^J(jYk2=g{$`T#c&Ap z-STm&|NqTvdsjaCVFW}Lv*Q#QW(ZEKdC~Um9Z(PU<=UEJ^|L)SaT$zHe#76M80m}Q z1BjBQIK^tu5Cm`t+esVd46DE2+|MmrC{PSP^UR}Nu3WG9NAB^v&;~{7z#C#}$Xal1mO-0>) zl|U6JNxb=C9tGSv_|34+8Z{<=_?3&KP+s0!%Inu7Xd<#eu>4gK#0*r=r+kUPGaHtP zYq_rLD-O+$iBrELv3`DI)x00Y-@bqX@pj(jUsxPDKj6$7{O%l35A2@zwQ)fO_g`8} zu4cl*SB9GH;uuai(47Ccx~mxQ=boE*2VwQ8W7bceOE5hGN_^|pq3$VEFTLsZnHYMV zAgb1!w=8_shCPR6jKXC?0@NC8c($(-5z`LbVP-CZft@#spF(_X(% z;7sWC0Ej$mWy=@13HnYE$m;W$UUtvfZiRD~j{?!&=ajOcJnB-dJyW=7iW=}1T5Nb9 zRQ-3uHQumn=rJaF#PjzcODV1m4T4yG;r#um&b;6ZJbQRt4@2TZPU$sF&jLK2HSXS;xm|Yy6+uRkqi~Rn!fnYE+}^Y+r6br9*EGL z{^5gDKc1oYezsg>ADV8uHyU#A01Ufnn)QuibpSJ~zr?8*7_t2wTiW^6)(?^I@mm-0 z(RFR{EU`Ckb?+@ky|mCzaqQpF@b1yg!78)REJ{qg=0QS35%PX#geFn$-CW&(pijta z&%VfkDGE@NkL~hKBm*_lyXQFhIZz7SojlxdQ1-ow^Sg);(A@gE)^0LEuE1XT-`e!P zcBoc*?YdXZ`Z|Eo*sH2H|MWpgJJ&IXw|R&-V%oZ$ybiC&G1=H%>Z+yVXOL>wlzs`@ zM?iZmAMjxt1vMvFe>g*MM~zy39Ly3qs9(xnvPDH4LIf%fv&7s+q~&h$L2Wv&4g(i* z{aBfFP=C+Is*9tmFnLeaUT#eSE#MgU^Pa=>q9@!o$DY84L&jt_UhlmLkbf$gu`Dh=vk}S|Z>5J*d z|3GxyaJx1G=Kq%-Hh!`wWQ(Y**UJ|-++Jsg z3IyL%csQ`WuX9%-Dk$L1<7dY7pQ~+61<&Z|;F&A9E)PnW0h##R^_SIiAmr}aQ$PC~P4y@F5Pq%G z8Abt@ULH9!k`K$R70jLWbo&JDe~#WhoOK4OQxo*{Iq$%<4xefve-2Rgh56*|Wd6U; z!SJeYlu<|C%cy(l=hk}%==eSAWe@u-8S(~#MtbVGL%$%Bgyz%4IJ$p`h<+#VN=c(Y z-5d5veC0SI&B!a(5ZG`mf@hgKc9tQ6jcKi{as=v&Rx&iXPXcNy_44Z%yVrexw{SMM zuLBpJ;nnTqiBT9XIWzw5>D2Ikh+gGl-`m{`3)DATwjAD$+OOrf2!6Z^Sqp-?KF6tO ziN@Ep>5LE}e7@-@-Lk=KZ7;_xZHrt#{VF=QR5+v$MgN4m^eL-k70-FwE^1ZwNU^ufISfGgk4T zq$+4wsoOFAv=7b8_)rLL9cV;uo))2o;jm-df4;k_h9D`|=Ec|+OrK@!#|v@4nkUd( zpb%cD{&U?vtUKpeD4YHSXt8z*X3BIv2dWI)X`S_L099`kP@H3B0n*`X!IX$k&>x=c zGGWC4lqXdL^U5=L=Hk@PyJc1I3=O(_pBBxa@rwKvvP}vaG>nUGcqtF$`(Iu-M-KqS z{4?*bJy;#u%#;2}$6Gm2zu!v5FYOPS$=cdDb06cetlrcAIG&C1X5wx-{%Iy3LY*hp zbRE@&!C$X%<4mFzBHt*ljJpv7#V7KLZMI9`S?xQMA^raU9Vd<hWF-;At{fLxNmlkGIhhsY9xg0h!hQ`w)Uathag@F}rf%=e-XiS{D z5@(zD&uRoV!H2yg1xCK%&w^T#=9-YOhlmIYDYBh87tI4`B?!+F`cZ= zXN~@BGXJf6+aED|-|oONEH-5P(jEK_gC?su-+iSh(Vl1Qnsy(a(YE*o9+=>bpGt#2iMR+nS}TFu<50mr8%D5UL-=og8$<>i#Nf z#aVYB#rTI5^X>Ol?uFsm8TlQe;=@p;=nKb!7sY7EacDz72c0hgUlMGmjuc>e%N^F? zrX~(1sB?x@KD%}&YWVD5>)A4cQl|t5J#2p>qTc6e<_oRp*Vkv88P8$)pw5!=N;n?@ zbvlI0wmTGHsnn;?p&7HIqwn+eKYcAgJt<`x#X%C-xjs;=bw%tTj>TkdVCTv%7;4ul z-cgc{k^|4jFHy>|b!+UdTe*`3MX%788DRjQuAH>sBYgJ+YNi0w^RMpF-KKki8$htc>C zMpN5gp+Tl-dA6_KhNV_PUC5<+KJMso&M^Sk^!p zA|h!$?|o7Cr5C?A;$?*z66=-e{kR zW;emJM7_{zVGBoTlFUWtbW2hDeCUNRt3W)%{(>VHFP5WG({0^;{@?Jd@vjAGB{o2C z;osd;u^Hy@_e}p-nV^L-Bf-Z$^n4^uBRwJ`(DoN8uBDXm0Bmxlz; zY8^zg#<a*T;rZO z<5)J|7Cd3s2ZcT=?-;05m@9vHxViEyBGRVhdxQR`9{dv9?xt`RG~ZjBDfWOB2y*+h z)CD=wcw_wT<>wghuJ7Ia?}s0}5Mlc9q!&*jYJ1+Zb;`&T#$Oy8iCSfAG+wW91n@h^uz`T4;u6EXlciY@!d5$~!h(bDBf_ ziVEAi&xr#$;_Z$;$|`icG7DXi(}PK0|FJXdVldy>obvqU6*R37un@}l48~bD{#7@D z|NA{azE|?$IRV3)hB>ZD9bwgn}h zbrw0(+BNEvmtce|+#@iu90rf@J{5Vyj#?);D-GC05xF2K?ui%1qgs0s>{{dG4l9dE z6NM6V9ZvXh1kdn{Gabhms?}_jcpV1vZgQ|Uv7lb{8*AH?A0p~G9|b9GdL7xnc&PYi ze_t(*IquYLkz^Z$*R_-1edH-jKh(TjwxW)P?#-FFeAo{CH8*D4H0&fgXg{M#TWnl`%S9s%_pEFGqJekhAtYt%R}{rqtG{&pRKh04VsHyGlfR1zF;}T zaAr7M8d}ml&-|8`gNakQuel$&q8h_0(bCFOFd=;BSj9y-Jo6lD$j=s6AZ{{U`VhYx zIwuZm&7~iOu{-WIarN6!TjS{_PL@4*#tzA=i4ECkZbhP6A^-8ZbtWv{wmW*?4f-wk z;&k;foO$wO6=#^c8c^T1SziwlTet3GyE&9~?qfQWxe-=*j}4fyd5D@VbREt@L^lUw zTtyR(MM5R;O%a{vh9>Uo>C6b^!_2vyv(|ri;n`F+#C`WZ4gI*@2THFi@$6T`7+pT^ zL7hi!N)rO+P_Lw^^!tubL@oOix_wy_7Ao2{Cik1-*;Wrb+8vihbCDNY4_aBF30$pb zycgz|5Go~0b3FNgDw=qC?&v=|7A2D3%hv`Z(F413Bom=m)iES9SQ@pTmzdm9>J7ze zEMrR{SY6cg39&fN!keE`M%C3Wjw5<#RN_cY|)6<|e#_oy;U z8q^iXMr{oUh00S>;SYqlfy5M*e~l~wL)p6&r+7**dzs4S_V+D6^jOETdmQe-vwH41 zD3+oGqsu4x{t7t*+30$sKGt6hm3&-oO8;LR;N920+n092%10j00TbjBUlx#nDiUs0+=dQMm_D;XBwDTtQ3 z4|~))WkFNQ+wT+AVVFIXtgBWR2!)&?{Tb)2U3gaGL}!tN>o6m{FV4ZY5>bdMr=rHU zLwQQU0Siw0{y|<^?KTuM95n7^DDm634V5yge(ZiD1fz)%|K{k2+iY$hrbn6nvuUGo(Kw<8?vxFAn2r72g&XA`mx2*Nf#&i4 zyA9MyNUC4x-vo;ia!RWYZBTff|YlhI(&X zq51WhqZcP{t$UAw5r@t8kGBB%;V%)PSbDuMo}snFaZ54AQ>124&i;&=g5lIBr@9|t zeIv=G%XU+`OE8}Q>3zYVCn9ZaQfwnIqG`*G^De^5s8r9Zcn7%!btPX*Z~Rq-qCUGU zj(>Oy1qafsdNsUJR|4_NJAu70S|~^{KEnuEdsUK4dQRdPEaIK#<2g{?PO(|T;dT6Y zM3%jHFqeHCt)+ZO+hVB&Jtp1=GF=8um1HQl;)s|*d3Z)G!NVz>dd48KL=Iu`HeXxwt0pU;HhPiF6; ze6c!!S`Cibe4$e4-M<=_bp_KkW)%wN{&oH|G!9#um@GJ=iDw628K{x)%qFa7EK3d0 zU<>|$`HLkq#5kj`^=KKb7WN8>{2W*}?g1+~4u*^PeNez1U zbb`NOxLldq3(FOyXE^5mmZA1S(z^AJnk7A4=C}ux`}-VNc`Sf3`J-N5Zr!~^|Ne$T z^>oL>kD)$d+2~kL8jOcl-*-=R$1~{hw6Dn0d2^^ffK=^$l8#HkSJNzZv*W>tYQ;Rz z@+ATdpRv3BS>zpd&%SAB8T`SF`b{lepx*#ynv*^_*dIis5#sMMX)HcV`Vw&Wktfz4 zy#HsTp6+?9FZO`Dt9biqtX{ssV0Tw--cwky8eZbwNI-wtHFimlDZ!XplFPMk7`{BL zk?Tw~YK4X}=~F2P)0ast^jX{)W{7^*q|a)Q#!*YutJb=ab-I+Wyzh=()mv3SU7vR`PHh(m0wMQ&goDWFO zSq2`*=ynMTHxmPhMOeR)Rg+ClSDLW;VwzpE#>w5>kT348SA27wU$_njU@p>sD9_1> zNJmb~ohRv`dCwn=mHuit#&Fk+8)P-e9#13$r)@z4ZizjN?2n=JdD^vPFbyc z4~9QV{d&FY_<~R%8}%?}o}S;sF?f`dZ{KnTV(9VkFN*7Rs~WFl1xTd&1NqT#pM$U@ z1RnfZUGriQs$)|Zmq$&|;`I-6ski!|(lpk_zF!d4R%9RbRduG1r|j*E_neigDj7s3PSjl)1($_1^RsiwVZr8O0OJI&^EEUZI!oyq`NC@iR=nxof09F z`tmqp@h{@!+h`~3| zZL4zp+huEL_R!L=vcdXlPvmzWySLg811@92YU3x-m}r!Ce*a}ye00-RIC>YNMp=3$ zoa2R#3jtNkTV>Jw%f-{jf}TT}Om^=3L&b=;<>%)MITvAg>$T}zn=P<>=VH(e<2CGD zdOy2+V+zv&B0K9Da%~xfrUV0;@<%%E4FfM!<(sXAV5BeNji1pwJWD$M?B1R7Fw4fl zQ*~pVZapk!&I@eV*^Oh-w^_Kn^$n^ICr9J?@o3aoy+r*oA8P&LI#Fi>c}_0XrlB8=3H-y zKi|oXV>$mayLlTKFbpk3zbcbs&AuZQRNfU>Bms{>xt~ zROQ6}_Zq(vnqv;RARHnDqQ?21r%Zq7H2>G^JSnQGBC43{7fRNZef4 zuN@XkcKB>a??BX_k6I2FCO~gm<(zf&2Q+hgR9RrBBP>ei=Gxf_0_`Zbqrk*NL~~=% zw)6b(FFw6;@s+GTWZbQJG4ExG@GoK?*L;#da}R^1i{fn&X^}7f%BdUBo%Uc?RX-1! zSNNlRR;~ufZeY~voQH<^Ma4g{6r;+kUtYZo(uVTam0cWb&tZn|c*~BIfAem{r*R4B zI%s&tM`s=fY#ZDb3ApX zD63`a%H>Bqh+vp&tMvrK{aQsIY!2Nlg%+#joLYqD5UnPOXP=@XtX7EpGC9S7$fJn{ z#|{IKTT&nJagd-?$-#OekZ$+z47Y#oYAJAmD(U3QOX+%O(2qkL@!zk6?YO zX!VYLcNgjYA_}wo>#UfRj4HnF+pc^S`@E;`&N*p$PMAMIl)cS<1KNrQpyoCaNGA&N zIY;UIG#Jr;6qCh{@lD84_ZF(v|K0oehArH80LMsv+Ti4I9QqD%kAFM04Tz>;Jdgcl z(P*iE$LQW=SniMB$4$cU_?*%tZ6uZfl-;+tnr@)q&oF&$KC@KU0@^MlD|ab1VfO*i z*3zH>(^(BXuZRAX9+=enVbhIWluyvju z>LWYty#d4%9!3*Ym>%a~FSDwk?-=w|AF*M0i}7?vl=Oo2>t=Bb5q!iP$qH;Aw;XvO zx=6J2X_`=NjMfT!FCDvZ4T#WqJ)N8Oa@+iSapp(Epn09`<$zmU!EY(xN! z4{uH#Ley;mS`t(CFr&?}#uirw1p(KMTk7Vaj;H_3i!98KCk`|TeX4nbW|>%VFxH{GT!y1J&H6^ka zxS>=dm!n72FQb)gdC!-kWLPfxb2Ai1fRbnD92y%2^^)&~*;Wr=<5hpA6Pdpo4NGcp zP+XK?>2pkmK+`G^qyIK%MbPV*QMo2fCNCV@*L@^1gX*ggpb0ab4_e%YXDU~^U{IHj z279#k8k^pSk?@@lJnz24GgooCRJ_;;%Y_CUx`O8rb#s9-C8HhT+dnKV=kfx%UMB3R z_FFW2Aoi5X=lwAF==cF2U2!}czfEiJ?BYLvqWkr|UB98%yv9Cmm36&eM}B>~MVJiJ zGXbw(DPnQP%J+H$hapV=_Q|X7)VPi@_@%7L&wpkfR*jX-Hlak+n*XixVR;6u+FXv@ z@%1gL5FEICBrpj(znRY3YXKOKp2YOyUXQ*xR7*u$1aWPc-+$NWP!QxGtM8Ujpo&=C@{5ryvDMdr9HOz&p$U3}bWQFa&@R?cyga07zYl_MLg zgJCFA;n|V*Sf8*t%TMD5>h?gIk!6V#rt1x$?ntZL*uhtL=9f7~JT$gJO69c|IW4+C zs0{x4!u-&>dH-$}(G&6GD4_V$FXB0mqQU24i!~90^yj@_sLYwZ=!I4pSO#tjNB~9t z_pj(64A))^9T~H8T$DzLIJnFxxFJ1RB7wXDuh)(Y35qVd2S=VQT=97Xhcywp}^Q%>6 z*fve$nCl|GA1g^mLsD|XznQCX%y+~bQzit^y!p1m3FvaR|H$D?4GGvE+x+HKc zOzt69*h*oM?4AxLK`5sN-Jjc0Al zW!{a5w7keFB#8Q&9IhO3-VY@g+Vv}2xq-3=1^Mj4_*|>gOiP6P7ND)!gp7#2MagAn zTX$uoAo~%9yN}PVuSaiUI{)DicN#30RNi}E#}4x?KaZ&Ise#^;wHuy_mP4u=S>!?) zrc>9PF#Yn)K@sfypXxiVaEGqT2}N}tDeN>lP6ZSmpKk_zO)&8|@Pp{?b$yf2BuATo zQq%%azwwLqeg7Z3U2gjvy6O8B$Q)nZ1fEHS{6B|JoEN9-{GstbZ&$0bywT7L!7Z-} z-O#{FJEb71;NNJ66|1c4pA5OnHc? zOR%L-`Z_$fEgl%BGqZ| z_J{`aY@q)gj#WaLK^A6EH}UKh`z;@V^7iPiFIF26LD$KgEVYPME~v(S{=on(dXEjg zTjHSJ&OPolfr$FQxo8ki1_4R@+@4+K{n$K}Mb~joMM6xCl(M~b01V+z94Ybgf_lTP zGLmvRh^XjQe@`6iibz+UKWCIL1v2x3~W_f}xcnm8p;^G}$S*H=$%3 z>OQ^SP(}JXDye7cQK{h6t{(|DGd-m8yWH>2_7 z*A71q-b0Oi3zs)U&VA2fG9<@$-I z5~xeI@6GFhO(?(H$L=gn2Phko5^K--1IeP<_JVCW8uZw=Le{UuGj4sDC+l??=a6ZmiTHe9Sy%qzS~~^8xi-E zR=@j%#cQY|TYhtrFk~obYi`(!>DZHslw1Esl>uep*O1a2#&2A;!4-=1XaK?MLGm`- zJ^JSawk%$g{g~d}GOPB6KJq*oI{tXqp)K1HG3t79X=E*G{uA~zlBj|C|5=7}B^ejc z+=lq};$%!`zB%~Ad6#@+)YXzA%6*c~gTXTz-M8{wiGFHnQN(Ytb&m#O z&(-SbDH&iuvphDwqHaq-6_FSG=#Kec3w!FHlWz+`zgBF%zKRN-CFooF#fO$CyJ_3N z)>Jy*0?#&i_K;Y303ut9*A$2EgB6KJSsUd@%pRNNAG3^>2jYWsA6h?P_WM1{hzTk4 zAR^d^Z9Xr*1T^oK5@vtQ-yT-TzQg^V19cA>nyW_V10_^X$WwX}4J!R9HH3IrS=b!o zu@CE4(SIei+jDIW#@!ct$`<|~*8VdpieL*Hg~>t5K~PbGh@hZ|k`&w{f&@W9P*G7( z5ikp4RwO8(peUjwMUf~;a>g2R&PdKVGm#vw?~glu)|urx@BP;Ii^X(zb#>LQsvVyF z?Dc4gW8Ld|b1ZK>D>ggmf@aNMy;1p3gXUj-{bt^Xyc>Pq;?OdWf{eBMfB#M=kE@lW zdvUv@pkzDV;}SDwFZOqDboshh6ZO?zvV8Fj%X1qzlzO_Qsgmw}&GpxQ`o*F8(6|-HFc^4oAXy&u*x!y4)3%1C@=A&F4`(2S z`9;LN-vThNkwUsR^c<4U+??8WUJJFBZ+#TyEQMn|p4KCuH4mg@+;DE{n8*vF;x-{XyA&i=aF zAq&fwC1kvPHP_SS$Z6xq7=F3~}mzc&diQuUa@vT#quIMP(QCgn|t8 zVCELm6XPsGqjD#-qaU0=Q$&BSX1zHePgh=v%v43oN1ucR2(dxONoQB;bSX@;6?feg z*TgaFmUoC3V|F3&{_bmLL<*3f92~E!6-SfiI|CH@C!lfN!cNEW1aPeSH5nX*jh?9$ED$clsYCWVMn190T!qyGeH%Mhz9Ai-L z_jP5~At*iIaI5(vhO^D%%L~c*ZU|*-%R(K0uzh%3f1yZJi4~UO#+xsFBErBuC->9l zwE8)yyz9#94oh{&K6HBgc5F9LF9ep&xnR7x(;s?HwO%-mdWUP?7O-IP$g%bK9ZW=Z z7?So{n0aP}h#UHFlIrguXC_XwZ-W^0O==Vd|J*%mZGvO7lob*cl)*6pNBUw9GmP+E`TUs25YN~`<)6Dy16>tdo*NUsBT9T$ zU33dpZ?_^uXY-950hk_(uwb3&gp%RXq--8mG`scrIY3=FW@9Pcp7&>ftRTY>ef22n zNcN|$q;$g7GZI3BE{aeVVcEtqGYZ6o)VC?m-lMdP%cP|D*?4BZM6n}BAD~Y6T~bC0 z@n~B1m(L{^F60y9m%nE^9EdytiG8y8h5|?35sQaGC zE8pUVRuc0`=;w79Kc1?y|J`07-&>V@d=!u7jSifC5G_dij=JiG9WOm-e7>l5D5Pg) zsufKO9!<2{(23&2gh|S~gwTM6aT9e9#@j=#@;DXu6SF5N=Uk4OX|9L9`4-EPSJ_a| zm7TF6o)7innACGU-lJK@fZt2pE~szNBkJ&K3mWkAchvZ%h=xD1yxxa9jw+>%ntzov zp_=J?oqTC05&8blr)FJBh|I}r-1J2R$I@3GyC~~{2w#dFT1zYu^}?1rMiG*zBlt^% zSS^+xOIDF5TeX-&n|U+ec{%KSs(kb-DxDo$M$impUyxp{$ zZCD#f?gz;m)EEBWeFJr;zMm-kpa(1O=Jpy1m!a`IhUGl*|ENQ#b6(Zv!1UA`Ff^hj2koN))LaJrVJ5Qc;Ll5|Br zq94mA>eRKroWVfzALH2>7@H;q8_@jn6m_%Y1{?!ZwHEbvAL`q_ey1%z#;ez3iB=oA zF+as{=F@i@45#I`Zq$GF(%|B(D$G;XEt*C=f#16g683I)fN`qr4$DByUrE`(v31|H1kB`w z)NDE-2<6)^EDxUl+y8}5ViUt-G+rPSLEl&Z8{ccp25Lzv}1 zHN$tV4Mtcz{in7_!tmWC%?;8mKst7(bayL_zvr*Mx>w4!@E|m&$E0}R8)1U;*l)*E zZD{Z$M?ROC9FBQ!74>&9E$N53qd;s!!aBy zp>ewY>X(N=9d&X`zwVF5e3BDy&4i#;0W*j6LM+~#7hF6rh}8oILZFY`o9mlkHCym) z*2%{}Cf%M(vd8k`W}WsVZ0^V8J*xB#&#!1d)W;wDQ-3B0O%!RFT=@1G<~|na$z+eg z?399+PtSJv7?g9*wiLt7k1Aa@40OkE%Ux%w!)`|yfokZj^<38gva%-O);DywAr*+ErxrmJ(&2NI{is~f-k5nOw ze(Th^%hphUB6gf{R|nFxpTF2OB}5VuxT-7^2IUM?vu6f$etkTH-|Eza5@xrMKQpfF zF(N}o_}Pu67bMW&=9+Gmg;BIJCJ~S=v<%af3qoXhj2DC~bcgAo!vi2`{@g2ahY=O- zO*9VV!~9V*A0@7dkeEpE-56Dv&m-KM70KLL}C!=dvv}qUpVR4)2MbMGJqf z#fjB^hcptE&9SZoC_;kO`Nx@I$Z^QbrNj;j=j=-I(~?n)Koi%Nk4ItL{zc}FT5Nw< zSg4M4w#0bU0($)0mXACDicb9H;iCqa9B8Cok&Wg@g;P&%Pd+^d^BX!AH+TUs_c0r>qTqME!Lq2>Fub^pl&P;~O)GfHzY3>tjhen1Ap1<%`RTwId* z4zqEVHl*2T>~~o*i-7G2>Msi!y>`QtZaioF@H|RPg{VK1r=PiU0)0CX<#_cL#ve{$ z_w{pM;X{ij&OX$(SVcrH6qjL6#xtKhKeSm1%Lfck%eBbU#rO_a3y)o08^ZYChZ_Zr zo_(oC)QcJUl}{5O+ef4Q-=5Oi@1iRidu?45llD2;Pk0U)m@=8s)e32GE zgHh)pw%~oAP-nO2osPRJXmnkF;p;kmG*k4X7RAu>nUj|F^q$Al*MTS1ZVR97?}W*3 zZ;m?jMIlm%O7IUIY}_ac1LQqCSp1aAZEb#B{~&5U)P3Gc<~fkN#}5T~@&c7rFv03D zR`<6yyint9EY?mV@lc^}ksKmvq+ViTdW`1cBwnbv(XPW{x?>0Df2^~*bxk6U$y;D7 zc-;1-bvUAgxma>}m_T{&=G)ECi;v7%0rQ?8 zX2dC7l7a&B?SpTYuzH-m^0x;WvX2AFVs}@716>_uAWvq8f19IOOv6{9sodR(%BQT+ zyD;092&rB?6m-E@@zsx?U^rLSzTyR_~JV$Yb6-6l@H$j zbUlo&7reywO&ASCl3&qiks zexUF@JyO>E9Y__uS51Aw(YW6u?@;9o9IM$=$@BK_p}0)3X?`F9jZ4TsU+0noWQ%tb z(IZtb!e!#*m%0lEpInQ6w5%gq<_({6)CqB75crc01Z-eb4F~=+R+bRS3V3&7n{lJ8OH&B)3Tx$iLLA{J9|h~`hkGj6Ed$a5O2 zPs04m-(g5N2hH6Q-$=E0gNb454gFgB5bwKo%R|E%hEQV@{tPb=13$eD=T|~~lg~f9 z8<@d-*$r_?<6Wp)s!HWU2^M$h95gZ;E(wNJlV6i)S{XGtI`5CQyN;Hc!#p0w+TmID z&5+LT{t86CfXKd#9_;)G^V^qNnT&#`nIGvC^P-@uABxrsCPBMZ@^_`0UNjUOKEE}u z0ahf}&lOm`g)IJ=WK*v}NL;qE(t3XpwHuWM^(T#C?d3`?&06Nb%oqFU3pe|*bz)r@ zxbHy(`eVMmc6!H6+PE+oABh=%Zw6`mZu@(y{G?l_6b-q0^?`acsanbIz|IHb10l1& zH(|W5T_12`t@6e>Lzul3|dx02bnEAjc2~At!f;Vy1Lo9A({Mykn_+@2r+hJ=up5-|I zz(v`cFm5CFG^)o;D{4rQcL;RS0bkeVkSgulIxn)W@C=+wpPNyT^W-Ue+n zd^V27ZF9Vt#3CX&tB(2GeEh2ya=r?AjM?Ah7*B=AwIx7sw8m$rGC=?P+qNv3!HA^a zmshe)9j3l*V#)n#f@AtsIklWb?|B+|(IwSHp z;})2EooLx;W{ik0A}Gg7vGcxupy0QM`4>D>lfk&Mcnli!)vo)^hMnh2sVU5JUS=@7 zJ@Asto(pL5k7C>81IB+jFx^!2&ejOe#I@dB%<2~!Gf8I+Mh7=g@rp>sgk^O=z|hkKelQ7u41(G78pDLfD;4Z{*L5!^{nd zlV4Y8yu>itcVMS*PAL$zxpq8aw}%nZQAs;XQ#98YwmfQ2UZ{Kso#t7-HeU>D0^0y32iFEIJvFAq3@^I}L&XQHXn8fTpew-2l27x4LdS=OfzE9-zA-$DeOp33w+Nn5X5h{i zX)NBkcAd@B76hE|7`mbvetQ6|>CAHw~iukXCic=k3_64)+hC$j;{N)(+4O)iM= zRZ4uwmyd3LM8CG3kgG^RB<77gZ+e~3A1|M{+-L8h?|@|CYnAzfc7EZ7$) zQapraYER6@^ydph?=bz)^ZV5L#3f-1&_v)5-lz^o`&UjMvCi3dJ(BG zh|)OMd9fV^k4R$T0F}nphUV)r zy@*_0qhYUf5-2C{iaq%H2grMmQMYTqL~R#uaeio*LW5`bpZa~V-_LnUlV#oKqhxXNX!UQhN&)ndlJybJ z9O4y#(I{18zUTqyQkPZxCRU32&Um?5B-;U5P02XhZV64X-r`=4NrEJ`>%>mOM8KQ$ zyx_0igqD5~ojz{+Yp-^E-j(0m3o|89^z+U-RCWEc$H^m%F!CgG6YG@-JPWghn3h=} z{Nc8)KFAmdz3HLd{F$*Z`Ra{QurxQGmG!{7v^Utg{xhERaw^Xa%_jFqog-gBg#vx?_%Oqm19o^Ap# z;u!XF@r?oMc*oJPS&fDRz%lbbY}d*5Mr(I|3tn}43WS7vCjD`*QP1%FCgzkhJX0Gn zd`?gshOX`Aq7ZdpC2dm0M8N@ z=l;3UQwHi{jT?WxX~VMz9Q-BwK>>PGrFkx(EVNSMa7(oNJo@u?C^BF30#pqP=MhV+ zVd%uf*+l9H8g^(Iv9PD&n4_j{m4(|Qa-zsw+x*{i{F;I8ia|e$Y1Y&v*Ng)N?LIy2 zfZ^cl#UiQ`)(?A zGeWKFEWYAqzahd~W?dQalkkIWv*95-cc65quI{noLe7T=vtySqzKfxELk#CbGGMOV z)GOtoH=aS^gwgWvFjz8Chi88O#sANNw?MYdJ1U&_8%=FW(easnf(TK0jF(#j zVZ1NIle;_r~d`27~()vYJ&r{5XEAlE_VxwOYXs4+@hoomIjJUhq!sxub){W|gSkwP`LZ%Dj)9<$PF;+~x z2~_t@E7z;vqW+R9{?S1VG$qoyz&Wi2Bz|H}on+I0w~t;Q`?)81t2~aWL3B@QTOXd; z_vC@ztqbT!XxHkps>W7cmz-*@xI{t+ll`tGHN}t~Ylg*>rxdxaXJ@$y}2H)|&>#k9S zwJLEVqOyLzy6Zm8PJmB_^R=f2&7p_5o@w>39$scFA8JoAD!}mb2$$m@F@3C4`_U6dwHGj3 z-Fb}jFdZK#l*r1RANRm`nuoq0$zZXh)uBTDZgw8%`(Jt1i&CmH#`yBdY+GdycHc)U zSxuQgDllF|GK<+<&Zz|$W78D6M#Ok)W^FPf8`$5{?)xcNVN&H*1BBj^9{=^2ok1S_ z%E9>|8b~kq#0osAMij2ed%gzG5ow(UtDZ;!;I%LLP)4x0L}HC^`(Bbh^asg57m3{=CdcAK{tSOMNQ0RAI*=8V_b}~^-lBpYwvD3 zdp`OI3@@Sk`+N<#lbZ#^IqraHZB;aR$nDDnH_hI| zF_quhaDDg*8jGy4OnqIAV@Ycqi0b-+PIJ_3;pYC1V+|o+`?2^Ercc})u0j}IguG?Q zm?bOo?>=>XQE_Q6Dj54>STUQ4M&1-JMKEMT|2LgsbINBlpO<+f49O#@la0UabQw_T zvdfJdS9oA%$FT5{y9F9qRnC2W;SL&mJtU<5S0A7KP&cRZ`VbIBdG+P%{bBOgtru?P z*I{YLphJ%vrl&IAJ!vbKa|4aM_@3^qU!ISH68OV;|CTLqK1=8K9B(s(9uEPMDR_U}KrW91I#T6JqWj9EAA;j-+?{V$lSn#+Ej% zji}4ZE0>)26-LibQ0;keJ}N~pdT$|D;l)W9trt{Lnjc5)o@r%S zKHM;@R{AmGj|C!n|6wE?D8RF7zW0_ba0arK=nj6_JBX0#639j>22#|4#RaA9cs8L3 zp3@&+qB{4?q8WRCJX;}~=3SR)n3WfgUMc{f+8t6_x{TF@&Y$1(Ige*f>x7NJ}f>7<02n6z^jD=3J;BX~#~?9vqC) zReW#Hh5p=oH{joh*=uvmQZC1jzE-czXx=b#KUL!@BC$V; zTo?^ORjQ%njrn(B)>+_rZgCMB^tk5DR{tM7UkR%Z6=kAeIJTA|orv)r^-4Y`i#R(% zO|jdviF0)LFf=`K^u2;QhVxrK_UPnmBP>s-SJQS`F~|vK6Xox3zv2ZePu}a*rVRbH zb5uLNd@Mrslb_21I~vja+1Mcumozl_^v|}iMr~+%deK=h<^*b&mwm;znHIl+hJE>e zv;+sB!g{bmAg=%hg;I_kZkI(A4|`so$qrb)F(4|ag2gANj@s_?m?fgRq>vMOx#>VE za~pN|W`%}7Y=2aF4D;Lf#oZlI-E{(yuMhg3$w8=3*}YRC7R&FX9RAh+Ok@F(o8l)L zH*AD)0h6P{J|@so!1f{Vk0Z)|UfpS!@&XNzOGot`N6;*5?T>HE0eI%`CuN?EH9&u- z1g_PR0_2w~KCig`>ZJ~f`DQ!oVBBbl@vZO;G*KKO7B%*FpWDJv#gotTcYWPY)#K>G z_R(R{WtI+hL{Yl@Wa!1p-~IlTWPp(jO3eCrStIZU>Tk@rJ|V1xx}A2JN3}V_^d7|z zRcB~C-#~tDBY&H11W1lfL(g7b#j_5Yzng50N0r+zn9hw!qebx}rh{Tk&~oi&>F#Ifu!B>}-wN|J!N2N#}cuIyFuG&7I~lx`M0c14{Fofqs1IuJ20_R^+z;&=ww zbLC-i5SrWEkZxLd3Mj52XFdX!_t6YS1{?Wler7zA+s=%)i}R?b1qnL@iqf7_OMo;`Dfy z{CJ&X_CY`qpHVq~iWj>17Yp}p|BdZ;4wpOw&Pt+wFT01l%5?cXn7`!N!(*3E3nTKw zO4bS%3`=eX82I>!Cw|p_~hWZ|?UH{p^2a`){6X)(7MbyztjvE>l(cC*; zwj6~v97~2=)VK5o804>{ieB7_ChRsdd7n+ev-GsKZe3D}F>cO$t|gt$>$7+VLn5^68+&~S)&<{Axw7E`Xj z@|>z@cm=a7`mVla$$Cf2`^2-FhmIPpqtfkP4L9!2eiyxsXB27G#Ob@C`8`WN;(32V z#AVrSI?i}lzW2S%_wE%Sze~`&M4|0-a5YKl$J{0Yo^}1sZ?*Y#|MbI-`L|?|e`0+5 ziyzuFT8}o+{!e@TJ@1-z2oN5RUNw*K!84RiOTP`E>CuQH^lM@Jpg0V$QH#_#vAFU$ z-_!8&jd#$fm&)3uHPyfNCeQO|S1i9{?Eazcl6MQx?3+r*qjNM|A{u08932whizd)r zON%2pcb+k%p+_n5R z4fX5CU-S#0)#-pyLk*sE_W!Wwo;aGnh&}~ly9I26zG8J*`qggL$i`s2VuO_-KU>d4 z!9d-#Q#pwiUx4nRPS!npu=>Cb&z?n+O))!YHAPO?rV~5&#_=X+JU9{mHWhb%;nQ zzl`|aLeE@yrq4UtIQ`WJ<#qZCv3jQC3x~)x!D6WHn$(U{AB9kOt3~9+M&AE!7cKDF zh}XC}La+Fk-k^aCT6li_n=|egjQ@F7t61!e21<4Z3Gw6po`3C!TrTZGomHsHR_hl` zxrho*K8nFJ{tS+Dy@lyPlez-Dd=np0cS~@3YV&c3$Q8cdyoCe$J>A;mce|kp;kLo> zspDvQYt4(EgOzCh$MZu?7(M~ooj&fjk7YtLAGTyVrF!647VG3a6i4uEsps_88^$BD z2p3Omh$K|ih<3A;>mdT~k2g!97yq79L5|0htYCD zOWlyptbIFrE*Fg^f4#D)E(}Pu3I##R+(2=w`X#)v6s`F52OF)&@LqHBk(Mcd3r~@S@vPz_9M?Ar!EB?>8<|RL zH2Tg*N=<1WoQaUtdUlBcM!SWYINm0qBHQb=o7}PbH@zk2>R#^)Lk;J}4m0>;dA~%% zE+>mlYyu`v5F{F&R0K zw}=P{c3(W6Btc!3)2Y>aM3|n9K3#pfA59;6l$V`MM5CK=LLZiKXpGT$-&Q}``3a;* zuUhFoOuwVH`yV!*!uXO>KWfB0U!kotJd^n`ChjAB(CIWN5(M0C;RikEgnW70#Lat!J6?NNH4#IX&0Hi$a;wm0OUF6zy+RruCcfii`LOm$^w zxEnnCua0-|I-k*K;)gAlBxv?J`c&_KiSSYf&*E|nZah>2>ha58rCc@9P-~RIWe55? zv}?%=nE`VfP&upB!>Q{sXdrd#!1~7*P_>WzfZMuRG-tvc8_O9A9j`o$t~t=_*)47l za>&~-zV0RbH|k)dHnciiYHq}M{!pK*`}xee5gdzSs%#9mIU;y%4ZR#e8z&f^Jdh@B z>W0Y2Ug@{V{X$(&Z6IQY6Wu;g94#3R8|ZNXq>J|~zI!b|VWo{IKE_O$2Rn5z~p|NOo1LHFtHYW6J@H)GYEC-(XR6c`2LJ1m_(bwtH_Cw?W zeU*|W3?DW2ieGxJf}XE(n80T7Yu9Bo5q#FRfw388i`ca+NquPP(EA^3L)GZd9kFz_ zi?!aIet$^_6j=|;6kAU>@y2{D- zn}Ni&7OAsW2WF3TMZR)W2K1?GpKDYDjHhI7V9Aw)VJ^o+M^S}{wtM;DzVmh zcm5FF_l51Xdvx#+9~zB*b4p{2D_T%I5?mw@i1K}B&mzTG+I{%P4NEbbuyt1PWWYJ; z9PIk9^i8b4{cYoE72A~JPu4B(5x)@g&P{Rtw5^Jqv8W<$>(1GSL*w}Fcc zOrD>bSXQpW?8dmtc*eha?VodYX~iixG;#PV{}EYUsQ+e-BR|@S=5}5kjm{&&pu7G0 zv!ChCq5oh1;#gFKw@D2d;~8ha9sT^2rq`hPwV8eC{B*pHwCi;@&NoJ@4*jq8(JEEV z;dQb-jBQ*^|KX_&Efrs_sBxISO4L{;r_M|N_j7m#k>@r`+<`2YE#5-G6^5x(|jY7VX#1)WFQrIGIn?8ZBk~%;jY|^FRAV z;@rpc=G#}g_pVIWMLEw&;n@Zxb7yVD(Y+T9t1Svo&|ufl_1FwHpyYHn5&~|aehbEY zu7p@b65x5kTt9|pN%f(PiO=wi-ZFnKxW(fcBA*KFR)@B~30}zws}q&Y*%v{_!N5*s399|C0%g_7_ITy|AWx4=dj_lT%{N zcsBKbeecWmpea79mz*0*@Qg>~c}`7O(2Y0Yq?(PS_$eTrGF!G+!T4(kw|aHzGYJ2E zKaM3ZL0w}X6;ZwH;)uWxC3jli|KyLSYj?6~>1y~X%s)&`ke)r$f%(I_P99SZ=K6{p zGAI@c8$F@7$CG^$6`@I4bC&@RBh(naWjj~n6Ie5)DrI9J5$4l zhO}p87|P$`+4O!zUv@<>kf`jO>qx`rqJgb4(diPHzl3pcud3jS-!KE>5r%_LP(F2t z$N$?o)Eeqz#ksW%zD%_{WSpd+mMaB3o)1G&Z{xRzP3N}5Lf>WHPrM<>xtXjMPS%ED zfx7fm4}ew11f{I?S+J6=nkv*?2g8ZqS=V^Cz=+KBS>{iosA}eU`JM+~V061@(7I=0 zcvieS$FU_zGp_xo7^TH7rRmgrsnf*jJPvq11 z<9qs8(I2TG!Ny18I3^jn^r^-=4%H(wupAh9d^J}9md$09 zy}N5s-!&o0ih1n5k}AGAZi`GbuUH@dJHa1jH}6}1Rj7_Aew$vTc!FWk#ycF0SRjdu(^vbW1}yOPvaDUF z@h1Xx%q8Y!@LrU`lj~OIRfWFu8pk}n<_k+&wDfg<-S=rj zDXV=bw$tmH$Of=ZK<$z&MZB=R9S+zwT4)gOB7QVY#1VxHDYw;<#9 zfSS|KK^Ph(o{YGD3bixFk7~63gp{F;9~`9cI0lWx>igwBsQc{d0f!f~eI6~R-?$@+ zE}&^j@P$^j0W`_;KL55o=I0nFnGy-!`|Xb)RHkd?EPXV z^kya$k+!n~sfEuXP89Q-OkZkSbaB(gF&9wBnJh8f*zA*iN&1#`cqS9)-bbbVIHv7a z%6$8x@XRqzak!JJI7aK6zi$56h$v1&q0_hZQJ2N5%@0~-(5zYe2g}blQTg3HI!;$Z zpgu6ZGweDi5I=s9qF%)6PFAijR9@Pfi}2s(#!e_J;hBcKpQ;9UqY+8NGkQFgC?(q{ zzdnJ4YK$qd($}ld+6$aW${RKqsPMyIcaMjDUgoJ!5-!mHf)br5Qiw*EEt9vO#QaTD z&E{sIwRU)>b62Z5uSP&bUePm0qyMlAjkkQLcAP`wv&TZOG5MpmoM-jAxvy}nk9L%G z$)y50xzKTJ-@gA--?q`aNdz00x(6~|XSoseqn)gf;Z{`oqjrBwvkICn;~}iwIVzw#?K5&m1<`d0Of%p0TW>@6_2KM0~vC*JI&DDAcyw zGqYz4nkW!<_B&XOXOMmv>-0<*t%b%mWG2ti?c-zk0m}*Qr2qb%cKuN(afl>bOnFMA4>r+o9XQ6K&~tO^`p}Hu zFy53^!EWtAcUMl3RbA^?q(xD$(|52x^)^+Q% zi^XU%-+n_@vpr;9o9tXPtAiH)ulKt0fr$LWbM%Lr;lN&!%~EE)F0EI{O?a6iB-h0DqT!uxG{`I!oRMMTZn>| zL1~|>P=U?Q>i(t!?ik*6D%L>Z$>RnbBbC}7tx867X5Bstw~nHcL+7*p7(GD*Ctb%^ zH<}TF>&VgL`*#Dix~fZ}@F_}GdB7U$ojwwcJ3W>Ob-|J2SK;94}vnDbCUq#T_J6 zu*XwR@%0$=m+)F}eDVh(*C<1h#0p*i=C@|ssEwsTt?^E)fK92)E@30ftOVnMuxN zu)O89<;IAuh+;_E5jU}bXJ%E&%r$0%X1{=Lrr#J}Bl&d67XA#HU4hcaGz^8_D56={ zAHCa-w_-q7pZq)^~h_XKvrzvKsmuh_AN}-sWsXD-IIbsuyW^8#I7RVA4zyMAP`< zqC&GeOl~ima@${GL1AxXY+cs=?w?m@vVw#caSYn2C*IoSA*%W|hwbD=JnQ(xrspBK zv~gN=-&cE>=?ON!r}uCh?85Rfs>l}Z5(+Q=>L0ad68=Q|_c-I2bE9fc6=89cS=FR1 zwQBMRf0BABM6%`Y`K~2hsX)^Qpx>Z-aL;D~>TjGcOF428>YcO>e6tP&(#5gwu9x27 z*~(t>dMa1r7^imsh}>QQv%)eDzX@(dM5ehbZpxz2u>JMWQgQ+!Dg4-6t)+^n_WOs$ zEYhJO;PaWQj9C8s!i5JT0Y-FwhyRl|^IAOSJE~8}xo{7kggRd2t?eK-L2oL-rdErB zRts`owh80Wz&uI!`RjV11b<;w4s(K~S1!7a>^L;L=`*jgG_B47pl3|Gp12awXx+sG zZE@_}TqR^h?p(-0guBn`HE&9o+f>YOkDlXApY;&v*b?K&HRxft8a#Gmnq zieto9=NAw=Q`0Wu%T3qrAV=8MgO*h zKlzc+l}Xf9=K>B^+B$;aMuz5o^?vQ0HsPWDaEo zsL5|k$~*gE`1hx3$Bi_6IgY9Or03q7&(M-}*DlfKSVXv>_!&ofk7ryuKlJ(TZai~Z z{N95NIPCtmveUt57|`1G?+$f`-lAoxn2>cMMQFwz**mX`!n675Y5I;C!2F#=3mMP< z$iLG1`E07z4r;f%Hfwo>1F`r%$E(p}cxKb%*X~9f0rJv|r)QU{|9gC}{eh8GFu$7@ zks{9QzumVTI(n|x1>w}tl7yi{RJhRge zO2ge5)az185oyBYbiygWRU-W^-FNg}-fm?f*Ntbhd?cBlgz@ip?O8bmwX?nKugqb3S>So+gfSYZqI@;lw-HebYZ@k=(&7A3sYfzdfJFnxyvuYwukaFT;zv;; z@zZE_wpZDba}LjdbKtmr`tSN%nJQokj=}cH0flRtDj`6!MV8)=0J_b~JPUC9ACs#{ajlCZUQuMB|r$ z$_bSQ;_fY2f7FSe?|z)b_5lMY_r?1GXwC1B%a?{-XyN?hgv_5bz6v};P=i9k5*tuI zdVMG`?M0MN5scN}74S?7!Kxqj*rD3WQp?HiH!v$SnZ7Cf4)j)EaGg2sg~l05uU1yc z1KuT>_55aOA9Z|UQ1NhV8`n5Od6DbcyzNJ zQ3lpRJB{a{A#~Ef;ng~r)|gWHgb#yR;`^10-tI8nIPs|3@HZOL{c_kC?=rdNMn`bW%3h8B>`x$72q&h+d>Gaev<0R&X5krcPH-Px zhv7BImO-bo=5`~p;*(#0;)2n_mh>PyJ6e7(BDoo&Cl&g5h7;@59?HGPGk=}9{~=}; z&-m6P36gGL=Rk+<=biXYs7I?TP9p4RwB{?bYzW3%(f@sS z-|iT!u3^HXy7Y+vY1E6j+`jJD!Ly&;7$3Y8ji^VwAJ)I)0uu4jom*DCFq7A}oyBz? z^{qLy_Z_^4R+(dRGdd6^M>c1Cj_=k&vm}?eiw9Jo*TC6xGskB{j`j}x!k`OOj{`*E zbP`%n9hUBV^#RYU^xZ9evJ#P-SbLh^3(=jUQ~VhtHUB4X&RXG)q>Kv6{SoUsY0`lP z-*wv3XAXUt8GYhbzac0lI5F*Hc2$H?D~9i@H*-b*phK00^>y?AoY^M#JezW z<;{Qij|C#@e9XR5O@_X-Z4WsPC*nEWRYlIF8>1Q558K-rF`k`njjq?5mgwpW!V-n~ zr*!WiCaeTJJboVtDs&5?3l0p5^G09A4==r&ariYIDZz#kw>;GxzH@yc{5*}Bq zH0P1n@p)4B0|rF&HQlwV#uAYa7%K;uOCvJV^c7nRJ@n_N+@q*DML@hw+%q4O(b5&C z(vV$r_;}Q*5s|2KeG6)9`noYd8^diR4yM@#6cN$<;Fac4*{6uO?L7ZJ{|TUO7~5#( z#*5nR{NFvuj={6ol#g>=wM4BBBtMc~5E`)C6e3~#D1n9?fj3`n zsH4SKan^6N1z@VT^>lgZ8j3T!TQhxl73IW9<|a`w-o|9Hi+V?b8K7ay$#*{w-U6cV zz}gv2F&J<;X0ZE!95f!e)RDJu0>^U5a`J}{0sd^zOlmtsi__y7>!&1y`M$%9VZz4G z6;fy^yDIZixdSAwdsBKw(h9A*M(Yr_(etws@~@kn@k~Vvo?#cata;&CpSa5Zs(c3H z8a=7U=RNVPYe5;LS&Z*UA)`=9K>VrGxz$FeBn9vWaMgC{E|oDq_Q6;hvpO zd^*O9w|3h^>ox`Bt)7Wi5Hc0Q@?%GC8l=58G)JBNI|&ap$#@2fJHMYgyhTItl%!6+bpXwaMIcvW_7-*xH$|_%^a)VjXlNg z4{H2jCoPW6=a+oJjGvQ;toOrMe-h*ES|dqoD2`$BIlg?X_!|2Zq8u1$*6N_)Eue}m zvx#wQ1CD8bhUQhzWFT{q50Y&$zhbGN1K0RREof=^!N+(X^PjaI(|j;aSjMrRR1fr1 z&c-qO|M0#f-wqQ)`x8X$VqwtqV?b-?Up)tRy-97o2pa0X^2p@A4WiB+@{GKL@nTel zy&ir_y@J-hSM(oc!f=Ju;VZc|jv9D2SC^mHeubf=B6*`#+X6hJzMJdr+tPqHk1OEW zg!ylpQqGi=%Ld??-}~SC===y9p8}=Z2Deqvn%FHn+b$Z9B3j@*3r-8gh;ZFe{YRDo zQ0`7M80{vYYJQpisL%3+E=jqrO~D9lb!fcPze{fTxq^*-0Ds z7s{KV{ky&Yn}jGG%XZUMzFmLhb9R;LURV^7v|u9)JPH%XXC#2rB-%iHDW zwXpgI9c|ovyM%LZAzxWM=;6vRESr5EeSMV=4Lh)({8EwzvmZ~TjrY;~&M+jBuaf6T!-t~G z$c`wtlfkIb$s&v~jK-UYXErP`oyhV;eIF-VFBWsbgzVVV%h>y9n)E4ue%~ELJ+$Gt zXnhomWjxd8sOC?SOo%j=MV1tNhsa_Y>YSbeFkryW{?G@*1yUHQ+I*YKVRbph zI%$$N{y?3y_1x^9g4UAzL$;~kKveCwTw)Jeq1^a`eD2@?B5hX4@yo~foTmkN-cZ(C zpk`04>qLJ=G{<|sYN7($Cx~yYcgCr!qL>qT=G&cIq3&9SU#{OZG?!`uhl+&J(Bp*a zKWQpxA|l(8VS6eJr&~JQl+QCVN+L{G_ zc8oTJ7W_ulML&F81J%$^7MsBr;EQI(1FRw>U&7Lbr?xLL@1nuu`bW&}9>U~_t>K%K zJBDYVY(E$-%R)ruHVz5D4^9I4xo(zA#|AuuXXC8Nv4Ow)ZGuDU5zOve{-`VO6LJ{H z3mwiIcGB7XsHQ6`my{ZT$k&W|-x>Tz-c4=P#ZUgJ#2*S_*9=GGmD5aD(Hiu<5?^@n^#&pYgb#%=+(pDM!R&1A8vp%W zcoucDDy#3pcs8roMEljC(KjeCP|T<9J=`%Z11=2MpusgCOhN757` ztka5Nxj@TrL4!<3U4!ok!u<7DOHQWx@MG9H&$gEU~>6s#cnnsMA)~c!Tm=Nh+h3K z#Q68qy|+WRD$ygf26{ev$w=k9rgjzm3Qm>rCFf-HBGdJ)8=`*`UF%>!-C9 z?m%jZ)Rj{AFg&yMFVV-<4p1lH?yS?AgqHu_~`Jg)-qN$#BO*r^e2Bd6T@S( zOr;fm=hnkBKa$z0Rks(KMiX~lE7etZ^$R^6N})Yth#`*f4^vI<#&HlvFMzE96A;_oL~o`wrQ;2|zA<^)2zjDhzIqeLs~nhl(GC&OQvMpn`XI z*BLyg=`+y9dWb3BB@akU@?+yU7w`;;e0iA`n7_MT@bEP?8z;2tqg;7M*cy@c%!EWn za6w*>QTo$X8V(szUAqp6EGWQy*%wBE7c|@+jz!>m5F5@LWjM?cSWgwBnSdcpm-1hb z&#_|MkP!m)hlCG_?7-rXWUxW}8?ide{hiz2;S_cw%5h<)7U?6XOGJFNtKu*;{W{sj z@)_gnCh_ijzSD^h&5(4AtyHDZ(v8Q5z5LjLYLh|Li@^MQjK9u(scGZFvw!Y*wO8Tc zUp;v5__aqgJQ1GRINNOSrwt8z= zlPR(R%H)pSJWv1bU-(2McBmZ9>OB$k2olFL59}$<)-VL5I&nuoln2L9F1EEjiw(!} z@MM7HY0PgyD5$rzapZ%g(+NuFBo0Cwd)2^11dZnm$F{L>XMWux>ho__|G~Zv$GBtU zagkpM`aY3t#N!=;W4b@_{#7p@8b75Vr`m)0A2ZVxU41OEx`2yYEbYrf0FmDEBy(hB zV(;$}xzb#l4W%8t&6DE}(0H#pUaU9};^I+Dx5O4md7t0cD0>1?efdT=93Y|TOU5o` zCj;Q;9$&}M(zmGB^5!4-aR?2-=z%?k7|(KF$Yy;L9)|y3hmhpAxl}eU8Yr%ubK3~U zbnA5GlyLFEf_IplY~Gd1<>iIQJBQAk%dfz*Y>@QoS>#3JyJ~CO^_6f;ysT+q^^Fk8 zs%9|w;|}U9GonbHISbiVft+oZX!4F{TRk5pC6V2+0A>a8oqw`dIa-Xv4qkKGe1vd?He*0X03Y+R(?IQgvv0T{QIIZj9>%8Up1;-u6{) znqLmjZob*<^*F}c=ccQ#e)k;4H$u_#adCOnf@ZVNjXiBRgofE#eedTcp$cl6(ViRM zQMMVQ-1b5mJ_yI8F}=*(DT9dTg`1TsUqGK%=6l<`uh13$z55hU24vr2X*v6 zR!k^5f+lx+G3<$rM#W0MqvEV?prNWOx1V`YfLOu3lqA}R77o`HObmPBnOv8D?f8QE zxf0y_eO_|-!pZ?+$i0JX(0}D=>0_()u+Vo(fBJ*~8vb>gKj!WK$p^)Ct)_$t0lr#*FK{XTBciduu zCLY~={vsda`z=1z^-gnC4$239rfB%#fJk`u>&XpcH2XW)VS~$c7?(aC&7a>5M1^*f zN8f1c8iiiTw()8_OXKb3y4E!u3tsZw*W=}=v*Ov?!(y0!a!zM&@~vJv z+!^iqqRR1O!+QRB)?Gy_cRFk6uK&q=&xdbkL>=R0-&=~WBkH`k-tk9IQI=%QZ>2va zsM9wstOe-!NKx^D!;f^n5K;d3mB*w2K}4G4@Ej2vK$92!>kqu~K}1IOUK8h9)SD0z zf0kDUE${5G5n6eNV?Sg0z4C=4nhnWsb#2OnzA*J27CWE9qj;uwJi96pw&A_#z`WIyT^j%rEfT9wkN9Z3U0yL6A8aN zl*Iq;FOKEWvQYGfVU%gqeq2B3EYvyu`~?IKG?>})3RsW2! z^7heRXx!9#YrK3BYM6lK@LP_t3VKo0+%2KlO zBU(0f&=iWvMH6++vyYyWv3=)B+QPB|mY*;vr>pI5O2so5RhZy8cwq6%pGz*jHW1!v zbGKXGAIQxk>z{OO2l9DFk(N#>Dj4vzuJogz#feLmI*gCeM1`zQ;KTn@=PI`8{2OgV zda14@>W=AY#Gaw!{&TW){h#hhrm7kXqNR&rda63sXz@Xh_pb~KJY(&S>z&h&P#t;` zTjf59=C-k#n26cp*%Q3X7^_*(tk(OFne`2@D6;(cXQ3((nYw0E9_v6#+9N05Q&nh< zW9zd`3YLgs;5UUYVuk+9y;oIF_@g<7yWY3PQ_#=GEb&mzPCT;{&S6$40Q~xknTAc; z(cJ8=g6j_(Vcvi(lfkkG78C*lf_cmSyWe=GT9NQCLkH2Is@AsTGfrr}LHg@$z92-@ zP`WHQWPoSR2_imAGJ=%wwnrA7$AIRv$OLZ7MA&Ba+U)bg^mv0+FO&2(%! zx1Rg_Uwb`Z|G8$&4>;i1-OBn?2Nn)H`#YI@LRq)ppD&Jz{inQ!$)#pLU$>y-s#9X!M2bl_k?8?9fX2Vi-)^gQ(Hr&T_X$9R*-PSPV!`u@YNJs5Pv zC|C~ro;tE<(#?P8wGWf7F={kpJQfvmMmy4SIiOG}Q$xat0K>hZq1mKhrF&2|DQ zPt6u2OM&>-@-%}DCz>@mvEzAA4Vqn+9Ez#Bi3)a}NYw2*h-Z&%-ZN$rf(i>Yj=MNk zpt<3tQ&9$(o-zKO;36iTj>?9g7yim&hg60>zMD5)5%ul5bsMD<5%JmWdmEY7(XF>7 zx08nb>3pbf0nZ#;MC0*?jGQVjwTopyl;6dDSDY?y3Y#Yb-=j$4lU4_<&G3L-m42bBuD8i73@v3DcG~K<-%7rG4 z_i_e2E5!8GN5lXeWfe{AdwoY$ybuxBpMQOR>oGjbh11K~)}DxRwEMnse_#Vx4-JU!tON&dRImJ2^@vdPY@yj^2IU6CWNyEOYpS4hR{u6R=(;2j| zq4)RU$Mp4H`)+TAq|c&kjzL?|%F8&D^j$@0&OGq(>*6#-7Muc(&Dl66htj%R zhc=y20)Z3mz=Ld8To>UlmH3x|wO!=Q4ro*Yy zt%uR|geGXf>TA}8lNotfPyDavXz%U4R-3!3Pm4d{7*&ewlTWq4+{zO(w!7viDQJW6 zo+DV^9z&b%oA*{U{0hu!tEIfNpwqK)OmSk)%8qpSY`Xq;-6XhZ)I5R`1x9?HmElV!(k3NSzSfV=wvfweYNocn5~G^q(SY!8n-DX=b#6>>}UBz7>!ivcG`p0;hAfF=It?63?sc!8f#nk{p)!g z!S6^}Awrh0v&*&wZo1fnb#N?B?{x|od=_rCc+vr80s7Wd7 zC*Os}C2nmW;uMtEJH<;f-2h@;@o@R^PBfnKeSZ6*3tfAZ?HYShQwOiOK;>Cy(Lv)xXiUGxq9Wc2M4wZ&zhny0Y}B-c?crdUmSfQjJoOGupXhHg z$X9{6jD|OXXc7@GKDlRQ;s-5hqH#{?m_0F3l&KK(f(|c+#-xsf$$X{p(?Flrc?o}m zadJ|REHg$NDN(vYg_Z{=Pn38WjFq~isr|Mf{Kfuy`Et8iv|Xi zr;d$%g8F?g6I}bTx>m#ENpg?%Cs2=h*xbe(I=}uu=cB=$-9cTxAI%oHug|i<@Tl#R z`5kNDLY!9<1LIhF01h;-;?^obT#7kS;F<5juT;@AbxT zDHETo(j}HQpo%`_{FjWF{usq_65folLB>xXk1-!ZSa=(I$X?R~sQM#JxKFhFARL2z zWrL+yIS}?(_52>u05Ur-^Xmv}poT0z31n1**?Ue6r#@{$F^cEDRui%J_RdefVUPI* zO+r8GvPDB_zqcCy>*Y`~^_)ivh`=@r7Bf#rHqE82{&aRHJPd9Lb@C z%~RdIHxUAlfG8L*o_!jN*ON-S?A1KAfjZw#t$vsb)U;=Pru+mPvwZ;vQ>PG64Ygf< zjQ3-5q(|nnI4VLnkH162UpF56jb@6EzkAa0PaZDXd(#f4GO_P+g@qfAeVa+1Xi44D zfA_Y2NKPKg6({dOy~F7nY?l>bxPZ?+PEY`m_C?*}O~``T7u*TYf78`Fgz2aowdMD! zVAk*V?}T=h|GB=BJ(ZR$`I-LOp98i{Pf8*9Mzq+Lcnse*^sYix?ph>JcWy6tYPLb~ zLK6P5BIEyR562wNW0`&LD6H;(_=Z&*!|!C*dtWdlE785bGY=QGUWJY?pLV^Qk0*KP z=@m3WFt=GvJ`KpfV5U&;I#6!9k0!t#KX>(D6Y zFuSOoHjtTjIU%beAaCETJ1B_B6+vtAj+4Z07=GB^e&T%skb_0!kKWu1%|-WFu3Sn2 zqKU1SdMh6CFWcT|4da8wiXka|T?|LOCaRgVey$B>`FCxT!Ra8%n|0&4>%IfwhPTho zWn(n%wK0&nHXRzpTD(qGZim7<7dfwGmjWr~J!i=SLn!ut|75$&MKospeh=&96c8<3 zIM!xuVOA%q?en_hIF{W$;|7k-K=ylUv2{Zl-8lK=eCv%O+x@TkrW+sPATQZ=C;%%L3e-gTd%~-{R}YcqxB$;??39WH?(}7^7&`IW}vVs zRX(f_!7;{0+$inF>ibOwQ^^C=yF2BO5>JFL)fY2 zi#J{S1YRE{lW2SB9u<-hIYaZmLC!F!@}$}BT?2m~U!%B!r%i|V5OT;V@n z2ZVlskn76xXkk~wDFe9!XlSLV{M`1-Kon7C+x`g$^N~phdBU>j-!Ja*KrXo!NW#+7 zgk^aM%00zYqALhhJBtPEBYEI=)-}q<`xx&R+1Bdy{SWo@?I94Nb}KvuSxGkP)01S4(KrNKcj+3X#_*Zaw;N!?$hJ?_!y#BpRw9CXutsV&z_l(YP96Jud?^b zZRiaWz3(1bg;E}FIUXa8@ecR;Z<4wZa1K`9nzq*(?SjJSrR#A_7#^SSBkiYDo;u{O zpK zctr7|zQW=~DJ~PjveH;y3^h$Z{reAn)F;OFWX<^`5ZSM@eGKVGouc|XJ-=bR_Y*IR z-leUY0fA>W!!H+$-;I85=g(mKC*Kk3uAS<9tVZJ*1CnxKyhDirDu^BHeIcF?NnhjE z#eaDAujgU>iqGNRq)8x18WdCq2?HsTp*>xx3{6HqLqE}Repq~Tf6u-tVYGth8K;;9Q-4tlGcY&VSMbA_!_$%T41vJe*W?cgulVK{F12%&yx2=S7B!*3_5L3 zw63;7^9$FWr4$koz6ojH5yNm1g9CTov4(pAxmDzSj!+KvyBl{AIpsgky z6i|_7o2k@uI{gJG+@i|jay0)Cn%`7HoqJ=CigcYv`7h#8R}uGE*0-&Y`r+Qp%oJAN zea>2dLs}Z+qZmEhVRQF_08DP_{V++vaBuCoO{RK;Dwy+7KB<&TXYT{CX~z-b0`Xtt zOLzaQKpR(?l`bMPt;q{Xy2Ipo?Q#DXw}Em#LWgV}4FpflN@J!K$b8)OWOq#w%vbyR zi9c(1O|x@QuF2>a`>j$mDX7+67#$2`o|8oztUv$H@gVSe^~tFSWAhNK zkYlr+3Y3nEwM9Rx(I84gVh5dpWOHQZ^-(#P%lTXx@zVjve9mOzn$j6GI`@sUaNToQ z{&Yi6PV5FOwpv8qtzd@P%nBXxFpNK(WKnr$isB4}U-5k*crK{-**BN{%?uHQ;x^r8 zD?rrI8G$#~6`=0(b-Tn1R&@QGNix#sGGs!n;|J=JQ*NO}relfw7RsS4c_sE%e-xU5 zU*fgZFHoq-VQrH)>;CQ+?bfGKG29p7Z0h@=ne8yOV57ICObN{nOE^A#jn(~)dFm-1 zf2;$Bc7GN4)aUyH9}fP5fE7L+Ae*N3@)vcJP|~KY1=Nx#zoX>c-8F%8d07 z3%#S|SHP!>^)jiQn4gKX&xCKs;KhHpQ$w8bP1J)w-^`y|ZT}}P8O$9lIe5V+BbXSNhnw9y_{*ZbF*3=zo`UcdCY+_K^*9SFvS9zpb z%n()n)%o8%?I@|bebRU%5y-2TG+dc5JCL+mm&<%E9VVJr?|wXo)d7t0C48|R_=00# zk!ChL^ac@|CN2*O(D0S8=H*(sahMUNIB)gHmV3e+qgiaKOz=PZ`SosDLE$YC7`=b{ zC1wA%|240)e#~BY!D4sn_W$ku*g4{TTlHMs1vD8He`xyBRy47){z1}lj1RG;N>pcT zFAYZq9n-Or7cww?wmLLa>|WUaHGdy!lg#kH-iz_@w%@bQ&KIG}lmEYNZ|;_2)@9w( zDC?r@HRUuDl>0?TSG?oTKfia>Q?P7=T?uBlzxlL&N*-l1DxJD<{wt8(bk$CBkl_>e z2eV(p0yviBqc)DY7;iS|)d}_%mvR^kw;R-d{{Uql2#T*z&ZeIS7Wo~4ew+X2xUakt zk!MiHctE;Ou2*T;s0h8E4+nTg(&6R)*-l^catM$23n++gn(}&p)%ze`u4eH#lmV0* zBRAs5IACg4@P$U2D9qkAzsbAP2xh){s|)kd#kpzEXJquam5INBKJQ-z9ksPUJ=4wG z`y8w9L&kr5cjVQk|JlFs6O6U}dS0+%-LTla!SjF4CskfXtaLA3{YGeDrfje|u>S9s_ZvXIW_#*}b_q1Bx~i>o+=b@HH@tSw zs=_KkGOl7%GYl)W7;h|>g_=u^dL}>cu(-{pIKGZn-wzFmIS~c!E2Bm~#Vm3LcCIdI z+MZolX{764qFn!L2VOGPAC;rCG2=~t_3J+|qZeMHo*%rqhE^VQ-@*$d;^^3Vw&?g1vF-qBWf)s4 zUNwcz?kGiNts_7#yu~Z5w-Jqro-kTn*MSHR?cJw+7*O|EZEzMF?cBpLUB1+WAcpMOo3@%8D-Tk9s}N9zN8Z1HV;nX&oKL&bP_#J26?WBM_zZ zX()NU1e$pxRHs%-0cuCAQ{)?3yc6cKHi*mU{?q?L_r7t$Y{<`<=zop_Rd{xJz1>^X zi<3KaQksHen(_*~ENBgsqpwG!cF~=eh~Ta?n{y@-$oJJBS)5OR<#W<5S4{7qMt9>x zffE@28&!tHH(2r!X4UGc8FxEj)H&3t=l(V{(3N~?;WwtIk=|+~xqZfP^5n={OpEFn zFncB;_~Pmx9J83N&J%VTUIZ#T{uD&#umGVkAj~mA4fSbmu97PsfyRi@MKNAMMEWDD z>vxHUXF(G$nX}$&Cc^KBT#`@r7QvVeGaof_7Xhsv3mlHyd;x~c(I3EUtCHz5)2_?&Rovu*f zwrK^$)S*AQ=JyE+9Y7*Kae13345hCOK7CDa25K?mMOz;_zD#uW)w7VmKMXLrTQ2XS z*?Khn<@5JI4N=rGw^zG<8RNy6E$+MYl`a0S97a1olNg6Kl1%+d7slh)sgkF(e$Oei zR?4bO%4>l<{*!%N#jbR6wb~zZ^wGI=M0VV<&hXSKOl*(CvAz2S${D@^0Uez~4H3*?s#=albcV)Cz{cPYNY5vYGw zw!gph2F7KtyK-_pr+fe4nR~buLs8WK@q6NXf&n5p9y+*dCl_k&J9qoBa0$$=SbS+{ z-2q)`ufMn9x`2?NZ;{w}9yLu1aK`++k7fglzsB|O;#qJ7k+V6cfTG|=k#Sgr@)o}* z1A6*En%DcrG>Ah!=IuJ(dk7G4$oJ=-Y7P|4;B_?Qb^vM#ydrMfBLzQI__Vr>wV^nd zyJS=32WV`gzSw!x702qo*|8)ltXFcYJc=b{zBd7yImHXt-f$I_g>_`|&0WA2&}wH#LPw z$rp5#zE|NG8=Y1jtCc`sg2;%r2*&5V@{7fhjqN%#G+Y`9v0n!Sy;iSTxQk<=;7f$L z0-^omOOaPTzOW`tSh>yo8it~|72WfP(LhR17|G})tR!Vx^jbKfWvwRN=fYRfyi`TR zG0!1veS2*+L}ne!Dhvpp^U49A%KT4@=p=Z9k+B}hA!=j?W^c^z*NOXW6_1Jz@zn1WbcX(v%~u7#V)}CV=)|)aRYg=T#ipH- zEJr6_i?6?2&y~RJ!rm`eqx1ym`NJol6xS{$WAQ8!cWpv@sso;d`tFj8=y4$AYnt$1 z--AYcZ^sT%?MK9i_8xpz)bR;wQ2#O-`nAQ+RojtkLMb6X9)F zvG_9K=fyAfgr7L(Y`Z0?T{}>t+w}n(O*&iw5H?@3SJ;i^drvN>r^Q%?Bhur5zQ;Zo zKl^aNIi(<=iwpeo{F&^1{NiWIQ&bQxe#3Qd4YU{F&qT5(;91<_FS_|1#xZRnJpPlr z7Y4^z--P}71jJ{XN}q4Scr}U4YK29y;i#~9!;IO^2SD;S^4(gA<;!>7@Y{K&g@z}> zGi)*ywmS#l>-hbAJgX=org{8HJ#!ysH?_67ST$gJUnu%oB#Y&F4qwtco_g*OG+Ekw zxcY4)igN>eCNY) zEIYi|!(aU~E(L>&PXei$!qCfn2h7rMz|=L|{I@%_xB?=vS$)6!&=w7E({nxe0plI4 zGrQ)0p!OVET;oe#8Q+7tE{fy!Xn#a~H*QjU`tH-(pWeUbJ+Fo3&61g-IUQU7>a`!P z?Ho7pM1;FqqnGyypxR%GTkbL(LLFX$O|t49m|Qb3Z`!~l4rKZ3Gs#~uKMI8re{v|% z7sv1<_QrQLx;Vl=zjxvB789m7S1>u+7_}{bGW>t8gT8z7YaZ$*&=pM{-=`%F5BBEH zsERxSBKzNUo^Emc@8@w0 zSv={@qdIi$QU}<~gS0NfpNAJWA7r}RQ zWMRP-{4W^HpIsSdX8;2ALNspp9U4s#x4m3P)9+Dj;gYLtK==Qgr?I5! zM15#0OuHUe)BPO*)A+iu)?WrN>@8=vJ;v?yDG@WjK$XO3%Hs7m8y`moz)7KqDbz;*@A{+ftg4V!G`NIO(^Q{hhx9nAh z+QOm>OC5B)H)xtqjj1W++dunN_&B>O@tG0oIu(00)NA0M_YH45FDYlJfEKs-{-_mJ zM16@*nOg2&LS*~9=V!?@I~;w{T##qlycdXdugE(`B~d^ACmViw158NV9+2p}2BlF3 zJ09L}PY>-2cZ@ml;IBcC)#xqou= zqYp(e`Cg#^S^EZ9z4QJ0fi+o}^?kRcbJYvA)O1!oNXPtL$;!fa3fzuk=Q|67$+9R7 zCkgW}%lqqQc%ec5!ns#H+ku?0%%yn%tE)?#yqdOos{oqg85vm;5CFoNH><)PsxVG@ zx9(mD7fc@J8m%v~Lw`&|=OT0J}^FMpk^z{XQq4&R2w};i3hVA^TSG27X z@O7wlxTpEXz6UVcZTwv@+7>$0Rc{^GzXKN8F6=%f$b#A~l=;1ipFsTvNh8gBZlJ!j z{Cek?Y(V(7?~m0h7BqMK^j0UBqfV1rJ^|)A97vuE*SQ z&yKJ5fW`Hr_g!pnqxu>Ro(q1oJQpC|epq_8XCIpHjc;nqy$^LSs*bpP!Rj`VKN&Zw zu+hmQ-Fv1P#MVubXm~V4Ra0cv?A!!o?FSoB8_p`z|QH)VaQh(#~ka@<{o>n1yQ1`h*QY zt=5)aygv-{o4Ay_&u&6**NU8;4~n7@1(g$_UT4wJ@S)Fg?2lk}LtasHK1R{3tSox= z0Fr~$ceJu%c($qHJiTk)^mf5`xNG3eUA46TTO2v!7|V~M#^zc%m7R3_V|4pfR|CIz zQ^_8nSXj<>*^a=VU(i*RdiTduelhGh4Pi!We!lDXo7K=lG47&&k}W18rIdWE>A)o`Ws<0fF*cKo0>kk@n- z;=&zZ%CtAO?(_#B4i{75OdF2b*Ieh91s(qp4)%Z}F^?8`5CJ z^3C^>h!@DjSixAK7Rw_e95wGg#=%3^Pv5}k0~*S-_rYwHyh-oTRj5Qy+NCD}#}pi7 zLO8Jro9}N$uZ39Z&|tXxkqeX85UG}TG37J1jtRD0cl}Oq!}J-Z&i-wl;L`cpyQpan z%HQ>*e+{+dQ`5^wEF+>c-=o@V2b&%n7kzIcKu{U`Hh>RN4&B8+A5CC zk`$IrsC9J)gmz-MS1O60avIG;#6uPfCeD+nkLCEbtGcDArCa`iDz^>YxRa&qeU2Iu zAhyas@Mb>73)s8=+Gv0}76+oL)$c1;qv_Ky_V)3K(klW`ZV`0fxhnae^V;{Z+A-5& z>;IAuSWLD$)3WrIZa%tQj-)X^!uV>5sl>+9P2#A(=nvzMIU6WlJ>bw-_zD|up@^-q zEhf-*dmCk;h*p0A4HU~4e$o|%Ua`5=KEf85nk@aa%6bQl_MH=GaO0xu&nyFRVpu#G zNWb{p76p$2(bEEVQmYX4CKsN1~ zs)^2TRQvMs8p{kB=8HoNKWft35i{|nEE2W(bidDj(!OJovK_|H#{4E5(CP@o(@bgY z6P5s?>V|m_Z!*L9I4_?Wxy9w0HW{Q!x07*c$zc?Cv8s<}7(FDB?GFu%- z>F~2av9^6)kdM_@T^v2`IFLifk45*Lbtm&>EKj}%Pq&-{4|F8xuJ$r?F3^cD{||lh zd&Y~8p6RGY_C8r>K^4U&iEsLoeHQgkt=qj-hz{q1G6mlKW^WZl!vZyuRZp&?(W1Hg zU2Dlwk>fDlwFo20zT4(1ze2M4mq2P^J}Pd=8uFIHcuo7`6qAE? z{>CwTa`@dtHu`6+1jwD!%7Z2!bT(Y0S}Ys4X5M7M7M#liOd=7V%RsxX((CKp=p8uCB( z6wp(?CP%a?@d@s%?*(R8?IAQOz%%X{pGvaq)Emun_qMPWl%Om;phA|wXS z?*58~cN2USnlfN=CpYPm>j_jJ8yF@be)^yNUwV=;{5kav-MPl_(Bq!^Zw#+abpYbo zVI~;(Gpd!n&JC8@@$b59mC?}YM`1Bf9saqFh7W&xf3pmadW*POB~)nRk7L$&yJy4L zU;Y28j-`rBC5}bLSKr9E49yi)T{`v)^P?n+?b^|JtO;`PAs=_`(}LN|jwPL_(=fq( z#+P~BSw!XwJ8&q#0#*XsgA^Jte?>7*O<7z5hV#j2H40db7%E zUO)1I4VIj3m|mD5>>(I>Uk|3$pHaMoXmz}x)9vHISNls)^zX;y<8qdWZ0ZDTK23< zNdRj9ep5n61jAp?T(a<;jgX@KKN7k2klFGjOm^?ukdN>t^NBI+(ie`ZNj@k9{okA$k*_Y=uMDrix820h;2|lLb>|xD*O$0Y4 z0tAl@MGlD-(D>0f+R+W;=V6svpsL*Z48*HVZcinzqU30OqnugnT+K|GrM%wy4z+yD z8+hbv3?xe3ZED4lpN7)YfZ1{}ke3jwDaOQ2{@SiJmjBC3#J(8#&Q3?&tMUd zAE;TXsQSSG=hJW8A9xL9uy3K05w z&iN36VTN<-W4!^azSmq~uYD=e5y*`C_9pv?Kw8TVO(W#v807CvQ*L~Q(zr`GX+C_= zEOhizx*_KOo)@(eseDYc%VES=rnTi=GHS`r(ipUT2(!WCE{Pr3{ykg!(7nMR70Oz0 z56`hmpg8plJTKV1(L|T@PW$NdwEL!RjUBkpNvqonD@5DIz0LpReZcfi&&qFFWeGq$ z|7&9bcO`TTHk>Oy;(*d?w(ZN>CxQkVMA!H$|DN}8xB*<02x^Sm5ttny1GAMJ2hCzi zP+!26uMw<&^*Z}6>R-xf_#@O7nz+V&wGp-22h@$tV?6S+pR zc6Z;z^5d$UFWtTxh4K9POXw7#(~n?mFm81FstfX}u{Y56qpxSRc>RdrUi1z1k0^3^ z8Ztw*RQ>p5EQ!``(T(J-;SGQFr8jMzi^JIaD9)2Ck>=m9_ZO~lDIYkE2xBL{h!x^c zzf?hvhTLEMUXb&Mk=-I1_)6LnzJk?P={sw+es?P!jtf?v|M84_iQ&*kh!Zo{e?5iK z4-v_3{kMU_H}ie>6%2QuES9bo$NL}qA4mK0@6@v6l(7?hd&e119JE^BL|($ zP`t-`-^MiBxr0VDOQfG$VEE$ko>%+Wmu>^~`tuzV6^v-YoEmGfqlB(sNnh@1?=Yi_ z)BanIzidkN84?ZQ2Yi?=>uxQmG!M<-5b(&91Du*UZM zN~{s(-{axEek+GkJ{fN|e~sZrDX|ZzN6vYSm|vk=JhqBQ&It_~&)7>H zr^|}~UbZ$TUS*`CsP9H3cl`_@fayI}A}hTl+DJ`{RveWfonj$`V1eRrXc zem!@X?09u@bUhFsh}{xtq3QR~o7t{Z>Ou39qiPqt*1Rdro<`jqcH?aQboLlbEWEBd z^n?+5axQ)eQTIXp;!B%M{V~4#_7&zuM(-XVd=iR3EhL5-9ePtWf78XiaBR|@GPV!s z+Q+dT)7G4~!Opwlt^8*PDiR@D{?a?*?o_CHr0S~G(Ff!a&Qt1gW$a25kjU3}`1YO9Z(C%cC`H31BdQ)UwjU3K1p~^_ z@)t)?t$wD5y9SmI(q|gZaaZ9x$`|VY@jy-prp;J-uZosH=v9e>EnD(otm2nD>nkkp ziMqLk>rxxdpM<8$s*imJ?EJ6ja~)l`V2yfy?-jp076kZZmACKevAO{jUJH}i-ZXt5 zwVX^EvXvpzWuCU!x2e`j4VDC!Vw)rnPkaA%CM>UI<%{=}Y0N<=MR$Z(zx&y1wH( zC;I1>%Wm)WpxHU#e`dItW$-nSvjaF3-qP|YP`#0=S->R^7)Vn3b~+o&gIt*{&CC|Y z{LX{H_4&8=d&1iKj{--=4RI`AIT_eEFuRH3&*F~9^lRGh<#JC|OqS{VAW*7crM2K% z19KrY!~#`Y=-;@uMq*`x)`{Y>3&Ylk+?Xy&`iSAe2Q{aZwbSGv?u+EH(r$j3Z9Qou zycx4IN1wO7vz`%$kml%FQQg_X73^K7nffm*vM%4{nwp9J;sHD_wC z`-0L?qz`EnS#z+fQDg#VfrwlCaC z3p`(83#C%*pXQ^iQRHQL6SJ3e{#82p%;f%%`t~y>CsebNxRf4wAUpjW)*GyYK_#{q zERh5t9-A&!y!#Z9*SS&zxr=cu*VaEV;wV9sxweJ;=SDykwdPXVxdG;t)ALQEX}Cw2 z<&m^W4W`vgLuBuZj5cA{@Ju?E!kzw+KRz4p>(MAVV&hGsUWSfNPfD?P@ z=8^JAm*e1A7OFV;`I};w4N%Hh%Xyn<>jM2DDxRAMZ|M7~cywPb#=A#}{HlC$_#h@% zUypkRwGH7I^wRIQ`#*=a;@~l+s|-M8o={qGS9t zT`&Yv9@n=2hqM2Vr!s#3z;U~2L=u&RN)k#Esq07??M=HhNP{*iDKbh$(Ll&nQITxd zF|+sHj=j&e`StzdckcJ&eSSUOpYQKa$GPuw->-4KuJOE{&lkM;Nb!}kufc!nFA#p| zy9jluo+Uu-p0yxyjPTP_;$)6V579~dMVsR1|IvZ$*^Cw^hj3^=oDc1@Ga-C`$t~X9 z9DnAoU7RJ>xRMiRLAH(ex_MIEaSZJTV^jUx?ZB|muT49&pTu`%>c-;*@1Q>?eL!jT zZuld@Joj?bBut3#&N{n|$g48WzYOVJ^Brh)K~}re2%IL(`;PLG%fvWSOpJw|>8yq^ z$s;db_GrVP;z@cfD-uQ?_KeW=m4Ll>*(qBB-v#ZXUT#6f|IAatci#)rSuZ(#eORtF zGikc;ID~x|&A8;Y588$9$_iU>>f+e3!uDpmFi~HKljDt0+ZhGypzo)e%{lp6lF*7C zvxz;Nco!23nj0ok<^gm2zPr)<1b&_sncaG-{BJx^BBCF&=>MKi>lCNm;`8P3z~hes z@raW%H%UB~>Q>L*zZ#eoG_8_(T=fHZ{!#t(7xrsi~i>hP!RiBr$WpY+X^M}C0`KpQ*kW+Y}?tN zFm9yR{P71Llm*nOR$bS|U*5l0ZvXikDN8pW-d{-6(G&dUzKZ?z4chZw=pS1y18ilz zYUzL+Y^^b{mHTCn)Gy5^vyIA$`R~cSde~V3hxaE67KYM+@k*YNs>rE_k$C(n9JR;I zVdNv0=*#Se5bfihUwDW-Xl}v#(?3rCCJ98m&v|tqA83!?U6EPE;bDdm36G#pdPF~R zWp=Xnbw5JSVydRrkh!E0+Q#{2Uozo}Yq{($bIyh+WA!xbG&K%uF0R3m9qV_9gmZWZ zustyARP5YZ=x2q)yEZaZz1*Q|ea8xzS6fV0_Oy|Cq#x{R$d86Pr`f*iO8`q393M{2 z?1rJ}+jH(UZYS=OmR+WCyA`Pg2dst}ME!4Qw$$d!ElN1JIqZ_Wg&(jN&Ab1Q|1NsH zw>$Lox;0QF%bZsjm|*+9NNXujnM^N~%6M7#XbufXECs!^|nuxxH%;nv^6-({O0wny#P<#~*iGZBJ!)Cv9p>adt^(M2_^qm~IfL{}n1YV8hd;YE3?1~6*w2~JmkiQGDR!zUr_h~owdQ`7Bn_8i!D_A-8ubQ*r%~zgWgWD?h`O0?&w{Q(CUw@*e8j4b!K-I4VPD(S%j5e+fOfWHog=>^_*mIW z&)={UX(hta@oG9aA&jf%&)JKeYJwvTl7z0*n9kY84T7229bFk%cJU;H7d*zc+TYj_ zv#oH+5Dl0Ud-ZeXUd54R1&2y+7hz}flYJ#-CP;p_qIp-;QEZZJ6@JGyK<1Zs@!yV) z;=tWQY}p^@q0`r0=jJ7GaSS5dq@J;O!l*XDB?JC#6|c-Cw8U$wKg^s>=cbP z%#(Qt6OZg7!)wg2OL)A<@`(?TVjrDg#=CJ2(4?yCU6>$9+N13v8oiJ>e-PcwHFB%;)Yq z%f?BED$Dt!i;#X{N5H$By+BV2GupnH=)-9JrfmP|yFB*h&u{k~B6vC(GJYDKA?L87 z?)>JiN7}%A#%~ZgKnK<)i=G)VC3qgyD-FQ0pEy`kbzIT#KjMIPcPS0=+Zc6mU2AfN zE)J`8>OB?}#%xKYafMThv2SH?%FX}@2rmm?mb>6Nc4hz3Om{nlJ&%oFUu?8NMs{A? zfNcr1hS?A2yp+NeZ|8hbM-^nG%^Y1KPxK><)KUYt>lKoD6nmmBFq)xxs#<<`GzSlY z6qEM`smaF3ez3zK{Ei>i9ju6*o*w`;#w^{CscfLOm;Zj;;D@DMky)|D?Lz8gUMr0QOi3PW z6rUUFc(DkgDSK1wO1q%8c*U_ zjy*=5UO;@V?tee@tQAbIf371M#;x0TP&oT>voIH)2U?}#>+O}f{9?%LjW(V0{r}vb z|9`gMBY5%GmKKiR43trl_1Hi;eyy)icyQSo>z8ejKP07rT@ZiE=$kzJDcn=DWhn)j z5es*%_`#{8V#ujk_p25~BcuLpV9Zw&cy20Itg?fXHzEn1O^b7!G{UyVqQw;^OR=@* z<1WdglE~IKc{e?+4I=`VW}Pyoz`$K~E3X}m*l_Xuf%97#Nd4h7ztj`3|9zld=lEsfL2}mz`y(%nQw=( z+1I)mjOnhplEP@i?q&5RuM>%Qi)}icG-^cfQn1xDyIVg`aj%oGr=?L*e7KVlz45pX z53=V7{NB+(!q%4MPbSCHhrmmL0A%>F$m{Asj0QmPB5?@s3;ZRPu^Q%?w8uoFA&P?D@&q za<1ndjE$9lvxXD5VDib#`zLl2d41a0mpijVN4Tz||7iUA;}B}3=SN5K3m2RGDWXofuGy42kwaq-FS_47q;>keD5Ul_~&?vku^wsp?CAoMvo%9{JZL$ z7n#I-zGVA{15~az}8^`hw3pK`8m!$Hw~1J)bXLUr%Oo#J)|Fs85z+dn`2>)$`zMs4jv`d*Lp zh;{FfacAebxz@``0(*WhaKCGdy(cWo^ZYkq9m_GkTyhnTtiJ5PRQrx4nj^P==zM~V zMN1hW9 zEtU^<^L-Y4iRq>y5$$#z*sYpYQ%TA|M&9~MRfqGThO$2S6vdv*zj7e(V%{k9?0dmS zHFSot#cMy%R3fpv{pD`OwTf7-bU@qs^d!<=HMTWKKY@x|6{ow^GDx=@0KdE*V3fH( zwO)4sd#~^a{bXH2pJ(lQIS~(${_}vliQF2TQ4ir;zj_e6exJ@0ep7<4IG`W@#wbv#EqZnKj; zx*ix|(|Ye!AKWG&HnHVUV*`-+yL*JvMKrLx?CZ)YjY7<_!l_Z0v*2U4bMs641RRrg zj#wnY>1Tjqqus7e)*3h{^giK^JHeaKVZbsL+N^|SYxg{Ro}-2|^Ng=c{JC}Be=MvN znDX&OMz2nrchEf?G0&61TqoUmPp0q#mW~OdL z{;_@k;D+lm!N^>HhLw2yCDNVm&-u*g$JV)_1`c-vfPQ6}s={q9US6or`61?hAq1v< z4^w(y6FhyR=Y@6MA}fHd+U#w=NEc{+m20+dJp=Hs!=&@TEcNp7V!)RjME)|T7ibjhDw zJ#`^u8ZS$1i#`wZG2NU!t>I9n%xLY(n!=Hni$X6t5c`(dcX~}qVHu9Qyw+IbCO{Gp zbko75&U`OT`)U0)78!>I)$J^~cxV239v8hdd6sJ5 zN1nY<`f%)7b*wpZQ%>~O5@bY9JZd=ThIHTe5{I92^J&Z|*i4-_K90Xl{Uoqy7t)hYjJog2!tRu{ z+b8FV;phASn|R0E4rJN^+eW3R^=JgLU`fV7EkW#Xd%xm_4VT{yQ;u!?`=cu`eS)5O zm)e54nX<+oFZn}P>e)nHc_}jAr+OP1V>1{^J}dZveG(h@_30;iiz6ejqWP8T73_Lc zwf^JXV6xECGsh`kICUYUjIB>N_P_ypLbs|n>E0yIQn{Auw(>k#V2#1S`M=^Yl=4{rf1FXcRVCQAM&SA50CM-HON-z z8=Th>jstlv#ksdQ{5{yTap1Om^C8alZH`pFCC%LHsVjk(|ClMECpmg-=+9EEM3O+_ zc_~YRj~@G+u4>yx=fF(mlzsiL`|wlk=dKgOfy8^23`Kqw&Ve4a=;|x;NYKA*wZSD} zf)A6Cc&~5!lG{MPbUN)FEwncpjP?}+3)vLOElw{2Sp9pQl&=NlwewQ)V~r2B+a z_0qe<_cQ`hFRuD)$J06;?N#%k-uqJC$JKgBalE9V(pZWk@>@(+hKeG+>!j)_#$zn6 zQoh-#z{z7{YI9O$))D$Y?G|0+dT8A8M(kc}VdQ1siqkrCFSAd|08>Nk*rU=2q^nat zgw+QkD|z%(nmP&Fnd%a=R}4U@)yndlzIxa=v-J3VQ6ioh5aRjVW~+j=yC(QI3vv1t zNP;U=%}dS_I7^lTFdqb4U|W@}_y&(7m~m5JX7z(Md^yt8EFhJHOw-L-CG{^bXK~d_ z@iGCJda~A2l5Gggy(=z~NSA9G{WaBBI8!j-xJ_2{r z*`p~h;3rAqmoj22KjG*=;?RM1_fzJX$e81$Ik3kQ#zcbEEdv&jc(k%V88k~mRt$fG zoXdP{|LuBj>ta!ei*pkCeC-0Zuf6agWkVd&MclI0q=%q%-KUzJ@>viTbGf7M>Ph1H zmftUSYxxhHcG9Q%Ypxn(p7~Du*IWb`&o7m9cJ(0%ZlAj@;eaM|d3|!~9VhY=g1cWi zt^JY&ebIbEB`X zJV|JA&+rd*Z)jch+5TpGFRS2 z>b>RR_&-Qr6{RyCF8bFVF&II+2z=Iq%E&fwm#sZ~*!}Bs(O}O2Po+S<~RpV5IX- zWbPCvc%z43MciI9kHZInQzyelySM0JXOM%ge~mVF7&<>Ik~&Z3;fwaWwZjH!QG1u| zeUgcz)T=hfeT$K{?!&jQPL7y*>zH#;TNi28RvYcTMM@Cjb8U}u(hKdkrCvuE6q8rs zUl#}d`L_%nF@>Oah2PNLx9UKhoI7{I^E-5(-Z*FH*H)xl)=QUVMpM1+UJY+GX*Zduw}^t)GUGF zgL@}dTs(#?bcX0;5))Ywa@!1w2D$CUkJ_+velj>5;u9CG`w-f0%fI#WAo`SJ<0YF1 zHow8?3k46>y(Rki``q@2dL;EDMWT-n)Z~!)+~C4TpQmJ@)qMUg<%CUF zvH}-DL+wth^a?eY4&3T1=l%{y1s+Q7*+=vvjMz(Oik*oA#(Fo>9bZm<02qa*FG{s- zhn5R#b4{j?kOY{rxAa+&ShjTX!ivZ~U>w(NJ$CRH4j!uy^gdQfo~6E2;@ScN7ZRK~ zS)Rq41J^Q6R15x;gW_UgubU{W-PbS3k0$tZnCZ_LnkRIC zUg0q6;jd48e(hyv+S9c_b6T2H<|zej-xC||4?O+nxs4gVPdmSG&Mz#ATII1`F%)R~ zKW(v09LC=0+Mw$jE+a!>^Jn2^u6h|U4j+EeduZ19r)$^PZs!@oA9;Al2yb)iFSf~7 znt2wMVsA%I#p^N!>`)#SS*lt~;tlz}r(C8Rno~SyAG>abT{5d?$2IpN^SqSu7Of2E z4$^WwYDnNu>FNg#t&Qf=xBs`FQg1qzUiy~A=P2v;aXnXm4Sr8dotjDgf%Usj3_sk^ zfo*SlT%_~$ajY*V?UUyaGP9^6?!vDy+VyzJqpaD$E_}YG>97NotD2|F!3VraJM4Tb zk_Q>Vt;Hr`omgP9wqlFgexTbe5IYv7h<*Iz@Xsp$;ZGa0=PDI$#eoyXcCwo_p!D6G zq|v#53q$j@IlA*V`a$0Y???W#P6BO*lP0_B zB);6O;n$UrfYY{jw_2%h0lQh5jpCkC$cSh(Fqk@N@*`x_qFsHEe*o;-2I!e`n~)`Jwrl6b0X{kH3@J6@wX0X|8>YAzw!1sVl8Yx2+hla_-qVx^_ENcTEwA*EskQms_xOM6 zI}mYP%c~RfJNF^Waw2#7A;sV4GqKZp$F=uwI*|U> z;;~shhtCSTR#fub=zRcPZa15p4mu&Vp}#9gH5H~tbsHZo>xFJRyR1gqiGTe*nJ-lJ zK&sFsuFrXDIP=`Ifb>Gb+IU^I)80EAGh7!{IyM{Vm1z&V9()S|3C2b@^8{aNL;)e z*q^J@l@-Im`(sb~rU9!=b?9r6iOAC7@U&ug?wP#A*%3J5#Vi!k5ySZV%C9yLZO4N6 zr|;Dxh`y9Yh4_+sF;0CS*^i4I+4VM%xvi>Uyp)4ef|1!d<_q?;LfQHIZI4|!dK;K) z-ddy;N$AvMv@^#|<&o>TM~qaiDwob+r|0ne4;D(m;OEg@a8VIU*DZdnAE|~_!cJ8= zhI5eR`a|*d%~YUUo(a<~dWcl@>WCh#7bN~Ql#5312^@9r`tV8ikw%<|NtSyU)sDSi z^vKubRB&ExPKVUhRQy|N$ja4<7TW!s_W&@0*#Dj89S0}S<*9<~BFIVWT0 zKD5Fp{lJHU(M6E5LO?Zq*LGm!>08i0MPpZ0_4%61ACaAKv;T_qBWyI0c^l~e5=X8I zRy&Swg63TTEiro)a9qgmXrO}GKi3&^!hCxTozOX$tXwg|lh5fZz{;FCmGlD}hX>&DHf+XzIk|R|8t)pMe#Yg>uozSJY7^SpR)Pq_u7~z zqrxqXk-qkN#J8&fNNI1otg^s_*dIokVZE(I*s|^8w5nPTRJCpUexqqUHoC2EIUBMF z>Cr6_QF3C${R-F0Z9K69JE8_QWTqf8m(fU}2hxDN_-NiuN1~qF?i;n&WQv=&g~|VM z-%7<1bl&q#Cqrc;Ho7QHsC=LSb<+t=AzLf1=dZi7*pnsy2>RCa_0Mza!-4A)vix&6 zd`~#^;O&#;OA7Jl>GJcLWs9K8Ve82e4PxKr&ARPlojgFqSv}vLkRR>DUIV|)?{9h` z<>S|sKjr?M|I@B_nvJk@q4JEfxd+=7`k(QBcYp9t-g=}6rjW;9{zCTq!=cCL*+OBy zg4}??s=q;k#$fZRi&%Ne?$ZtZTmPP)u9Fp)JpWmL#i1)nixgaxAS}Zwe`U*F?= zpTQ81Ut47gv~f>KZTD5!W*S(uMEsxp*HBm|uB=#4479zm@(0Y0l6ezeRu$X-jqCO* z8EWbeLB}jxk?b~8L)hXiiJ)4<1WwEIY zU%=;DB&kELd$DMrcGf~ z=vC_bm_h8T-{iZt@hg_hE#9-M{V4`JeX5!Tx!6(QTVBb>IoFYHiED-ze1ulFJLe-l z?0{k3xgM+MbLqEY_onJx{!4;Tw^WWjv#}mBUrrv+4OoEHi;uPW3S7mBV(oQuq7iM(Wwai(37PB!Kgb%z)_UBIS0PQFqH+v=#LpVE6JT#&gbmfv|mi$8;9@;>hb|Uqogz zNqn?vry8LcvOtSqw7Po&GE`+A+X+9!#-9brrB`>u$Y%G$xBQZjBF~g~u<{`=cDda= zrWlX$Lrbq@Bsd_oy4prOv=BSWDqQcZ_rL(9#(fJzSG<5i|x;_ z>!9-vvOxA8DPH|j(X(Eg`i z1l=F6I#Zt$yt&u(ldkh6sl(7~4fjKj+AvQZR{prY9=fI$Fsj}VeVG#%-rU+=eHy8E zpY3{77zK@KVryx^|GVFbX3M89_gfG9OEpqSZCrjT(l0(;c)IUC^safd*WBVd4*sq? zrFDqYr-p2C3j0hK7rqnP-1nG$9{!(x9OwHLDi;a@y)&SFuf_68b~aM@bn}PidPB_p z=1F~CPnZxe7>dz43LRr>o8N!t($9sInI2`Iw}jpg+tA6!GSpsTamKL_`E{g7*wro zs_rJLD&LgQ; zPH*z)LN`z5Bhn^+tL*06%Qe3B zH$B#zKb8e37U^^HAQRY!jYfa+a-DZTH~4nV+m`6-HWw0qZC%M-ziyY3F>!Rd3&~lf zbH?2X-0P^jrj@9)0 zX%GD2X^|ig9>;?8q2if4jWGR1v*!LnB7fHFsVTHW{5bZ@mqx7V<@7a>_zh~+i^No* z>B8pvd=<`n0hM{%5|$9S-iit(osa|3NH<X(1Dic$Qn5O8U9H*vKiFP=cdY4_62!P=^-gth^~H1T_okmx z*WZ&W@Rt^S4d2s$+7B}CV6|tR0!JqhsZD3s*)H1zU*KKBxDyv2BQkH#k=-ZDIcJe= zWe};}$7OHGf|o`5Vhjo0ZR)h}p933?K(EBS*AmvJkPTa(_27j=uO8u;Q2Vye%T{y!zwhC(WgB=`<9FMhl$K~E ztRJpAM3EZ6uC@bfudVmPPv38iCYGdQ=v(?~=q-3n=2@*L5dPH=+C?TU+#MURVzBDAwR1a+ zCzRgWQ=5!cpKi>g6~4mu=m6etx*z`6_;HV?|CY}W`Z#t2`dEm%Nj~`2L<+rdU*8kKtIf8FzbT;ABivU&R*A&d)OA}T-`Hcj_r=?>MJUU zK8CzgFAKFzJ^*chyjRx57ZT6HpOvob3$SBoCi9W0IKirnczfRN zuWpY=VW4G`@-Y!LXw$6~K26F&=JwQ#qkHmk#KwVNy}BDpl4^&;#Z@qG#ReVTI3A4h zwRTq+A^POT^7}-yo)Wr)t>$mEHAT;%tGc1KVgL)yezx;&hHLq+P^;Mwb=+KEK6)B+IebQii!V~_uKG|KQK1LRN@}b%}F9y0# zUmUnP*AnwzN-te5P1Iv+N9gR$j<1ncF)=@6TMMv@O5e`PSHZES%V5MNc$9+8+7`aJ zv=Qs`9i3Y17T}QL^@ErE#F2hF_+XUOJg7e%!!V0%hT#jxpPm{M$Du_hzPqv)AluRR z)#c=AoIG;X{Fko}gBeH-kEArKvSp>9S;-$a;Okm3irQ5EV zCD>ciWM;ih2Rmh=vO{dGP+ll^bqOQzhq2#lc6|^-H_z>4*N>K?aO<>#@e|ye3k{P zfkr*^bBO0VG8XjR=l#Tqf5F9++;2d$A@eVuow_oV4UF^3PZf>V<7l9WP?U2mUOfIF zE5y|SM{AF7GRzrjubLAT&NckPespp8}DuInO#4Q&>ids_atzgVSkY}ZdcF5D1ge7RGq=Aj8p zYpQap&rcEuX&HcFt&FWro9{3c3sS_Mcyzu;v%glxma~&JBf@`>~6R ze!4@=Gr`Zk8iiO&bCey~=nUf#M@WL_?U1#HHU9i{JFsLlC@hK3WL_<+6VF9=<0$_I zVTE)BGVgl3h>lhDoZp*C_};;~GY{zpFYi;@)I<_keeZzETv4PajPQRLRf8$5WeaK! zzlU~9tKe@j%dlOTY-Z^h5bbA9SS3Z@G4GE#WVnd$jprIH%qM)SQ6$^D7#m zSNp;0WdnnlzvlG54&{$X7k(Nzu!n;)fv)-Gk|UDKAm#F=H*3FC0_(uS9dUDuVLZLy zg^&KxQ6+W z&;f5OA2}ZvP2kFzlykm%WK&=?XWMw);EWH9EPBx3%9Dv@#*+7B=6MkNpXa#ETECSL z$A6DLT+jhb&2p#vDlIT5weX{o$a{F6aA$Ou0)ek>M$hw}Zwq0dIw-=fkV13iMDl;;sY(2>dDo;;4w>)H#IxMnPl1^!c|O|i zDh_&+Woq&tqcbB>W{VG}4*@6!U%N;&FJM-5?n#STS=92Ho*SsY?Vf{Jv3C z=4?bt%_biAVLxaljg5}CT?E$IIj5bDWMapg+d)|s%YgnWCQhs50|c$$a}eH2oHLEV zPj?)?@CcgMP==J|Pea))^AF2((~+_){8RkEFxP(SdSn&Vl`srH&I+AtJ$?-aNSVp;I&BU%y!2y%i=` zO!s7-pU1U6vNOsaA6P*H)Y5w(WmAP6fqPXYJ@@06lj1b3-<*074s5u>vu(2uj5lhv z#n${gA4ty5{GL4I0Uej5?ae;N1GVvtmC15~$GGE8XX5M#Ug#bxwA^6B#vdL5)^7(t z0_#nSXq_d&+rrocqX|E3on1kqcTS=kb?gfqUfZ|H(rDGYaR|(0B=) zP+|dm zp1bKxfc=X*1&9c;vd7i0U%H%s?M zad@_IX!-6d408bN^PA_kSQGCHvl&tK4|}E=fvDd-+LlS)6atH=pQ77~&6;e0$XyDV+)?zlw&wnR26?g_Mo2I+G% zH~RJiYl2UC$?m(*JxAzuO|CVGFSIpJMCmjB*t)#s&DdAnk(9i4%%4n7||mIVTBPVs1?#R2TlUHUFftpzA1@fE299R4zBw~g~U zT%v)4UO&xMln9+CmdU=N_W4D?O8rx2J4+lp%{-SmuBJm<=)+@$fjKa8q=IR>m+0ef zzk{tm>tdn(%T-UYAIo6+hSZ0RCG9XwZn{Y>`|HnYhnWqH5d0+FdxJii9>{{$ed`J? zt4u+$HrY8oe3F=-nGdn{`Pu(mm-I(Z(vQ;*zy$NP+S0QZfPVP0PB@iYpN&ZyqUHZM z35+P=n$rhEvA+O4K*GP&ta0Z#PQCE&`_=P(2)+r_D!cDD^DiRBC$ic<=smRdNeU&p zeE=J?d0`Ov0b|#m3(1)`7khs{&3-HtinJMu@d?c-=+G=(C0#|>@o5*2l}8?P>@Rk9 zRF9m{NrncE=ghmeiGF;#XZrkEzqs(lSk>V*<#RFyJC}F_+KSSEzHpY*t!3M>A(Qy$3)BVVjeT#vz^QW;hk3Njb+0Aw6oXtHB7pK(wD(7PR z1|yn(oei;%H=RlJGyF*6Ejs!k*P8p>&#?6*r?(S)*EGt38TI)UP$8aPIF-%q7k7xZ z)#XG7z~G{_;n(v2sUr)WFGzVE$v3be@Vs~XJwnH#tIK~$^?OOot}H!iu4WGONT<;g zU%C6Hn!Tm)bbktT@jccHj}5@uju<_@SC1jRI2oO{#G9 zO_6!OH#<*8?SiQ%x8(mA-^S?cUM;OQMBgLpl_KtWL*(^FOdpCQ#YJF>v6f&=Lnbzl zn#B2?BKWdvjTAa9$(;AXTbnFj)Q!p$=h8y;c#ZZOFr}=L`klNB19Kr(zU3T@yNVWP zGw;IKj&AW@dlz79I>Yb+Pt3ocHM5xB1)UEmdaH+T12bMm#;c$P)7LF(4PXD}pX1#6 zacb6HDT05I5vd~Ov2GHQ`bTZ-U*;jjtSvt9DA5Pq#A8+17(@jctzJApij$v4T4qc6 zk}$5mE1V4V+xI$9AFEDz{Th5s`0>*h29G;LZ-E}K)92!R?jU=4llNq|KTzGNtzVz6 zgq&IHCp#0l`W%p6^j1ezi{OLn5I7N{_LaL{%Fr&{aX*EF2f&o`$7qbwqmU#t*?lx_ z1GGGO_&m~-V^2tYmfytVZ@XaCThYy`Bnq@i*#++jnGNiAtu=ie1n+V0LB8CsJSv7- zw`JCs*8?kTqVX4t(?5c7IjeSo>`gEB%(q z!kFCjo=GZ)_YqM}-muy5C{iTLSKXE2=v86jqtY6GwO=G&xi&LJsx?s0U3tq`$c=xe zuO5^p8xV0)@#ErBc8EH3RcAZfjAj7q_(H2~c?3SA=h3ZsCQZArU5usuB%O;F>!0f| zxqR@}YyoK)I_2}L>L+n-(3U5k6UgP{2QlaMye!Yr3s_tB>RP^Rg31Ez z^9-*d=%Jq@UCr7K)qisRR^*X5^P_Vm{*1QwK($cBy0TcWC|a=F}zQEbKLYp6gVDFm=Lilhso<$lCt! zsL3r(y$FULnj{K8B=o}KFCY7~>*E~gloRrd7ZQYOq$O+D*Fsw+z5jG{Em8umX|5I1 zh6(?7a%~?4q4wE8eagBjX#HkwJ|&fhWQlExTf20DRgz-HFOZJy(PNF}646lORZYF} z+!zPQTWnVw5c->qCR_APPFE81D0h~+^C{7X!~lF=`>8f4O z8|-I%ZjhssP2%tWa>e%GxBngwEPl6~rYg1#8~oobR^;X2Z?SxX_Rg_5!Vj@I6Q07HDo`=Cdx1z|`m?snbbG&@)(L;iPdEI(7udO8wXl z)cD3{tk4v$=P61}r8YGiam`x=b9c?Qq5r23Drdziea^WD?J=uNmp@1XmW{}HX5;`c z|Hw#0yPZK#l}GZ)A*Zl&He2=tj{@XwxUeEJr51V@nBO|FgCDwG{Ei2%*}(NXQ}I_v z`K^h4_wG=>6RnL3b$+z@qLBg6UMqgfWs>Z_SD;yX8?|M~ExC(OC}JfJswQ+VzHH!O{xC@IeR0_{myb)6+v^@tRGe1=L4hOz}ntW z9vJ_O&-#8u{)Fo}3shFogzpplN7?1uqs2A2 z`@&E|ZRwhL;#}^f9Mw?UP6mdOtg7OhUD&zYaC5TBVrVm;!>>j6f^4_UswD+4A*0}x zoV=wnbUfU#IoszN(8iBznQUByUZ*;CNBRw-%Scv^!b1-jU+WY6wRH^)w+)9*@4t*a zB0JsR&^dSs%vu-~!+(R&p+p@`#7yj1dHVaRdSu)bOq`14;0&Rg`umA&ha+j$X416;7AHT972AV&oYY=Ji=*Z| zE%|61wU=x^m>P_acUBzhA25ZkBqtAOd}Xxh{){VhC8c@Z8rp>2 zd-iEKuloUIxwjY1JL>?QnA{_JQ5(hTx*s%J zp7X<1*kkd14)vEF=0I9NefvCO{dKL|;_SE&%FNy*~%7aVq z5_*nh9+PsAfVPmFQQCT zI<4m$oyq^Zj^bowjrI1zn2qpD)@*{urAf+W9Vu=<^e`{~LEAJ;w`#c~s8)|;y`MgR z9;;yP&L`v}dkLOA2DwGj{=+<=UvyE)3Q>mMJlwEVoxs8M-rbYqk{knTubJ^ZoFLw+=Z8bJ-DKthTD23bVnyYo9c3sQiW@_FC4I9XDS9ji>Gao*55N9(ZG#}j+RARTs3iJ!r=CczEZ9Zx_On$@v?~uCAqmjt;6U|K{1La} z`F-z~*oi)OrASZI# zln1+pR!mZEgIv<0m$r5ThKiawYDwlk~_k%@LD?o{j~r zEQ4vKs6OcoqJEU1aqPgzd@g(k=l?U6&l@yW|G|M6m5m==qewiw>$cVK;!|3*p`4Ti(^JlX$@ zgTI1o{hv|Rp(DWjEIVh#XRi1Ix>ZY}x91c2FY0pb;?upH{1^`ytwER7=b_QB+@vtMtbLr8x`@f$;q1;i2G1mpqZfp7COy6HW^pxqfs~@@j z$;qbu+WS{?`qQBAU9x1t=pCTMeu}VbBjVrhY|`f9QKFtPP<_`!EP27-^WpY{Cg~G2 ze(JIRPBXuJO4!_6?TqxUltltO2n1U!5tt)QWN0@$XJ7BI*XL8*`VP zzN&|v4?9;XZ%N0|Eh#-O#{pwSZ{-Rly?|ts#PE)qRnRSR{3QL-1Fm(P;3i~Sx+f9e znlCzLBfAks?kstbgkCAS2^d))a}*4d}6X zCi~QxlW%~4T_!Q>4a2y`vDv>x<%{P(@j%cOGv6Okp|cOIn4&B1Bw6A3Lyq;NxAc>wBN| zjq%T~E>vWKoTB{IOE zb0_h5=5{tWYz4OHVg14H2mU#qvHn-2JVO2>ueV~(FY_yA|D30OC+U`WW)ZPIMQQsl z{h8RKR! zApJRCeYzd#u~V-vT9pFRi%K#%AO!VC4944n+o3yS_p8#c1P(KMK8opJABhZz%{P3B>fUlzNGNDVE)8-yZ_n+#hf1&2UM|p{YuBk_hMY*Fs70+d|qiTj4nypUXf)8Z3n5p z)|4e6OH#>ts-LsJVRUF*?Ofz{WV}piKF*#*_Q7xSj>u6+LRQo1j2KQoI+=fMz>!-! z3H@T~@^R7r8@8B{IMePLSxOR&)!1B?y^74&xcR8#&vnQu)zun5LEt;6A}dy|ws(U_ z6?#pNk}t3iYR0Fy48VB(ysXJZLogvI);4DP9qX9u96aV&A^qA%S=Ceub{szCHdA?r-*$?QfrK9sFrDISPbT5>n1^D?kKswy}~-W3>olO3d3L_ZyW&pM@`_L{h5W9I1k(cdK_ErzE=F&BXq_ZB2udf?K=HcJ>2IkdaTf)_W*79lQ|SShEziYLF?^`>C$`g6|bhPHE!bGL5Lce)a{zGMgOnlI}h zkq@5oQd;%kD(8A&x7+cp`Nqk=lLeHUveko!xb8DtTk|%B#ND6RN*frrt9^tts+eO- z+y!7hD0z5~kF%c1{NE{89|Ra+e2U}c<0iq-t2lMH?Bsl8N)#Loa^%WqLHRAK(C$k| zG4pJyvXyxs1Yh>(p)h~;-eH^F0K8$9~HKY?G3pEAvzid zQ%VS&$jAqYxh~{Wc<#^O$9?Jtfbqz0gY~Kik+h5Af#B(#H z_MOX)>>~4YJ(^-KvmguhjX8=O?8J@(OXlsC79tB8dS0w6KSt)2F?xOT!7dzY{k1N{ zhof&y7FywHX3)MI$GT5_n*jDT?Sy1G`AH7Eu zvY_yToOMctKmp*F-a%T>vOo|MXVH$(7Q(7tew7H8d%1y8>6`Drdj=6}oDIwALjEV!1`prfx! z7J6>@Y}Sovoc_L%1kbtBS){=D3_ zhR`P)dQ!=3Hk1DAU$OKqJDZ@N!XOWwkpms_y%Vyoi*Q19ab>#oEt`*jeM=ajwxeQWw zCDpyXJAtEzPV*d9By^{01CHw5EU>{L&zgnpsY=AUNgOA4MG$yq>iy?DrM%8u`|HoT zQrX9|O0lPK!@01gZ9sqRPag4q@z3=xcuLgkZYzz<)8cYec;#;#3mwXT&r9UD+Cn83 zzK|s1?F`k0TfRNv^qT>rWlnml;zpd_opLNJMjM&WtK4T57Xste_!dzCe`H#^iS2at zMV8?UDZ5#OjwmbrRO_6gIvne{KE7s)ET%QZZ*tovj^ z@3gQ{8}7O&&3d)KT`@ZxTK??ZUC~9@k&*SjEMgW;i!CU>M3Df-ap1Y8NbtX~rncX1 zSwrMe>zuuP&Z>lw`F}lF`%{C$u@}6_QkmIy))E?^H6M{`5-*yJmi6AS;(F==*zjx%Yjfzc1VPAi)Nio_wDhUK}6`8JRs= zKdDIOPxWdzGr9sPM* zGwxt#AR_CC-7d}%71tJ$1%iflq+kAqZ7Y}+7Ekn%7WQ^RAL2OU@zHY0@{`k*$PjRt z8_~D}J7wE0>`sMujX2Y#B=V_u9rkwjT?jb$6^FLXEB+xe3kN9ipSpg|#tDAyTP7P`16AVa zsnEkSWWk%yX`w=v&>3?{+;tCutE3$c*AWPPg2V0=pPrw&hF$0G7{~_5ka)s{mhw&g zCJRJo%NkifBlFyS^3!sH!vhQ~S82JVwfnHUb!=mtJjdSz_Py=TtKsrWb-5Apxm?o!jd-*hTZLqVD}lP8_f>0)~R zUc)Yr|L^ZZZ}fH&)q)QjPcxf?$A+P7Rg;KFoDp$9+l1F$3GGO2*f)65;vv%KjMYSy z$aDSf=z_~rlG-NNI&kRnOp_l@$XZ+6e@gK4)TkBGO{sT)BBJ-GOrO9f<>rm1?{qUp zs-$iI$-zVaZ{HVsufCZ2BuDha1?BvF{h8^7)Z!4q)7KJ+_h^6?4Wr6xE?8FzZkqdCyx(D;yQ8^&nzYj&C{E+C!`E# zqAeDL$S)xC3Q3hsd8^~t!UXzWcQSTTR6fK#Ba?YO#;%|AjmPrcQ9D*?MUVyeiM=ZB zJcOer^JghNtHEA{bP_KkA7X+oYA@Juk<3F;jMF~ggj1!yqhC~%ff=-OvrYIE)iwJK65P|HjoKCiz$I z^x=T`xwwJsL=x|c#Vx5ePhm`ab9?n2K3||QdqjyhtrYuK z?#hVXLcva}h|4z|&maRtCQsY{#DYU6_E(%1llac$=qyw)fUm7sTlUEd2d<{_)(fm8 z&zkeyq*+1)nj@90Ck`7y_T!lEI}H2X#^(>u*Q6A9Rt70xUjv~FZ(d%@aD4DmTO)4U_1G{e0 z+cz9qgYADVKIr&7jP06%vz?NLkwva=-hHSWo~(NOvE8eX%(J@p;`wKrfX?#VZ~jOE zTAyi-*1u83KQ{+u3mv+kYk&K$>T(W07qV2z-QcW`tgeq`Zw3D^1lNHd9*nK$qqhn*Z3@RyK~*?6_rWc)o+dmC=fe)a{~$x_Pk-6}AqJ0x;S zR20idzcoFnGg$VSf9eRF!#@37{d@S)rQ3R3e8jF|?(t433J zX8Oux7+aS6&J%fROBUj#Ox#r`#%ZYXeA-u&jmRweRsA&InJlO=vJEDT$hkVzW>~Bmb?n7}sXzGPKnZ>H7m*6$hEv$GWX1oAi`( z|FMhfHrU1G<4;y&6aDSMw_SUtx7fXSb zwwbx6i8C+Q4oz!?2MJxAn#%e=hwmqnd1X6|Vl;=4C2}X(WeJDx0cWPdjr4S95r3aK zEHUsx0!KdYzTO;WgFQ1=v1wJ@`sf{M79X!WPXgVn!Y6pCJ@#txtv_JQ$rE7cc3q8k zN6uj5r8T9GBdws;G~?QpC3|ozxo2bb={TIY9;nl#Hi+Xd{UlSIhp_ftKf`5K6=Y~A zRV5rLhQUiluUBOshCoZlPveV=A>m1}vDmpT99b9`CLA{jE#X2jZ)~QJeqf8|>U0iX z602Hmkia#2xVkg2IKf5{$ioF3Y`Z)kJ}_PQ?h zdvh}sr;OfMy@om18@eq~)u0aAWR|Ynvo#0lKl@@Mx9{!(eV^mrG(&|vO}YM~-~fw@h! zh$^xaOjMTlXe0Z#fkmkKMl$dF4Wgn$`WSK7*jw$>G}1&KC!Z)>hiu1EuT0)x>^X5K zvbk>&vNQuWxu+1i=!u#Q8kdrak?mf0=@s98WVAWZ4!xYl?<0L-{%K#JHEL?TQ*JWS z)Qi3L(<+f!os=b+$B!o(4!%&xc0k6jneo>IexGZe-wSOFEYKu)^}D~XbWO`V0`0Ev z&b&`ri4{w~vjb_Id>tlj>pL$dM)diQ4$k>rL;ojl0ha21Udwko8)&aSf9dwg$F8PA z)_0lb|NT9(&>Fp(QkxJgSaacBRgyQ3HO+VPP4XoRJdE1BrGUdjjXBF_jlSCX89)9M zIH%B~h?9NC&NC0J0ESwwBk$6WFl9g3rLvUZr%u|k)Xhxk91f-g2#c?a}Tyj3>Q-C#_5rMu)(u08_J^YzMHo4^0v zQfU1w?o8uXLiReFgVewKvij#qvt0rlz8>gnbd+_xX^XkR_Jyi+U7Y@Tyd$xl;ISN^ z?7owJgtL#yeDTvhOS1Pv>7l^m_EQgWI_JcxgG(j|duee_?1n=XGA~?Kb=xM+^`7ic z#mh5x5d24Dg`3BAJ|gPn!)sUakSM=__At%HZ*c{5*(}qPKG%tq=W900>Jj;it~{p` z2ggrf`I^SM$ZSBC23V|3^6Gw4~7IZapW?M%zS&=Ab^RdLw3+Uz^GSsM{Oru+c3-Hy{dU^;(p((V{If&)=`HBkBr%;= zH5*ww6X)xd6Y)Z8o~YWDtDJfpj5zKdzPD}>(CbV*I*9%(tkYQ*FeK_q;u}41@vDs+ zRF$7`BgMVISF6=|iVCI4f)@>T2dQabhe~*s(4xi2IDWFcn!e|p7u*4_o!VwFS* zvi}7mk|duci4xHJs{5|Y;-B~}@a>Gbo*8(Gn~&do=0%=mRk};?g$=S7?{zDcWdqf{ z{oIK?>OiYqLQZhwyT>Ec;k-q5_zQvw7n-{#zGhu_jA~$||Z1$HEP@e>48U{fZmnx*&piN|jZ^HdzOYX799 zUw|~f!aR?K$`~1J{9SfqIZg4nBogg|7dQ7dKG?O$8lo5tG6AOt2%Z7?m6fS8_6VN`C`AY_hW1*U;6m! zkM#d1FQA*NL*4{^_kxt3yd=ZGuA1Lka$6yoZsKtGP$>?Gq;He(An+j6s{1OI*Eu}v z*gfxk_}QKrU}$8Rolk!SRGQblp$l}Jbo#>jx;PRi)T#YpP5e03*{+qpmB5F$e%~i- zyW8$x_Lju+s!e6Xep8qc4L@Y|fD1qLPdlB}vG|;|J16gq!x?hH`--{y!@A!c^j^LB zHZVV(+3;}d0&J@l@Ew>Zhs=wSW)*vHBlC&%f}Z6gP__4R(lRv*@H}hS$m)BJJq3oM zdfB(hywf^KXFwe?`5XNyxdB-JOq#acKm^m)ozHrEkm!d^KDzpTN>)FlEn&JIwHJW_ zv2`n(tBJgGpG&gULEd-B46RhzXdQ)rR;L|NtQaQNsiyBWKebjIc4#uZcI-d;o=&!& zo!mD6pXYy<_Ncu zhwJlBXZNj%zM{}Ocjpz$1$S}ii`EmdFmC0 zA@fcs8=gBUfw9qf9f?1#l4r;5Fj&*Dge>6ZbNi9eJ)jA`KflTQuiZKDiZ^%09!E}A zvd!kngJIxhVSe=RLE^G52^Edy6Fy2wI@N|v&D zxOAMb`|0hQ;xCB)<*a*c6ensMv5t}@={L8w|BXM+ty!LJio-X?MrH=d*t35;aJf+{ zPK4;(j<0!xoq;lL$~~&^+ruBFb!L$TyAq#XUZ=j0p-YG} zQJXNxDURqGiiig0X9pgoGb2C|5!m;wup38P=1E*K(B~ePs~InbDI9-_%r`EyNGUCb z@SDoaCGU}j$g@)G(<~0=Ku3J`_}ZHP2bueLF|jf-7QsTiT)_R4+#k= zQe66S|7ZJFXExBXP=bRefq~R74eyr_x=#b+-eoN|9DEwm7o9}W1=mSDcOQ5^j{Ayn z2ep;u*FVFa2ZwtksjG0<_T5CHL<3GtVtJCwUw^2+IJNwK9=103rsea$APcP>Jtk}} zi0txhw!-Oufc8VHcuChU{KQwiv`dpy|HZBcFT!u~a(Ln(o=?7M|F;@o86GmReH#Fj zl%v*C-K?T9;X-w7;YaXph;S~_fu*k z^!!kY3ip_g?Bj8DhX-t-({T3I{ZoXUYdu<bD~ae0d?!1(?0%F&tQBtBQt)!d3CV8s9K(-thi4)Ndg5QPF@7Q7PuF`$e+ zzuKcqMWwM_Z!>-Alf5vM^JbN7O)GRJu6pfXKY?s?Hh#LY5j={{mlRa(Lz;Zw^(9}| zOX2@AJq*#ZE2#__apR_|6maCg0v zMrPJmS-wl_aOP*J>EXJ!Wd4^oC(9pj>>YV_@YP{69t-#-G5A&fSP9l@k4LVj#^TuA zGRt)Wg2-6^ruOz?3!K?nrfv6C9owy*q`99-B=bI2KCm>)hdjrv%xd)?AsFa{1&R)w z`(kIdvWMNq{W#$77;X}=gUoxa#I?*L6IsDQLxJHTIC9-td=rTL_d0|&)O!3vHvxNU z*6}1K*I-&x?%mv&Wa!NDp6Ag0*N zJ@P7U4VnMf6M358rN939?Xy>ubjZ9~FAkrmIfCQa&m+u@ACm=+Ei#ij#RsnItQ{GO zx!82IU$~`Em&{*QDyaNJYKX?qspV3Yk&S@qK7!;{Y9Cg46pC!&M zXTy+H*DM#bAOIQP+IjT1a_UbwsvLHo_+|Y zE(n?G;pSIXJ;SayMhg}_CvYH?veW|km?%t?C~vu0#?c?eDocN>CCxj5ai+w3Zk`A- z8y_9-QRCEup)S1q4Qth2XwvElQQOTmpPcUvZM~AXR)HJu(fyU)@ZLFRhg%Nf*MUnUCp>|!EAYWxIc~T2ZBi=+uGSO#C z(c9J8$lZ_2@gw=%TK3T-PC=)hfbbRVfA&dp`wI?(&vuFvybi2=k2D2xS3=+7$zccG6gIfYXff+lB&w>qD)p;YqVn(#30dJ(y?Utt07GGaVQI!PKE zzLNO2XpH z$jN76f5FA%+k3dyDKaQK#B%L9_*^VHh*Fvnyugx6p5+ih@Jv(F=UPvRd?yKfUft3+ z{~V6(iRgUnx(_qsYs|wq`4_pp3FJB{wQV-+i6g(*6gqjQ5r3;6@t~PsXt&4H@3p>g0dag_#2FC6=wz z-TuG~i;C;l7r>mJ4)OOs1g}Y5b&{479|b$lT7Mg`p+W1H%nwcT=V7bY0>Oq)0#LJe z{qyfBF))xh!|M~o&6|5D)bCd69LSE&P(g3t?5 z1~8-k=)yCV>%i8VC9SD;?qB;9s4iiiR-XdE-SU!u=M-1pE;7u%j;(qW1?@a9Wu`oy z136{Z)V{YwpIPt8GeS=eEP{$t4fdC`zXNkO{AwOH1p3hxJ)%FQ@XJlz`xzl!K)D*olJ zfy@OmN1sLi0Osf;l3^EN2SzWn%dhQO4z!xbhOP>PK2KlNJ`eqK8Ng6}5wple40e6a%catBlr^^kU4d9sB;o6M5e0yG5{^ zEddOVvzN?&k3y4qNu`mDH8w5d9okV%@V(LHC-uX_lcA*zjZKCrT))Sr-b(q{B?jH{ z;uq4^a^gRr&HlzCw={$Edt;{OucW<>0IFHZPO3@==pP_>EEsHma$?LdGFDmd^S6(NJ|6W=G22q1 z(`oqr$joV=rD~iIJ}d!swW^1cD-D6>_rU%|djS+Lm&v_mY>gF~bJtJuJ;R1Rb&-O` zKx}{E&Q^5UkLCPL%3^}4NIqCKSw57DeQ7BrLwl|R!!c6t(PbSH@9&@E)1!-lT~Va< zO{@(o|IiD=k{)1NNYr|Jk|VL6T{q5NVe=knr7_gy&pG%KlHmM@p|hM-kl|~8F5Q}l z=U7RrLe`~3!c>sLz2$G@iTebvnT+!u23o@1g}oP}vA^q<@61{fwr#pyB_GQN)UQKE z5fX3yuAg(2$__asUaxwyQT-yUQV9DQCdkP@lX=dZX>soP2-Hq9f4}<)kWhWULG}Dj zeBZo5_f1+O&`S&?7)D9JF7x}c^5S)*6qcne*>VQR<8Q8NT~UV)XBo{88#-_xdf{Bx z{^c+z;@-3PswQ-lA0Khn^#SS*rERPBhC>^v=Eds53}DA*Td?&afFXC*BdFJz#M_sr z9DK3{nm50Yef#_{m3R(MxTQi`D1^-;!L|o4fIkZ+J^eE@~n9Yu=Ge+&u zF|EXB{GJBGf+tr<(}?_CXM)=b>p%SXhdMI;;J{j7zYH!OK0E=#3*XQSXU`__QH6gD zi(7*2Lz}OMYz)*EP$zbnea;R4lwkLXKW@4VDcnszNxGmHuL<6 zsk!?A7);a76vt)^H)TFBUcLxhKIp`rinqs`NeS1x*$MbE&wPB9F$bp#{q?Q`CrQ77 z@}?;>!j%uYY4Z7oCcA-oD_E_4dmi+y&V1a`-+^`Kmblt0Hv{{yW-0$`fbJsbo6aTr za1sT|_>K{L(b&=b>d^|SH8ck(j%FMw2CAL(j!hodpf7ZALzC$qm{EN4f<2)QjAo(n zK!E|sEeqw_(oF$+}{lXzBnYZIuQ}@T@%=J`YX6WP{`xSv~D;}E5Qm*q6YW-8;hQSp0Ec+I`j<_IDE3(Cou@95@wmgwhoU4sFOF~}X?t6#bewJTfTvLOw??WPae7umcQcdEN z6(^5|Bay|r@0KoudYd(d6Th4wMk2srAF{wOU8~sHaTcn> z&s|!$aW%4EM=u-QJcdo$Y@PZmyD;;NWmv=d1gsx=IcrJdQefP*8tE;p!K{e~rkN)D zFqX;Z8J#BgV4vk5Bc*fZ0eghvMA^%Tzk7Q+z0EHFPd*MPUcc_%%pk#6$iC{?b+|nmN=Krk zSg-g-k!FDhL8sOM&1ufiMYZ!-GH4l2FQ|ryOHYcETmYu6 zN8aDrj!@{XAuZ5i0KE%Z|7`4#!kPtNZz}u7A=}y`{<-xX>`eT!h1M4a%_H4<<=u_Y zZMXRA6IXC>b zHc%@ma&lVGPS*Z@L7IrSYIVEIQgA#wk%-yL1g4AxzbnNv*0eCh_^_Fd6cdFtw3w<*JsCgK#l#eSkf7DW)GYy~~f&%5i@2S5^bn z=KY=K`~+YodW$;vPC_Q1l*W7>LT9S8OF8CsksQ{XeY2-|t2$I~8l0#1>k+opKdcm4 zBZrK{<>xm2I04L|^9dg#FaK+tksdF>=Y@B%efNSKO``?K-YtDO?i*3>WLO@$m9&W9 zT~3*+bVqv=CohDl0$$qQXI}%YL1@HdoCJ)^uaiwH2tFRhHfe#I@-kS*tFhj9$vj{u zw%l@Ec7y9VPRu?MiP>aeXTKKl&OQ#^6U&qic3yx|3lZPggB<(%_k5kRrzNKzD<8HV z#PEYS<@av%fT9I39vqv^H|H62@>}JRdisfZac6e^awPIXY_+c&!V;q}Qh8u?_stmZF|DKio!C0JGq${!J08Zzwo5xe>2b6vu;nJT)n7+1u#5C3u=)0t;YBoGLRi_!k z&v7p9XyCu;%~`KlkoWoM$@b&GI+naUVznC<$UNP1`FAsxB_;~K^Rt24FXH1@uIz`T zxUO*3$2TDKoMw56_kHNud;iR1)d$!<6>@#1q^DYtJR+|r*r%|&le@G@#skt%*;#Y@CswE#Qp(= zo@At+CjXju=VgJ~uiBg! zG)fjSUv@ZJPA?Ceq*+L056-@k8l9X9_1!#NMy&u@OotQW3pjT^u6STelB75-RT^uNL+pzi5 z1k&ydR!h(2%==$EDfMh~;5T43YNcJVkRtQ$s>%qs*#c9me1@p6Irb3x1$11Rl;>bc z^Tzv^WtIZ#l5)|32iM_GsCkH{dlI&}6s%(PIAX+8ox$g_G7xMRI&LJR0yI78_>Y>^ z(8`{=ak}L_{B{a^+&OX#GySAHKD~ShozfeGpVoT=)9u0U(YqX;2<)1zCTK$N8ez=p z-P@aPEQ1+;t1ZUL$B}*Z(p4*m7wDJG2y>=bq871m6>9!5SF*Hs^QdXQJLW?%LyeHr)i9PSo7m^fUvDsOy&{YdysYMaSpU`LVz% zSuJ4qrVUs!FFU8q#gJZPembsg6IM`X4^3>}Mdqa%el2~p6q=vSS@Y<$K2S|=i7Bi( z1AUg4PUoD~0(unX;6jlxZ1^pG=6Y;7wwkZ_oK>F;W1qBNS}WZ|3f(zaVXFsxUk!^$ zi7CJ`lHPo!$sHyWw$`-Vn~w~&jZTFg{FttwoBfn@8d&FgclX>`0!+Epcf>_av2%}A z)5EHd$YR+FFPOB&kx$X9-MsXWAwQshd%-YS&|8J(WE2Wav&ikiIeVbP=e=Xkq84-( z_FB5`x&^d1eA!swz6ELn{3yo@$DzL~&GdBk7U-?Gkp5WV0aPz6JEJ7ZT~DIYHy*0h z^Woap!-^fG1Ph}7pY1Cj<6J|WKh&y)5jsjIfPQ-4vNbtedT3C0!&p8eod-vA3x2OY z68hIaG;#mEGY0bN1$U(+J3wa7FzIO0Mui@B0z^!!Qz=xt%RZtjAjNc`^H7drUPcnEqhE$>&37;zwPJnRU>5iIh_k zKaS;}_s&x}sfUBN*7+LVJA_d>?vtNW#UU!=@;m*aY8Z(8@aM*`31mLL@%hiUonY5h z*%BPrMdt6T*WO@4)L+^~e>w25_d#mE{4R-kaX|mB`DTlo^ z=6S)Oo!(;GYY|xEd=!oxH^zaL_06V|ZBYBgm(X*95ze z)&5OQX*Llz_8nZaB)M@Zwp)iRHTfw4G+(V#07*c$zdH`>1V-~c3O-7Nj7{HrE8GW3 z{61UkE3-L%8UC@m&to554lUJEo9tA$d3R}-8_zyeCiq7F6ds_6{VqXv+F@_ir-A=k zmn7cwt!+u~Id~pywN}||mN5!#euXBh&l7wr9h9bp@As|)_Uty_ua6U9B5S_S$@m7y zzaSw&Cl|u74Rz{>@DpI)CH^W7s)cZLM3SBGOu?+fO~-hpZ*b zM-um5f<{HDk9{X)fjJOpaEE;aCar1~-H^BktS4;;gO3pL{J{zpfrP$BV6qvzcTR7C z(3cDMB<;!u)>ECJ@;|>|w65d*Ls23RN=;{;eBDOyDV7`G%3P$m8t5)}^QP8b#Lo-; z(uybEVZoz2`3I}_U>5n}5dXX*z!+XY-C{%NuhA|n7K$V_z-W(%YF26kvfkLf4oY4C z3?7=;t(*nGYz)tPpKA`a*E3(7n7oZ8H=BfPqaH)e)<}z6$KR27PFkl7jG19+%hb=} zAB&JOexk5pU_Fjkt=lq3&m3A0ru=wk;S2euXFrW?A^0qqcVX?E-NnGTd(He|ximKE zb~9ONZX`jE!cj`o-}g9tX!gqQnUKZ*am~qS1)v11^-B>u4_)|K)>8fdu>Po zvXY-%`PnRB&JMp?qGbVzZco3zysJ*)i%s3JNUaCDA3S+~LdF|@+10{@AsZMuv3SQ9 z8i8NVobkOEo*M#mdWu@vNkZ?e;8&%sp@uaK+5b6YxQm5BkpV`Jp63JeFkcqswJXri z#QBWH68abPD-x0h?oZ&i%`tGCM_-Z@!|6Ds(Xi6Y>LIXZjBDLm z6`?Hdz}O~-1D#*)3romw`ROou%Tnfc zsw32mE$EWJm<1G{l(F1*iP%{*R2;lJ57-%fIx)``VNlRW^{c))Q13V&4P#p=TON{`?~gzzh!2;!tqbu0re)c2%YzD@V$!_I#F5V8LTgXTqzD?Lm`o-x$-j+1%(>=Ug(b^X29ll#Wx?I3_zm`U> zQWAqnAJPViyvx{actPm?*G~8pcG2QlDWT6Znd!ZcvJ#-XgQ?acDg~4pAJLsdD}Yr- zp=JknA$7xHeNu}&c3$Z1mo2A5TaS+Mpp^^+g{SRrFUvq$z4rn}ofb^Fzm{Kc{})nw zeqFxuy#P8lUS45ZYK(vO*B4%ypMVh?`V!)Do3YO!V19ZOKeQ>|6HpmQ0;*JAVzSsW zoRI0?ke2igdPK)NdzQbzj&DtO-Sx+?W1^vi{`wGRZV9KfrV{pza@%etX_gX>xF|NA z?oGj#WkdH?FB*bp*R`sQPUONcE6|cz`5Y-n9fF*Y! zF*4s6e^*vt`}m2&e?jJXbo#`K?nvk`D-<0WIfavVraP?%a)3V3XD0lo0UM_?6jTo# z#;(Kf=bs+l{x^P_e`DJnj$J^S$gYDP!x}&v@^lpyCHRb6pEbNl-;e=8_6w6r-(=#4 zz=0I!;Un;awI=ggYAgIQBeh8B?FUAqL)~$4g0C_*jmfq%+>E99zA0B@+JQO$k;FdV zjZn?6BAT^{k8}T&{v1YC6Lp!6QoKkj1+ z{^;ASoBe1EBDTAvJ>2^XLsqmX==9k^2laPz;?Xpq?@<;xmv`-NT+sDgQIX(*%d88V z-JM_oV{WW{W0XfgyPfsB#y|{O>?cgLuQgyx!@fmZT@!%8uPakAJO^8n-$Z$8a52>C!|kotJRDm_>{q>Jh16>;w{`v!UxwAGB{nPp9#7?<98t1Yb%%4Hgb4ou$iahT(Vw*ps}8-M&=kIb*X7B<7OZ*zOEQk{F9C>(H%@F%^S#&-NK4d zB=XsGHxtFx633waT+dgZZ|YF3`{mw&=`?7cf!x^%jad2etH9RvMBjArr$a|Ko7s{1 z$y>wM3uPc>Tfoapm-(>S{pXuM7OvhoLvRMvX>FP*`5fhl*P9AX_-rtq;Z`D}$TbKCKY!}Y-6i=r21u*v-4 zTl3G_voLd`nb^&iEV2+ea_MI8ZGbMhmAQHZ|3a40%Fiv)2G}6>@#JvUM{Jw=thPjn zgBL=|^iR1I0bOW%e)G$U_Bf1cIkME;e-zTk_bvLEMBvRzrhj}Fe9Osmk_8+f*)w7V zFk-Lz=B^4wD)o^0rbTs7EtoqJ;46-KrzLnMyi#!FR>}?Q4@1zAz2Snn!4+UcDKELA z+yPXOefgxsjj*eV8%|5pFTlWS-p%hPV&E`;!H+d_&jCx{W^I`^6NXeP)R*efyp=wyvPdhwg!Pbk&z_UhO92sD>f$@HxS7<&IzN76_orrP7QkUvL1l_VrroFu$m z6nnnj%T<4v11-Gc_K)NzfE}GNYxx5{WZV@V6zn|o|Gf^Vrz-bJ%$tv~>U~}&SPYG< z3sN7~9)OYF3&Ht}R2bex-9{PUfrhPZ-|`Li!jKfn>D2R7V35+>hRfBUdzWjE0;fL+yG}h0dGxLYSo>1h@0^U#dkgd8nx_fS zhkNja-zW?(I~8`NFBIsuTIS!v4k9f|D>lWD<3B;Wj_jiR+5}*BDkTRI0yTg(F-*$jFuU>9bl9Zee*{b@lj^z$Rw`$W%$Bm!4 z^A6j^io*w%LWk5An_D~PS3o;!x6y#fU0~0X?b%>z z1oXF$UcZT4gVZfN*S|fN0CwR0K39!Wr1U0*q+T2bx<)l6TfZ9z{=|vQ_F4oC{X23J zyd79xcF*f#cqg!&YuA@A|GR&0^gV+9xiI3PS$S`bD+XTAjNZ`n0Y@9>KBh}@`mA9T zA`L4yA0hL%9%|ol`TXDc6Ix*TKnz>uyC?RP5;*D-xsWq22t8bAS!r{#UG^8$O)Ax1 zF?fO98cC)4u2PRJnS3GHmHL>TiqRIK^*;W z=+LzlmVsSZ%&XL%5Lb_kk7fHFc@jMBEHB|gV?%Mw-1>ax`tqlkK(SgU5mv?7 zMy&hC+psNSZOa46033Ga8S&k(i~V;J?emso0Nr@`SKrcQ@a4^Vi!aIIz_3j*xJ5R> z&s#L8%-24^EZ6y#-YAc3&ygKcw@qR6$>I{jFI>7ukg{Y=&GlOZKI+@LhtfF1S94<=`|hapRSgD-*Vk_+KXT2ivznRypT22CHK&|h= zf=sLw;|srJQHk$MBu+I_nt?rCcgiVLo-EjEc*eWtJJkDT$u>NGiFFy8w{CcQ0M-4} z>cxHp4^_{#Sa`Jc8@5VOza7c=gE;0{xY*q=a`~ z1jhcQ@!HQ80dA=c%4(NI_WW7o%$@wm&`H{U@`VsIXE`6-`7i}46*g>Lx{mORh7@1# zT|(jLdutLu7A7p(ge;Xc=ra+ zQ3GkEKOn#FU>d(_K6X{vomaSY2`VMerzL$n3SAwWPwup|h2|rb+aL6fV()0P^o>P< zz%s3p*Pc`%@qH{zzf|7@(`(*K==>Chnwa3F^DWHKe~%`U&+9Z|L8x&9g8`+U#eMY) z{{#OP(ucn3wUBT0cyHr&LO-hDjCSL)m5nedf8m|qvBki=?i6B=_kgx(&Ui&L9a!3X z;M{Jv|DG?-^%L^KKFb%L0=g*UY4A%z-~LjUT39NB(A(-7kNy9&_Z>h{HQTymMJ1?+ z7(t96VpcG%0TU_)L`5VhIY^Rd7F5il7%*S}#e|}OAc8^<5+x2p2FW?Y%n&3#&ar!s z|C#GM=iGO1-Fj8`ukvp??$v9r{?^w$ySvRNOgjzi*QeSG-*V6SXN8J36(^rT-s51b z)@Sbd8133O8rRH=Z;HYV&n{SXneeM-y z2mGn;kWh|(w#8u&Q@AVXzUY;44n$tQKC$;{HL#x@m2R1`04k#8H(Andq57I$i^YXs zP#JyM7X1ND=y>#Uaf%rCUhQhJ`Ic5v$DuQ!tjAS#EVN4oX#1Z?LX5>vz8==#-V4KN zJnefdO9vJ8nn~IEH=``^)wA={xa*(KTk}s;+U7&B<570^88cL|y5hdf7cNifl}?UL z+``>2$(GJ-JIHwi<>%L&)7e@=r5E04RGfA^8l_Z^SstQ+=a*{g3+Xe41{tY!{dJ)!T)Xn;CeadQrFD zbA6o<^=9n89e(Q}uU}ip++QA5Za<)A`<>a3?PN<{5 z9nB<$wmXtuwjdG4hrh%Hek1%MCAlE&czarMwKt3(wCi>xf*KE(oIMl{}&D>$SKDvD^Q?1D?7bvAGJ5w3_8a$ zr|w8~gn+n@e5Q(BT)MV^TB(RmVriyL3?NzR<8slhuW2AR|>1%f$AS($F+Jg z(Ce#j=d7){gu0cKIg-^rz`VUX(>#SbHw;zG*dSZBTLS%D-|7;7y%;_I5?!kpU<9nk z7Y!R~zrYoC`b{w5IS(E^Ui(AlXyqalx1jjS;lOLal2$3~(7J}I+)q6-?l*-rqq!fR zs`l{ID@&`YVo!|vlfC`35YXMbu zy?%na9~;oOp-a~sxaVGWozJ;-s=o%ZwD(D*ol2(h(L024$LvFuXsfo|9V4FSH5h%5 z%dC11oI4i^a~(dRPr?nZepePk*S_if{bt0!=s7UDX?Dzkd`^7R;@mCe(-pg2c z{%nlwv_@dfwtMkFI|pj!7B(&&zY4n7caJ@k#xst&>jD-|=j!IJLh(|nN%Oqq|5&d0 zdf98&#|-FFJXyT4YbSJsD$A6#IQUm~p(t0A%o7>l>WUVf7aEKzku z{*^*BfeLrytB-%8+^W4wMjN>Nx=mM)bG!AA_d-CUD=@9d&i3; zD9|23-=4W)o$V6nOM{M(>-8ZNxh7;v&dih0CH(GW!4orR_3^CpJgbbJ3MA*B$hm@Y zXM`&H%*ltYYTr!Dnnb8A|1fd)y7{R7N=AiG5%+hGo(vgryI4(Ly*-;S=e1cmG?@8} zJQ6w$s*j)cy5A=YrRODGJpxX{&wb)aY|%}StLfXbRYi%Xe%00$i=qrq^R~-&Sa<4E z6L4H7e>^Zn3u>5P)qcenYK_(k>+Zh?iMo3povl$o@v@8e?~a^;8a`f{-*99M!%3 zd)^oIwym%dRb_TJ{20FoGB)kJd@Yv`6)t-?_@MYZWPk6lTUn)r665_&nB_zbkN@11 zs3`6|c+EGC+MR3viCQP0jNbdU6}2}^-Iuk04Aj2vPnetXjV>tRC!Xp&5_P}SziBqH z9aWxO@-WH%9b{|hciuX$&hx(8zPKq`KggyFIDEJ}Qo(>nPd;m{)U&@nL(g=deO5hr z3m!|3JDTjt^@F{+B^%fH`=WyHpV#(laraTMB#h-XK3zgpp=tSI_PM~3hiH%W$05nO ziQ_*i4I0uOizTkxfIc3UxbA*rE`&N=R4(qK&h>yO)#~Mi7Du2vOr=;#JsVZOh3?9^ z^5DPvbMd?7vd}$k`+&0E3HUi^&o`lKIj~kOp?#G7?R%TA?CbgZYoX`D;_J3$Iw)Vx z?t6pOTh#5@qWJaMZRr1$z^DJj9&rja;!I?PQ0;cRz-y1a(Zx4Y`1M5Iq26zER)6L< z2bRqQp8(Oth&B0+E2mQz&b3?3PYU3Jj`s$;8m?>xmb1XidB+z6=YoEu)tY^%%KEB^ zaq?*N(oFu*JuL@dKiWFp=S2|EJ%%11NOM9)r8i_PLTIS!%gG|ohE*tWn*E-+54X~V zW*P11er1R}-d#d`vID?5^tP$;{8*@5vDv-w$qrz3FdNsT{PDg+lzTL8*6Q1~s8zej zxN@-u)VeXB2q=w({w7VCfGqBPWc7{yw-`=0Q4+nNOEG*N)aiP-TwYd>IvX2&LRBU~ z$Jn=(cd!2OTydy=sTpb&*#Q}!(xyCP?BL>LqIvJ#85wYYtEI7Me>sTz?Rv6i5g!B} z{Mb2l%pz2uY4^}$`9xF~_wZrr&PwR8_7v6`d<<;KWX)%@8d1kst+rOlFFenyI6t=a zo!I9;%EM8Ta@idj2ps>|sgow?p`?-zTh%2^5bLdXeeY@~)S8p$b!vY+uqxIppO{TdtRl=FvBF-dNz^%0CM+Wagw#|3K&3IeJ>T9EbzFUZ zEMeyrs8G8!cBeTlW`ZJwf19U?W}u{ zGP+oGx-Iv6nU1yUL58g*@HzT@idvQ&)a{s65o5X)*vI0kDjUk6C1JUw!+Z%Iy|91m zQTF3ojJoQ&_}^ANg=V{u6;6AX0B0uOGMOA_)U(le>5D0w;LYGMzvTW8(EfFv)q_V@ z5Jz)c!mjBys4gimUqsp*-L9!TlP}Q#p&yp?2F?qDl(8pPMz)=W_%pOTK7r@ZE9Lc_ z6`TqWll`4PFVsiby$>(*PrnVEK+i?rew0Ax$`M&l864<-xM6ad;$dj5X#R2crVQM_ zylGDFWkqPp=bTxjGZET5{Dtpqe~CCx^l0YkVu%$jb}+%x2kH*I$wGlC@SU;HZ~K}= z^f{`>Y+ykD!M2Cl@qA>7xCMUzAD$XIw<$`mr*QoLOzSyL%CI))$9MAMJzU>1xvEdeO-3&}pxb z-V~H8F(>T$okU=j98x$#+X(NPAG=m<9)fR)QagIgc=qK$cetp`npiF9tb4hEea?c5 ze}Oc5f;h6c{^@_EINYwWTd{esdrL1NiYS}DhC_N$;env4Rj`4Lt zyXF%#LX{0ijwJ0<<)@*%R|cz`t+?l{rF|Uy(%?kh;{>g0cYGwWQ&Ee+Oy{-3^8$M8 zCndKq>Rbz`3S8O2Zu5Y!w8c*X4f&w(xy7N=4&2|J8w{4D=nox$%ta*9m~*mU#+`ZG&kd#p+UB4a=ZTuixW^vI2Pb>n%!lX@(}1 zIW6M7k*M~E)ZEO^=b>GTfnzWd%L7}Jo|C}=!Yw&oj2OEWeFE2 zjAu-@zd0EaB_1oZdGAFv%15eNcdvo&`L~xKZAoCYw$0iizYcYHPkzS<-UggUuO45i zJBu3qou2L*GY)mGyKQf?;}vAQ-EU&|tr@C>8YgXD!Ustbhiwu!?FaVd@UIbh+0be% z{bk?gI;e68-@CZ47WtiA!mb@5gX#=wf8_P)qPm+`qepG#?qBP;@>93+N+e=DFFs-6 za~(Ls85<6D`$Fg43c1etn-I&~?|M}D1nA|=6iB(e1@69>UfKO}40Me9I)BH8Sx_N% z_ziyq&v~gl`sm8O-B?yH3g0${FgC?=_tX49D{_y%352eIWnKM(g}@2jpFeJTGD_;) zbbIbXA8?NeTD2hNChAPTeo3f|1%9*Kwth_Fxi1N|y;)$MsBs)p1e4U)`K^Q3U6&s2 zOXs;K6*>gtZ*}gt2tx*QWQHu>LA&YIusbQ#`Y5W@nl1Zz&JI+6aFcFjCABU=<5MzR z{=haKb;f$?u#~y`TiAU?Qcd~?srVf!OUsX!0-*<33klKlci!tWMC@^Qv$h#9 z;iuv)6MNCiP_a`-`(1M>>fdUl*voVR4@SPkrkmsG!dY6$i}^P~ZOrM|25H{=U(%My zI6QHuez!+=t(OR`&v=jePM--|Bm54v<|jCvh;N0O4CS)c`*%@`#MP*KDbvJY~fgtN_lzC=AMWZi;%YT&h-O4Heqk5TtRKa;2S$-p|J*i!N_1lZMq z3yLDPLd)c9KP~3PLU^Y31BHYFy1+KunRO2HfUTS7>_0Fb+N&xy$hT7WfFst&8}Fn> zy+Pdqd(;w>c<$#z97Eq=zwBcWyYO*%^~mQisAl8`{0h!SL8*Q=!GMK=$63N!edaY%r_@~X9{ZGw!$XWZ#}A>{YdAf)+Mey5%ekfs+uV2 zT5$iJ@znQUaK)o(`9`@K;2g1S^B=+UyC(`Ra^2@pOnpxS?cQq@2AR?*Img4#yk81B zXM9LwZ)--G%@!jZYpYRVP)WgL9qRiLx`2GBepT)>)Z-G@v&iu_eZ;kUa5&x>*g+TR zTHfPO$@#JxDYgo#c0N%OUo`9|n78#+Pe`DadG{=L_#6UuMf;6j)&Tl?&whFIV;V$x zh)J`x&!ggj1&QT;^B`bhuZQ_Gci{AOxLn&d^zW{F(1pv3wte32iSiE}+_>agHR^N@ zvN`qCjV`p+;J9iV&;44IUV769`-;y8Li_4Oz84dCzPAI;TcNYBz6v1D!nA}jv4N=F zkA7%*^x=OjN8_swo*tDNLlZjcHAm2C5zlkJrS;nMKH7k4O%?W^&Td8>^T#o4WAmWX zaF2nfia3o=&~|^Lx+y9hfB$9kuxw3Dp z*;)>CdAz>%?#@oA&>a2J&;U^V@T1+>#f}PzGGME1<&$%&~cABYWItH zNYNOrCAnY{uw)u9hYemw#hr<2E^-m54Xv7MKVm&_G**B0^X1OxYt8kwJY~7x=`xPU zH~U@w<2h8kxYpY@$L#{|_dM)@>)G?xaPJ3Vrb_v}UKcye(+t}Ls~pj5j?}u@?|JU6 z<*u8oF^HTMb?cwvd;9(US6^`N-=m9|=Z24Zq6r1XA^t{PJm=*?xuKt~m<1DB25zzy z3>Yw|d^SFB+9?={o?h{DSscoFR(*H!`9In#iX0U%K~nx5)aK7*p1q`lG7<*s7PbXJ zxwovr#sd34*6V~xDuGEKp=Qh2q&KFOkb5@U;fO>fR7*UfJv}D@owD*TzuXgrwz>Ta zWem8#&sLn)^(w8^g34{%$DeL)hoAd>Wp7jq0&7ehM`G7j#JHa#u<S+_Vj~_&>7Ysut^HF>fl?teCFTBdV52l0jhO>DO3;LfQ=yUb%MoAa$9)7hx z6p{oyI^!%2QA^2)wF&jydy0G7!+lr=^B}@9V}V_|Fvk4&ASPBA8RwLCH;x zpX`rnDD}eXNoMNfPuJ{IZ zs3WF~_4w|<2f%1PJyq^|7-G3ED0_T%8Zz~>{dzBB6{=D(nL8NDb5AWKYH0*sU&M0{ z5wG3)=$@*L;kkeD|CbQu+YEWm(;NYvZ}kRd#3@0Y#fWEzPQK&mXWPNRnXMk&d%fQ{ zSCq~cZA9(IpVzHhtPTBfe*IPIx8Wgk<7c1Pc+{nzJ9fV!HJ_sio?bq3(u$eLecCg+ z#vW~WKf2#4ymL9UUK*?Ml(nCF4y#z^(3h7y^K!&WdU3{e{{*O6-l4x?Cigp}pIaIr zu4pavTb3mMRO30{i>ICDqgLA#n|bg6%65*dl$@;zm3os>AJ->9p6%RKA!UjGw7v4% zp*{9NDSxz&g1fstQ~2KSq<8Ih8TI?D|TH~t=T<@e~{|B)@2z@r2WaHAVc1Jzbn1AcW(W}{m!ppm16wr0-o>6QG0X1 zOWpKaDDL@p>1dG^G?A|*QwL9;1yUF))@L+f4dkciWAbP5Mb6uq`W;g&uh4Co+t#S$L;gFf{+MR&(^LwB3^yH>tPKuXpR zl$5#tuj*W*eo6Ewo_e)o^fD)|ub}Y>jd^CYq7*floImDmy&vMGe2NOKrJ*9vXu=~U zW8iEWYV*&d&eec+X&+j{QE^}|kCZ$aH<~t5FyviZ-JXAH51j$<=KATyJaVKAW)Iw1 zHX1sat-JOKpNGz~mfnW7)cgu14rVH;_wwAY1m9Lh-`;%lCs*I1M`nj_D39X%-i+GEN*BnuN#D246AcKDF~3r4 zNEenqrgqSm=e`&o|0Fb_!fe*m>Oa1Z9)rXmoK>v01PM%ES+Q--_ zEuJT{pnHk?vy02kP}{R-TOIZXKBH;vDrHExVDs-=O2p_$a1+2%T@!df_vdsr@*>wtC<-(Z`mj zeof23TKRRYPqu~nWgxcc)6Qt-LRgHJs5fBu|zLsp;S%QXI~LxRoCiD;-#^QK+t z(?9CJLO~<*%`x8JcRS>U5}R!b5PjU6idn9my#B^SYC_k!N?w1vdHl-*7ajkZyj1z# zX+!i!LHqyJ@_+0{F<;fv+)+De{N|3MN)4M($j%fewNZ|!;gELXHfLX`5I52FnWGN5 zYyH;FU7!T4iNSVP^b=6u@&k27u6H2q-Bg>)%x0>+*RJ8G5_BJ;Xsp(kJnsSZ$fuYevYWz_9@XI=H))jnzpTfOpzw*OWVrje(;n3yKCtvN z%8}Q<==*33gq!OM%bnq#e_c0I^qj_9?!CWF57`ZV%3OTyda&KRA!f|5|JaZfbn1_M z7n0fCZtEW{-hMU=82DAq<@N>tw(M$FSQtf}RSGsd`A7vMiCvKp#CdMAZ zlv0b&ZGML0Qw@&gUgmyhQkgJg-)OIM|E`{a^DVecTtWUH(`h2|VFT|QsC{P8Mw_Iu zmTxk!#Xqf$-H{6{%W00cR`Z-!M3wI>Qx#vR`y$4)GJftCI{s+3y zj<`mzcWpfF3Qq87zL}(n3co6EZe*#`gf>O&btxM}KQA}w-wmqMzSiCtc^jHvMXld9n)|(Tzt7FtM@KJ&hMdYN zl7GdR(TyZ02^$qI$NyuYPZaCGzb51Rde_rYSpQ}!za;@{% zzLfRT`M-pe-P7dmZ(;=xPGtMugATQw=M#cuQ0H#HO+6k%P@J8z$X`kvSc4T?R+djk zJ&oe$R&Kb4%E!ODU12B#mA$qPDq>wxUw;1WCRI69l+wJnN9{Fwpt)6VL2@}tUp#Wg z>+l%FUQ^>KcBUTnm~k?0#f?X;t8G|?&&CY ze}l@#87X~f=Aajg9=@;-;(lkrJ|pdL3$#m^Nwf}8^9?$`X_v9x`eM{Q;o!uE_j#y)kJ3nA_6Xn{$x|=RwBAw=GkpfvVvR9;T9e{RO_YV?l zS`jmay{dLeEHKY?XgG~GgVGT@HcSg}MjTJ|ZvLMRsQyS?t=qjp$h{Y+=(k`YZJ?rne9wLKlV6$4eTzb3%q*MFp0Z^PLjpTC|#{&S0Ry zc?`yCS!?uT%K?)a+fz}0VVr%Au?*^)8+B>xsvVHy_Hdt`+DX*zRTEVf{}y!~`@Z|2 zq9nAxDIK>bZVDJ%B*c$C;7a4Gvr=AkVHGqc^~}qAq)ZoBx6mSMwkKkb@we%yjYKv2 zsgr20PSN=+pBxLQa79_!fpYs7e}~*>>B0}&T~XKTnCmu&7^vXFxE6h>@6ffCeK02P z5?v(KJz9HiH&4Gh(#LF1)93Qh(z6Xel$~5}&H<&5nSD@O zxa%Brnl4zS;&y|lKkTm)v>fHEQLmzi=h{1t&{4guyR^6hlJePGmM`ezDgP#ldCTn$ z9oj0tE9{lw^19f-gzUuyYmU|Awx!w_HE{m;)i1;CEi7&gU z^q>0nCm#flh2|WkHKUaMAl=LBd+07r_^#>})ZkDA4KrMxo}R%*jp5L`LVP4NdB1&J zwY3V^i;vxZdp!;1W!~|+r5sKdh%nc?oudlTHBAZMwtYj%mt$5Yt)sr*M?DL@q{7DN z(FLT{FFA##a`C~Owc*;gO{imG((5WEK%5Dy4&QLthBBV#%`nUlfWeLLZpTPOK$o!y zf2Twm>R!}X){$}#)$YIIXw6T3uZF_zdd3>s%s?#heve6gA-v<^bn5fZue?#$mlw;k z`;O5BlUArGWtag&ZQK|s4W4}_RK3BOf(vgYpG3vQn`ap^FQXrs9}b6&tOnL3VF8um zmC$|W%$J|Ff85{9?PuZp`Co5VEIpOU?>2_}Z)cGf<(P$U4RM-HO*Nk-< zLMT6r3a-_9kNg(=-{$98x0eSD8FB5SbBkEX#gp9oP8+)3N}fD_8->z!Hkfe4hKCsQgqpO*5A=5%{$AbLK27>N|MUXSk(fgbB4C0UcH;j`o{pLZA5ri50pQ zJnttZK$oxO*sz~Ylz6su+eXwIHucr!(rLhsx$r1>(FQ1=s-AqJhVD7@i0&w`S(v|yL4|ByI@KlXYe(#t)kZiLs-SH>OGdJyESJ30GR zOAl(ClhZtFJa^wg&)u7&UMDvr_UpvK+u}uZkq=Emwht7bvqvz>D#V$__i$0hsh4@E zHPW+mcbYb|AD!zP{v{uElvI~F?B}^3?4R1VfX=FdR}9{HVCM#(#vnTPcVbS))KtS~ zFX(&+jtd>K=6)Yr8&LSERJ4UVUVMvd0)DV~?dHbRBPFI2z*{EwgZT6j@ zG3!u;i_VL*`@+yUX@(x3O7>w+Ok12yjc0q%Hn zvoIX>_$y6dX_r^X+Y2;-X>~Hk?osE#L7!~t+AY@&QT5b;`Ln&N;p4#ezSVv_;~bjO zw?bMm&pe9jcY;zA_F4S#_cNaI-RjB$-a(gjBL7}M?;~d2OrIR`sbwwR>73PdlIf2TTEs>rd^olOnH~Oi%^kh_?E#uQ$GYko{*z*4>H&m@%uuC$M{>S`|?9(e|2xdUL&|I61*`8E7yW~;>$94v^t?0-X z4JrATpVxRjRrbM{PgK13`6u7qx<8;!l;5G_(++4+sFwU;Wdl_oM{cp6=mhMP8;&7M zGNHNjdW~l12A=i?=1-Y@PU|Z4h_A8w=$`}4H@^EQE?`3U?vzXVc3;u;*m~xr8tS|p z#K~rAZiNr1%lW}#2fmG{WvSgxc?UYQ-l;X2-mwRD_5@Bi^7Fq8f85o9I>aMAbe?hdyLB^@<5nmhL*+{fH3jZd`$&MLv8?_o z;|Q={zgm7VeZ`up#}x9;oqB=f6Nd1bveQgN4_{(giq0H-z02>4I6M zF{}D09D`23)oR|i9HC;v_G$M-@1WkpXPsy#?a`2uk~*XJE4(na zJ$666n<$Rj+*^)#9K3*QPXYqsAI!?DA8Eh7AnKDE&=T1 zm@~S^CI4-Gb+-mjW+|P5_Z4CXbyXkykKuNI<+%?ryz8I;wmrSeDvh{kkPMl#8U7w3 zO2De`IWu@G3i%!LIv{?~2s(7qgcpBZ28~|_#n%u0@g8(uyXvd3LOK6*pB8j>U)t0k z`N#Es=sJHuF71>rx>-B#ZTdE6-ugOW8)yQ1(Eqw1ac0pIjBrF*A^EiFqg4H!V8AEGeuX7ek}uC;h2X9vfi2YZf$^SQ?D% zlZWcA1I!km^Ux^m=kFAB8<=g)AxGl-Aphw6T+1hMh-tZ3>{gCH%ISDoT6z6B%2id( zp50`Fay`_epJ)f6qA!=54%I(F#eo+`1XmTKl8;kWJe|W(j-0gM)6e{n`^N0G^*#-h z|4Kn*@m?p$VkcF+wa-FD)}^tpwAMn}BHxkI)UqM3g>Uq_?|vvR@tUyJBn!w6Zymv2 zzYB6LXA406c1r(5eyTr?J4a7G+Y62JW_`T&r5Gy3HqO4hB@+@W=3grqAqc5p{oFBn zDN3E>!MMc9M;Uu&dGJFQ%2d(3&adW!(k8FoD>~*D%28QgxaajDlnpC${CSF~u4f0$Q&H_##jN zKk8#d2}0Tw(e4GFnvl14vD=f=+fnYCqk*3jn2`PYsaL~BHsoxWR3I%UOW8kVKUn`n ze$t;B$&~Xu$0k7PlA_3tf$hN1Q(N5=SBBDFEA}rM`3f=O?`4Wt@u%(h>49c5)lB)UGh z3(39mj@`?0h8dfV=f*7#khsz%ywV~XG9#3YPRe~jnY%7sNZ4x+8M7sm3WO_xVIuE; z>yZdGeu@1P`yu)#@{|5#txrF8pW_6%C1bnv=Bz|amD{)PoUTGS)i?Js_!gm@EuO-@ zimE6l*yrHOUHed>I4$;h*jiN7B(J#OQat(*@BNJ>b{J(#vD1_SIFNU>?_SQrM<{Rl zT8F(7a=>`lerM|83#d>d@FU;muaIUnbXQtK1@Z-lYM|qRKKjc&zKz56W(K1vB zIjjS{vwh?!{2}9)*gvr!qJJVk=}(1^VZte$29zM3I3h~M2qk*FeQ^7AAc}~+Skw7K z21SkUdG|uu5GA_Yjhi)H31x{zUDeKdic&`P_32$%gCZ(|TFVP&!-q97J6IwIQBLpv zz1qiwA#-J&hhLI=8^)Wi%AMngniDm1r+74}G%Gd9vST^InWg3~GZjwd%vtVHgjzP8 z6=WqjaG+vp<`xmf${gSV#pikP*g191`w#q{oBO`*>w7()5Abur5F5OW>`3pJ^p!qK zKKRt9kLGAzOhDCH^RK*ioWhfm^W#uq)ZwntwcHcJ(`vqWW|~AT;wIdaBf1J*GG9P& z4d9S^2r>!1$o&Q*LMIcGXDUf-$FOh9#R|JzOwS35k+7mX0+PKZZ|7*kou9s$j@5H0 z*3v9bG4K3?H0CCda**O-YM^m#7m59HJQ%!ccP#VsEOYs?x%7&6<81gzfr>bvkjNg> z8z?5w5e-$cyX$@B!N7D;TDk9V%%L0a(~3=>MkJJ)#DLr2{6HGR`a7mis1u%rxelMG zS@zb#1aGMnl7CR5?qfbcfAYG{0kf(5Q=LlF6$MXz&G0vkzPog~yJb>hF}I3iYb6yz z9*PX-GTfZcZPB6goKM0HN7pdnjJxwM;+|tCTaWzid-A)+Y-8`>T|PZ;$GxfQIvo6bl< z;LG8hq{`1W=;~D~11k!$v&fj7GbYpko&t8z;2Y|3U!5Ef`V7B=sl***Om-j)4+5XV z9}4jdV?9?sO{})AKDY?@u}`ih$77`yh)3FF^eT544o)gR5O2lN{f(>kOtR^)L-~ps z*Cssj)X`)9rblkgNhB;}K6{#v{Ty+ORO-l2rn(V5GSGX_bS&DRH=Sc6}2_{C^9Ix`+ z-OE_qwlY7knJQ-0!{Nf-ir$YDT2+ zfeM@(tRDktVAabK`hbKVX08!(cWK@?+gyt_)2&34P*A0i7q8BQV9RZx&!M0asXtWL zrQugkEKYNnj%ZlN+|%aeJ0Mn1b~cGAI2TJ4*)B$=G)TEvE*$&BOp37L*zg z4GIGXIob@QIN}l8@NZFLj&}OXVUB(jU7I=Z5FtWQ4Dw?A1`uk zk7hGrXHMsqGE=KKhU<4~P`_e&3nyj{U5DWZl{v|C@^$iRVB*>jxiqs>mJr|Ly0~&( zfFeTjzTjRSYo4zzThhFnT%(>8e);M<-YAUL{P&>mA24%i6j`Xh4f_pseF_aY;yb@NdRF@UQ`0RI$0@(fMutlX$p^)5n=cgB?=|$5Tcx|rY>Zi`>U@|U_p*bddZ6OGYM|Qk-ptLqRg}JpKU{T6$P=)o6Fh-qbfhP6m<12QO%e^rz z-YLvoOr5U@Sn9g}c*9xQ2a;sz;?IcYm94pibhdUb{4QHEm*5H@7=Lw5i4!-+JBCgG z>sW}n^id&cLIa?_TFcD`rrd@WT(!T%BY(ny-mbQWksC5s3@ct6(JhlN$5x>e9XTrB+%LG#N3=GgMS@}9 zF0SUc*$SqpDt{YwrHy~Tc)LikG=-dED4|UE{QBjht-DezXb>D?YEvjtinb19?du)t ztR2Taj(PTGIKa*$#3>&<1Axz&PtN%MHm^Ay;1SzsP*2sUK~ho6NU9ZtVgc^<84GSC zbXCo)X=Lwa?}a~;j$^|#8yAT|QR)zl(WPE$q&Y+b^nv^CN0CjP{Xdtb)iz9c9yg;P zCkXMa&qRzF^PhsM4Q+etu*O*w$vEA!GlD||MTmC&;={lL(nB*_Af-)Melqg*R9O>a zJItQ5TaOtODKT9(y&i&P16K|~D7Pp2V!Jmtg?X^csFO7ltBFJB5(wYCy+)VVfl-H? z(E_%}c-mkPVBj1ef&x69ZC4zOEw^ZC->5&1hy0bQz{|^o#?;YBCU6f;tQv7;6KIn{ z)N+Zq6Zrd(PNh-;4~`&KGDr)a3OrDVI|FlO?De!CTS*M8*6@$(BmABs#w}(#=ZSg4Hi2>;}NT>660XBl3c{^zx_l|4(<0!^uRg?kA%=8@C|0|?<3_dhIX zy@Bi16S$@WQ5_A22Nrx$`RWkP zBhzbL>&7XIjFr@qPwpmA)B5&rfS9|h6Y*u2}0{BpOVRD3wHD79EAzBmOox9{<9Mqig0NkI^ zM70_k(HfFB9vabe4$vTHuVj)xpc2#*=AmPgA{(4qw;j=fJv7Nd?&}KI#{6Z6&Z`QV z1id-mrLBqEDd;e^AcB`|GpdX&H{(I~ct_SU{GaohL7F&mT_zjGU9#>I>m`V-)~@Yg z>MsU{yE!A{fnHSG2LKf~S4lwA^d)NKZ>SLxck27`Lwb->B0Cx_3RtrKjrOnL zMxqV|iS~~-rhtxDq zyA!N4m4I+V6nT3ATDFO?LCccwo94N+%$W=GtnR0x=p~g39!O(IO)ta>O-- zdE`%~CWX{-&HX1+sV3|B6_3nVi=e^Q^Q-jD?iUq9(X)X4>&GSo?Pf2aK8)l;A1zLv)BRwxKy?p6K|-4+h9^r@I=f(T&F+RFHEQb?P0V^_^x zKP<6Sc5XZ>eNn6uI7pxM1guvO>ukKq{>jGM>%n(K|L@?-O)~u*3S9vZbj*Cqs^S#9 zLFn!xuO1qTy3;1#veEXYjgaz~afqYEpWyOKMp1>`dwFymoLgprrH&w25x7Diff$!p z!0L|CrgK~9maO{O;|47*43UHbSYhNFfEAVOWS`SIS2-eDwv%xRJ%ELlmSK`tKy)A5xN4`&dk(hj^D)U)?z>`yi%;9UFY8#tLPmkk=<(`tZR3$H^6j#`(F1;2sym0hI zLF%2!0rf$dbU(WY4`2Os!jr_%=7vsEe_JOhUi{-o(QEI-B{?e|Eal<(08jJlW+XDP z(X#*Z$fu;@Bu<2Yg95>7r!Vm<-`5Vg_I(ylhX-F7J-rj^pB;mq1&5Qw3&NnuQ9y<^pt z3K{?OQ~6RxEw@EolM?Ush)C+2=tH3zCp^!JOKKQ)*|5tPXq`wZWxuMcEw;Bjp7h#s z{PK>OLcinogZ-VvhJNCmW(~chZ&^#?_;LUIbYb=%^XbJ@3dZu%dtB7x8V literal 549650 zcmZUag+o5QD@Y!y1EaqktQ=f`f`CgNnk; zXsbCK#rlOIDv$6xE5$1!WaOW(JBCCK+et+Z8DERve+~~4deFbnU6(o!(j8>O8Nju= zKZSTbEK2yRS8_7H#N%OE<@q<fw3jHC8Uo+GjuaD9iysPlX>-ln%uC^+i8nUA-9%bd1W>0cecf`; z+pi2e!k>FBDQOgE8--U9{LcBZMoa#u6)DB}Y4l!|Ry`7xZVtE+pyW79G;EOWwNDS=SjSK|#ou*HQ5r}{VAA?7Lc1_6vZk~8&4 zAFc0wlc|uR2#g~cmNCuYv9t3B!blxoR}`{02Mb1>E1uO{_ojwQ8Gh+HxCA^W^R&JB z6bKUySmakH)I|t+LfF4HwTYvVcIMcWydHXU#Sf;@P;j-V(xX- z&o3M>p=KxeGKM7;Gqed{9p3Jcf%6XB$6kI}2nuL_A0ypMpo@@ig+heLHlQcFeO1>~ zk7W!|mq998+FDD3fp%1G>f->k3uS!MtR>|a0q&E!KMTKs(r@|(GMm?-b+vXz&;7+Wzl1y8TL{3mE-9! z1=1nqeod$ob4Zrq-agmcob!+52A0PHi!|HjLx1No!8|$|2U57MiW@@%EcneRZXWeu=C^B z^U+^;4=;UX&LmSj=6>NlxiIJK3t-Dj{U#?#zfjOY$Uz;umQmT@vBi({Gq)P4QZG;L zCyUE|ord|~bkY{#-hAoz78tr_EGR8LuBwldpPP?fZyfGN?_k#`e>FVZ%d#m%6)b*= z?Bal){7yP#P=t&aw<_i1xNWHXJ~;(wk2=S`TfIbOEeGiQC~ zgyO@skdrk_^{84nuD4QMFp$=xyL#wW0@BOsDgT4x=CN@6{q;If3r4VA1>>rR!~$Ki zx*iSX<`J~Wrorw%`&TO)6pPPi3n^M`Ce+d2vilq0T{R5e*w}cf2Npi&)-m0YKLy{Y zJ?@kvtN+?vJg37)wOYvg*)14)De!EU$YT6XdE>xo`HwwHa1rv9X5!Zi#}&38@dSVP zvsOm|(UTHY$)JU^N>RJuEDf8tS=i!}Ssk|PWSEj7K1*X7xa)? z9q~j*l?l_js*>#|h^_=?LO;FX1n(`*UjBG6GZ^H-`UEs;Z3oRp9`jG8cl zXqPTTI8fh=E!Y=v0*IDQ1)+{}uB|V=sSNEJAoH=skh@Ht`dwVf`b8hZTSeXNrKY(T z0q@)FknTvb9MR?y>^QtrIn{FA%Q&dyTeE(K%S*n)dr_;X;CU4rx*K%M>I1sN0+(pz zTZ506zoSAD=>GhSN)dXzzs;#rk@cfzoqG=0nh4N=|51tS3Y~lGSGYs`Rfb=}32?9Y z-dKJR0f2f={nf@MdgjlxUun7PD0E>ol0I}JJqh{}Xp^Vg{dtbakr}3<-G89AWWa&4j`q?y^n)EGM??Wca>rl?{cff^eF%Rd4)RHDv2_fHyscZfd82*c(C|H33X@HhilDS3gzR)+S&CIx6qMKceUQ52IHe( zhhCgG4;jlBy&wf%!c02GV!P6mKh2{4`8(7gEQoJsIp_3Ce3wMN;8I!Xi$Nl6GC@Ck zXxs=8QviM`3@=@s+98`qtU~WVFceS3&0=sg@QPbDZ}+eYj#hj<2|*bTN9qa>-wJxz z106FkM7(p2tHK6LucvENazr&DX=vPNuxtBG@7DJIOcOCXYBV%)lg(!@U^dM*+e?(f;W~Qp$B|IYxP#^l8Q>;6T#=lBn|1+$jJrQE9TI{DqS$= z{E!Y@br&S$30FLnLhKOiX3ssUJW_QH4C^2?Zb0pSlWF0K>njq(l_wNB4C_t;xzf7*ej^gFY0l) z6Bs*Pqsqef7=K4&X7(x`9b?i-ebeeD9l^~o{m7pRMAw2)U2!9h>8Bo11mS5PcgaEz z5QrxY3&MHC($rac1Om}BNzem;OxbzB?KZ7#3k|wcLbVS?BVra&I z6vaO{*g-={fyY9?ED=f5KK1uL5fB-F~>O)M&Qtz+diT|c!nWimRz6@DIa z10hn!+(23;O^EFl0Io={2_Xm_p&J73UML|82G?lA5E7*b&-^Dkxl$MP73N=u)^A6+@bD<*0x$IDa~t=P42n zDi57oVw7BXxdjow_Y(~Nj+v3!r9AT}l`542zn*MEWdri(UDMHv0w1Y>yCqj;Q3r@) zc*+3~zT!$lf{g)W?8tuhAc%TV z507gkF$CT`BrqDmRVL{W2Y~iDs6zY*LZk$mkuo9`tph)h-IsFr{72n0yMMJ!1-}DT zz5fgIuYcN@D*>qHU6t(L+&5&LAY~(oz|JwA7_u))x`c+F7AV$wKzUf)TgoYu^nj09 zT`DWkaUcO<38Vl|0`Zj9^%oeB{Qe}DC;a`ss}5Z5=M{znv_mNtAb`-g1O3x}vReXx@;?i6V-w zZAB5#(M_jf=;0p#ygF1mu8Ywh`tu1`im1N}#c8lp2)`BBTOsdm zZo-1CD!5WdsE!VSl*UPVw^1pQ@#6ezo2eq|Ql04!^4BM;O-H+M>_g^;1%am}EB`0! z?MI4IF9g^h2Tr<+1-95~fB9Uv`kCkU(jg2MHqO(&nlCnShlfi-+Gv|V6V{E@9lTnx zY0j8eu93Egir#A-J6CDcv!(fEr)hJ8CvSwLul5}o`ssCpFLRuEx$(o7k`yuAljX%e z#poXCy$^T|F0aTNIJpRe=}=G(&`YnG#Vp)DI9z|V;>3OTuB}oiI%JdC7`8OB96g3t zRWuWOK}gnz>ZVc|8-Y|&G69psvDmtcb_k&R_yhZZoSWnwGD@poLaE7^h4MSI-F1u- z#H9|7oy!T3V-HasuKCI*Ei8V3wrEDmnI%dGwRUPC0Ud7r0w>H~3L*J1&G1C-Q@(oF zQDtUsVQWcc3$qtWDEwwqt@iTGtk0d&JOYQv>RI>)k7EfO!|lg6L{RmG6v`G2r>@s* z+W<2nIPn}v-DeZHs57sV!X_M4mVdRrTKWLLF(A5Y8}Q7V;eJ!-;N*~J@9?zgVy~2( zfMUG*BFIVg$<<8GGYK=db}r>7xFtZA{$r;vhVEqf%R75;qCyk8JVmBba!T~e`9dE0 zHh+oRJs21O2Oedm+Po*51SG%wjfZ<~WH+PP`Dbh@QPQA1_#r;6GU9;7YaGGsxlu@$ zp6t@vlv8iv6*HSy_slrRSvvK>Gcso$O+6JGm|mvEL?u+^Det((AyC z;K1}6g9*}`vnBC2lI9mhn9cz%nbesr%3`2aAjX-m-Nt=E`K}DRkO$PC*1TUX2N2;8 zS6c~wg68n}Qmf+Fg#msOJE(!CbepZC7v8UP@3rT?M^aJM7A(i8Q-}*->bOp6^F`dk~XL zFA~(viSMURH3-P2jCN*Lj>4CLuq>IFJSB?A5n;P;_C~*(?NH_<1Ndm_03_^zbf<(u zk{auD<5L}=t3I;ySp$8-4psP*NbzPm>q4yP+)Yc#bAQPozf5>b!h>;Boq^5x#Mg9> zqZ@<-AhQ*aED1a&Yp9+gv#AGbA1)T=XsjX-YUv2CDgt`vnVVCXEe5@?lB2JJ%bQPt z)=EO1Ac||YMJbb111vu*tk__*chVF1XdQ;U?*l0s-hJL&bqlP{mp+y=c$xj!;%*UK zOJ+v&t%-MCUGXO8=ec_#io7_vb4U1Qi#)9ZBkNMU`vuM4R{%U<8=@#frVER^ob9ZwiUS( z7JRkFU*!o3Tbjgo3VQ$3yjAgAN9_Pnpqd_e*4dGqG2O7wUOUyL4FPb!v6@eL-R=#g zxaybeh>*R(=y}aXt@Fv*NQXr2G^l!(2i5l$IHl;71f~`NSTJ=`$a%@^ZrUsGSYDD&hVqBOPOFy;|kuzoXSJha7usQ z^cn_8B3{=0j9ozm0n;2muPq(aLIT<^c8vdU3;pKfFB&StkXUqU_mE499o+7>$D}l? z`&zJLRD!mR!m8HMd#>VD*x_^Lh5qJtUqaBa7u_4Bt<|^3=QkSldG9mNP^9Rj$7kf| z;;&$+O7U5Pn~UUO2~qO41J-zt2C3ANAB!-zlhyklo*k!n4{1Cl#=CVcl7De(2k?~I z%R(x@QZK>QMvFxIzfJ;a&34G!DC>xuNEy#EQMEiA;QQoe&{Mj@DeD1z{74eo_=c3} z&92~i9Db!+Us4N_H%&-!7nmMOf`vU`5{*YI0_B2-h!LU`$1gme8rdM&OhgQ1cNrhz zHhWoVLmO;tL2{?YBM*#suf9>@Rd&7|NAB(r8P%(9%ma%6r==ktQ&1M~s_#zX_1_sm zl}1c2O40>6T`rn15{C0XgX@rw++AzyZTnviWSkzBSZED!k9p5$eP1XD;A^z&_y00F zkArv6`{werTOs#S2q_A`G$r;9hFW8yKeFIl8wtKvjO{^1|D%mqZ^Va~UTP~a+57EN z208}c1}{-!MYSt&H8)BifZI3=h&wdI$^^9P43QEA2mHBxhXChm;$R(eWAg<|3_afQ z>@LcE1B4gzUIVnp-l!n<&&Xa9z~XTFg62TTb0FC{P$WXaed_HPB3F7YZ>4qrsGyx* z<)~Q2$v=Y>pL4t^fwB6XjFR*uI+>WUU7Mt@n9YO5ky~qGo9`r9p65$!u16o~FcjN? z^BJ|NA&)4COh58#C1SohhZfoS>GB*t`tr)G1%zkowzxV>?R3 zj~c!W%#o&wd%$i3`N;W-J0`wUXp9#T@5-~8OlPNj8+_Wc*ywwXcm|wk=1tERoGwRc ztDc*Zw{gs@hJ|lg5cDr&VCghKM>aH)R_bJ7@H$*h5MjlNIMiYgtycGkZjI)mkiKoq zaz#*ApD!s&!pW*dZ>@7iozqlDb%t6spZP@)2FJj)1eJpk;h*2IVsPb^}Oje!QWcD!=MUVVFG*R&Tjasm; zvk@-_hev#EuZzc@Qn6VwaUqOy(==vkZ;wBs3vCxUSJc(zwYWjy2O;~_w<*#S)YX+x zY>y3IjPi!uth<0b*~16Bo@NB%1sEY@4s)(DQFK<($#1;zf;oqQa|#<0KCV`NDn9#w zVdpG4cmIRJ!T3E&1C3nh%0P8`D-@X2>cpjN>armZD1^1Oi%6T2ekBQm+}?KtXY$N2aKG>25{kv*kU*MoaTPH9f<#0<*1FMe+DGE;QxI>17_cjsH^#fbB5o!plaA=sXLUiVr38FCkd8geZH+ICIB7s-o>F_P>`{sw@} z&^$pQ`TK6`ttRk(gW;meJ3xyAx7%^IY|Sz!1d1e-?px-i~?kF&r-#P zKEIkhI{Xos#jcDSxI@bMdLsavw*PF`lvg(I`T+UHwm|>rEZd)b-i5^G117Xsm&0w(^k#q!}K#gzT$RZj%;diD@d|C%1$e+#Q>C&VP`adnu#xxObcO zQnm7jgkMc06=4g0MW#dCAHcZlgBh+Kif4JIQ@J5_*S&IvMArtvabSy-%l+#Osqf;a zdC8cYLg;1)uMKooCk`+*fxPUh&Bs|90$RBu!audP+C6sC4)+Q+@`pi zp_uo^cKGvVqP^1Mal;XZP!IA*f6R4=dL^Q02c{iXeJAMy`+<5^2AP6#tgkEh5OT^7 z2d0#KxV(|>!>hckFc=%TnNUR#&XHV2)bnFWD~*l=R{fu9PGv2I%7LVbEma1>7pZ6? zbom%arPk!)La#FdcQ_e2n>9BspCuJX@!3asbfwKZ%bgWsIM8ViDy+}nVe+PhyFR)* z=8L^ktMp_hTcNoY@jGwa#rS+LEG1RCm1UJb;a9FEUAt6v&!>x~r54t7Mg&gMP)-wwmIn66Zt3x0`{X02S|Re~H19SKY5Ns%JHy%FK1z_;7Y@ zMfsVVx-i&Df2Vg9bnw8VEmekA>;e&|_5G`F)p3KbTW6ZE49^__x*QZare}% z@ciAVS%K^64Dv@YdrSrgKpRaU)du_9+G4cmKN>Xscu*urqJaI<2xy)ltvA)Gbg`X& zPa9FAOS(xP1j-!Nt@Ou*uw@ZDKcnmbT@u_xAD@BlI|v}Ex_S>n9s6lBy>1RqN9~}3 zkbF&deyXFuxyKwhrg$6JFn;XreHQ;b#jxJICtoa7kJe=tB&&&Sq~QvAG}pzeKh}9@ zR&F#P_RDTI+BNSoSx9 zZ4DAGEk&62_kw6#%Z*^Z+a>VN92`b8kAYxJVq3v>PBOQ?oh!@!^$(Nu`c0q)(|(Fk z_FG~3r)wGJrt`(3J1h7vEgCVwwM5fex@BB>km;S_6LZO>i*d~mVh zs-fU!9d*gri5Uus(dDutnY2E7DjL=(qpPo62aik-VO4Om`;Vk4%c%UReo&XRR$jDz zvYdCqZHeS&!L*SYDA&jg2YFEOTol;4*z=K*G6#X%oIa!D*u)IDo|pI48fTM~1a}xD z`{^IqmXW3j&g&Y}8b^qvg~9cGB)*LOpe}-{$#X0vG3}J~B2%1hZ&%TdTV6FznQ@`v zE3?izQ+-?r%RZkv$50&Qt*Y#J_EqNigibjp%I+Y&@EfbBGH=%t`qw{50hwVWeJc-4 z5=)2xr{L!oNv+H?Rl*5@+wSx;xpmJNQ_<6)0f|IGb=UMO$9B?6NF6U$ETE8_L&z=s+Nx@y^WAi)wRZNXiI%Wz zl3lIfY=BGoffewrY#_d0>*QLl+FIxd%IWU_iP-9$B};L@4;^V-tS;lq66p@*-E>6PE7ZALvPgOMv9R$u!>gDN{qg7o=vqY_7i zh{5eDBWIH*+9!b9(;vhPrb_xv(#;KD=&eiD+LSgLtGHuZTnjUD>_X9rT~nF#gnYPd zRqxxORZ|yTa{=il043c1@$Z=-vs49_?rE~nSd%X92vMn#pqbe~@^2d(<1itMj0LWp zK_+l@eWi+h*0K?Y>EtZf*^Q*^o+*c9tF-?ls>wbp&DiR!p-iljRLOF92exp?b^i5w zf=1;m@UiC0L$?Rk2s^7Jg+Tj?G?C*EY!mmE^CsSp8>{7dH0%-aWgQ(w zDEn(Fk&^F@Wx^k;U%05LY}v+5Uu4lB3n1R#NC`au-Uda`Wy0;RXHUNVaK2SNx)!AwTV7%->>i1J zA0N|gGm5`G*#yEc@Xi?+lj*%qIz{n-4@dW#R3YO`Bz>_R_v9U(lNA7V2yW?elX)TB z%XtEuE`ETGSRHmLjXAQT!rf(b2?F`)KdUwV_#7(1nC^7XM>qWElv9cIyoZ2(nN9d8 zz(Vxb0j-H_i_*_$U?$=hgf0^{bE(lwpoY+aP~53EmMqkG2C+JKxYQ=z2ov8P!e6fD zA@|33S0RI1WV?I(KVIZmYf>AU{vId6H5_M2AXj&^ytc1WQQ7z^ja0Tea<~e6*_#>` zUFxOTbL=?Dkr;!;-LxP{ncRPy3eE=77$Yli>({PxO2tu7VX9HQ8BEk^XynjK2&3(? zgM$vF?BS6%T>{YAOSyR|(_o|Z#-8%E085Kk-`8N5TwKwZ)~UbReWkovEBTp(^~RRx z)5@3X6l3$H4V?V4TbT<%lHAvk_V^lby~oil^1?aiiesQFp7n}EY|11F{V;?V0jJj= zGu*i!HrQtTPVNG4{~r5wS>g>neOei>Y2B4q-@VoCqsqU|%Wpse2)Vky;Fmn+gZ}{^ zZn7V6D(^V^s`#!e>w&8fhs_yJSWxs@I!Ci`ZkgnxCf8>^mrEnQgN}#u*sbaihU4!& z3j+LOG2=A5w5%cSoKj5&xBc}qAoe!Wv174J(sv$d@P>pFr?0Z z8x`Kp@Wp5lQ}gzwtiXN*Z;UC3eD$+C)pvb%gw|znEQKD@V0hh|8XYF-^*(17Vx{z= zbG2`MJtzFe;13rT!?J$0gC}*KaJgNA`v07X27u^)r+^>{^OUb`RVqT7ttXLd#$0l7 zSjP6w)D*EJ^Iy`*&mym1q#lzPp>o8R3V`^)5e^lT#y9wr5+K{oK@A%Gem*UO$0{-| zP4CNxnYfn0nx&ni@h_2ZgMk{!>$k-zDpIXPzovPu!bqwVx70hmsB{J1$CrOrV)UGyw{pFqt+H|VWA;}k zU?h+rZSmwh1jgs1{w}w71YgeME%^_Ib= zf1tAc0w4!Kc_u#0tMD^5TaWW3vy{iX$0|e6MfPb2MWeOuHa~LE(|_cw9zfpDYze)it3g2PauY-TE*W)Oc=#X)v{dO7#e_QBt_ee>L=;`wQvPRzgWRQvB8W!je?*uLYBh^6 zM4Oo@3Kx<+SYPLGRQOTztJ}cSr=M=92{&*hixqP|f{r;&!@0-uz79-jSt*P4mi@-| z?&WvxMPXWBX2nr}9TYSGvWKP&Y}!F12k4Hj(kcI6VK4qIb|q} zSMwr2GH>s#KE_BM2*=yUSCoHYuw4Fqi?Vun8RX_;`itjak#w9x4X!-eWe35DL(D$% z#{p*^{~-nP*OC2iQpOQer|A(u(SMWj_qx$f{!y1Q)0nT{>x5e;RDVeSCE{i)f^p16 zdDBC|50g*f4HVYpC%sW9hhwd~V+VP|iTJtCEdbZ}uhsc0gZ=+4gS!8bffo(w&AS|k z(1#d{{kLUA_7gUm_u?RXg-vXZw`7#GIMD-qAm&=PX>5PE$|#*Z3?Tp0mw0Un2BwI1kB*Y697a}aQV!+mutTx-y+HSNSY;X z;-oRUX%FG%1pe9w)I-Au5cbfF38@lj+`zwWV*9Ju1B5+1GG<0D)Dqg?1B zJRv3EX9Jx>H}B|Naneb<6M^A{Sujie6ACA=Xuqo-nv@2dz9wTs6y5x9^7{W5d7gjW z&cBV?MvEMAtg@?HU+B02tY6NlN6a*eZK9pr)tK5l7te)ZBvNg=dx|SlJJ#L{AD94Q z3*IO1KM~~0n%;3UjnH$Gi&{O2sMxaT_&1iD3~5JZRK!tqK9^M=JtK*o9a6o-Db?1v zX52d5k6d)wpN`DGASu?<3Tsek^7Zz62s=F<`<5?F^=;9}%A`*{qTY3n`(!vq)6VYt z{LF{xs0Ajm{pk_aJENkllKT-4Tjz)ME)HJxPv}7z2g&<9r?+RHfSxa(INrBM^Rk#r zBNVzz4d4g@ym7=rJ4`V_twRbUQ=E=US-J6rUz@<@*#F>**M0OzqpCXb6=sDulpqw+EhfORkSN@=Kun|iqAl_? zP*R4d2ar7tgS&(K^8&>vo69v|bg?h97EP>6DnrgWFk+8uk*nOct&83JnVnQ|~IQPNR%Lfq$M447Dh~SFWBXI{h;ct+47p+iC9H zL#Cl-g~qta3>8)S(f5VQ=q^()w)$Os93pfN+@kuabm|&TJ{Qgre4y2ilWFOZii^>j zr|=(624x^i@-!)il-f%B}nxE|Wc|ajxqK>gI z!d`lSTi+nMfdBnOwk!w3wg&aKNkd-DDonqPnB`ftIMzUxH2;(N9cKbzIuyVB(1p`j zY!qj&p-+tWSy=Wa{Ukfwr%Kw-z%Xf(Mct0nTEr=}1XFjmsgn47BH?CL~o}*$QX%d zXuf$=GP3LbY?qRI=0{}^66YG@pXDft3}JcpTNdu|@PA%PPxizveQ0qPg3zLmXbYYn zDT!*P)!yI5@0R6n&mzaY*!ftz13js7#FrLo0Jm0qO4Q>Kflg?nBNtHLL!59>ICxp! zW4MEQW*G0T7)GrP$d7Yx6+~X($kC24N_zca^Y$rr#*wM*w+OzN!YXH%D*fR1L=`K` zh?VVipNL?3(nOs`eXAjPBVrpTdv@1IGOCkRGe8r_Wh@}uh&=d^fRw6P-flaYb|j)$ ze>=iAL*mHFf&YW~P%<=`yY{Md_|AaDQ|8X88gqf)k>egM67Km24`6L^qP2uOrtX`6{WZ;BU#^l z^F?Yn@AGC_M@Z(w!(yDXKLN=_1ofKOhkWn#dTp<nVioD^l3!e*$%j_Wi#8);u?< zkn%vF+iXaUowhNeIB;`ocR5;zQEbv?6TO33$9Jzs*A3l|5V;y!e%^>S9I-SU zyD81FDV=%(S>GUliC+*BUGOT~VHq$5G&1!X=)dT~MaW-57yfyJ^L=TCVdCZy!z2^S zx4X6rS|h9Yk7sk@WtO{)OmAmt|ati}B=R~Fzl@TYg@cgb8<%v*b3=quYz z1(=quTY+=rhBo|M_`>v$#ULG4*LUl@t;tW;R&kuL_WTzacyCiZ;LOc!t99WyojJAp z()j5gID~L80fG*fMU(of%Jq-G$>gLeWn?$bTj_9^DL|^*pFea9spVFh=?fnv*K>VFKes?o!L-no|4_pX!o?*V~Yx*01iAbnX55BVa20Q8C#D zw~OGyum|O@y2pIiFnolcY4*EP&ibM=)_RKUU=T%FT*F<;b7>kGy`i-B?k`8i#`~tA zH(XG>4xw!^pKleB+i(0+xT##s5kglgmLo;SH{U)ftB0$*ga!xsjVmR5O1fqHb7Ee> zRS#Ir`R?RJhrSd(ytIHxFxZf4byH=r++((BL;d9*W9ECvn1?~V&3m9X1B}49lN*{E+VXr8sx(<@tlP4syr66yxXtea2hFcnsRtI6ZmcT z1DLpS-n=44bv=wqwOaq}73?=pQ1jsPd%}J{FdH^IbFU6|gk zVOXKGjqT~Meg4#DH$6F`xW-6;GnoM!@=E-CFgCM@;Uf!=sZb4aL zs&X<+Z6J+33+8$c*VT|ZpGF>!xfSPpYW`djy&O`^>)F>y807tncDU!`!bn13jr)V$ zWNxOurw9%q!wp0tuG)OHPAvLCD|)OhlPI7)^7v*g?pw5evp2HJ^Dyu8g4EY&oA^RY ztUJGl)l{C#vqB&VOdRn#lwWO@<`!0P)_Yb|Wr~iiy{tvt^b^gVHoQKQWZ*VMXLxcIw_ubF+7> z@eZc*3>)!#a@to#>;zsS$5TKhKR(OgGwU^EpvLs=;&Lw@d%@k8nm^l7dK=81C0851 zDcT-ki!ej`paB?Z2n$^!0CGGv?q0H|pUrBWOu2f>PoJ$BmUpGM`V{%T(u$&s@Mcz! zQTkI_lK+OP`{K*<@L90M(xc9Pl!q&ukm#Q6np8Ko+A3q2vn#gMEI7!iNK~8r$z;Fl zexqT33brPycyhu^-CWRET>gn3w{m^gF#kM+c)Kp`NuzUSzjQSxct#C@H#gUw#MwF* z8r0RtDaq_czt1MwFb~eBTU>i)sY_g%{ByG+zLVUXK9#)k%vlY`W^R;R`DIi~2nGa| zc6w|72sY`pLnm%BFWp46pug*^kVLWTvu?l^vvyb_(%Ji zk?^U%A6lBhzI_gIE$L@4+nB0q|IikzL227yve5D)qkE37nFIR9QZjb};=%6axK(ba zgsjm@T7TA9_1)lSjOhn#$snehsfXJN1h=btwlmduX(@{|*s=iWkUrq)Ze;P5C5}tS zS0uEes-_2pnZt2|L#LPfYcGPDCVT|1)cJSTKYTdILb7&R;_2;t4P$H^SN0?y^_KJR z*YDUArORhg4~W#UB1?>U2WUv{Ab*NY7XZTjpMfJ9o1mc7*ulY?lp%MQ5Xil-{nK4w zqg2JR8RXUQgY6oDn-Xjmf*YiKC}z4`j5O}RNU=XiI2GFs&o09s7ppwiT#00O7`uAw)^=b7iBlPX4n(EX%4;+lPZb%nOvGC*iP=5V6V$*?^H0DOl(#m?Qg~`dgWYaB z`-k8_hHCH*zt1tB$J5`lt+4naDe>pQXZyCGU zaKL?nkiyUR65xZ+v2bQjZo4`hp$U1P5*yF3@YZlz!FHX-JUUP5rdWFg@``a6_c}XCAxD0sq1X8 z@&iU9q$E)Pg$m*K`1e(+lF46*gV74le5k7(4}IT|F(I+>H$o&UbOsL+J+i~on)2HA z#wBw`^W{kc!Z4^8H@dfa1YL+?oAWOo^z!M(=y(GlSqU{=eMy#(+uv$Q^b20N?azj+ zSg=^4#jZ-RcN&$g-^4p>f0+$<)Vyx>8F=>w@)RxH?YJZCrSo=ciUh}6(F&_*XH}ap_;En02$MuGMQx0{0k||E9cK3yb zoOjV{P6}|iVbN^91CuS&9rk`b3z!*_#Aj4Av0GkRd0xe@eR8KFu}wZ6Toh1W8WCvY z*_-44{z(WBOPo~z_#}CzRI5GMpmSEmqs;-{sP-4p5(G)f9EGecT!>8-b zs3TeKDx;Wd07Ico@%a9Hul`)t`F!v%R|FRgwQ691T|CgGk;R1-eGg-j}~1fbp2Fg!u#Novmm9~h5!^Llb@CD96Hy{5+P zObG~YB`ye#E}sD6hhnX*UO(e-*M^JP*K`VR1cp}Bn)$(OFOC-pI{md$QByF!jzI&D zp)VL<4Q#zBrg;zD@wN$Gtu;@J#u8@-sskQ6cETTNvkxDVs6Aw1a2b+;AQxcsr?s>w zwJh230N*2jxop)}uuDZu%4^e|tcLW7Sz{H|onQBh51L&bHrZw7XvDB>Fm{?G-$ipI zn2#&bs!Jv^w2u;H7UTBpH$~=XgC5WIR75?URj6YQ#b-64T~pPeAVvVG&iZG+(RM!% z$KggL;V$XguPjP4NjV0|40LVnaw(kvPTj{S#cq>M(uaI5@Zh@%F4GT&lzG^uyzca) zMS~W;+aqyw*bIqUBJ%k>uDMl3XmlI2v$mn3=__RHO?;;jwfowOU>ID}-C2BFBZ9@V z2>X6r`BkqEMJbs&ZzK=+_)wQ+&kW^bhPQ!R(FVFY~%wbd~UFR70Cu{ndEPFsp$Mzu$n< zK!+18^!d=68n}bPH+)jsSFlQPpFJhq;`(=9q*0!CN<2nZvXB#61)Z1lY)xBdy-ATZ zN(o6`?3_MGXC-yH-wcjdA#i{sG%E~OZ%Q|(|rdXaqw>MAohi~fx`SCeg zS&;nrm#P_j-^*XAJ6&JYn+jhcwfc?+aYGOpyJW)STavx_Y&863+kbBOFEveRYT(h--($+hN-(orQ zORodNMQplrUuhZpCx%;a&9U`SE%yaQ;HN#=$8CHdmXj$Wu{!qx1HA)&=M+{+Xau@g zrTZj;g0d=)tFb};g;-&bv-yD<7esD5dqN&5U(tYZ>e-T?132<+N#wodRe2{web$XW ztJL+-&wHY`yi??FqzIhYmw6@$j;)dLsg57HyB(iW`fIa^HltIq)r+UjTj#e4ahh^@ z(+3|Jm^1FWqhrK7J8jthEltUbc8x9Xw!~HicGI8ur*0Z1FMcl49&rau3F`Wytcqgh zv-7Qe?y0>qwn-86#Kg3ZR8*7GaQtd@L@0cO;|OQiLA>(M`M#sKQnwwIYkl*)WnX;f zGiv0sj05(_t%5n!;GF<*Tpx4m(V+m7n8GMyEA4^jxuLYWORHrJFhSsA4X~lt)zreH z{9?V(%W72ks8bROn|-%Vo1gf`12#fXRpYt3F@6#8^RoflcKPSQ&%cN(aUXB>aG%J& zF#Cd%jlhBjZuPKS#xjmpK>z&x;j)`dw4(q&H;Ck*yISPqRD*0GCCDi%O`m|j< z!maiysiB}|Iv`D1U=e*}Vok+5!_&kf^ypyNR1LSI3HlcKJ}@}@H}xVhjWR?da5iBy z3qHMAU*JazAGV@~*KXL~ehw9eJZSL}a7^$a;DN)EiW_d9tL|rE7PwL`-+;I#q3Be6 zMG9IAPCtZQH+zNjCzQK^--~3JhK)bQe8d2wY&S(prMBWbANF*e1?$Yx+TqEmWhUOQo z?gY!Um3l`iVn}e)zxpW|oG_PhBQ+vM7?74+-8~upyarl-$G@IS{xC+R_2ZmkeT_2i zyCap4$I1Kcp|T$OG=(aPr)D#In#Sl5VB>ngQg$)=r*IvFqb!J`o%J4C5m{uEU!-Tw zJxUl%NYCh_ER%qvNR$v1b;SFdZqe7RtqYtG0TFj`?n?u=*isDs{pI1Lb_0vVg3?>a zC-v}r@#s`8%$jEMx2vkG7;tt}jEN`_=x=VO0YwaOl~ zKJn$@0RbC9EMv?BwUF~nXRo^~S3vR;MG?-yGP^otq-svNaquD*8pz*d(i`D&Tls$g zPe8E07vBISIVKL9h|eK9aGOw>Z5;I7{WQCAc#e;sHTh;P=!PRiWovicL72JayRb}e zCw;y^n_urQ-BzMsx1+B6q|b-dNRfNFZO=+?EPHo&Rz~?YLw=S*PMw<%;CILJweoz0 z(B#D)+Ev02{o2h&#aat7?K-Orwzeg`&6Q*e(WtuYuz+2^8p!yAhk~bsLL+@IFms8i&yZ6`neDc{2R2njaZFn1V%_EUUv@b>gdn%cW+z!?s5PF9Zs0S6ZuV!B;dMqtmNc1H z$JQT#RujvF$%}+Jp14YM+p90wTp*{v@thkr{G=F3mnK0U+ep>PnN`r!^67%rlf%#+ zyddT3XCgFrOI~1K#Ddc-Gx5@*Mli79^}5urFX(!jE!^k1*}4qJO5PIJe6nQlOT~Mi z7XUpUBoUSot%d|Ap>R**K~sJgY)0)D+b!qvWbHhUU8gq@S-D>pYBeo^3DtsxwnL0{ zuX;l1Dx17RB!Y2#2HUE2jOVkpQ%LUCO`ynZ90;{dW#CxERfIFo6M9*7g|lvGVB=tG zLom}G`u=i_>(5twyb9x%g>I%#>2*lRiMqOjc|#A0z){q)RwapHZ>yufyWCti&n_BE zoxG2DGve1h;_ll|JhY+raO+9k`yM#%wB7bd^1tzBz4QI9mJ+lvk%e`UPM?ba)W7hK z6tiN+hOp~rjy{)#LZ@>p#p@V+L)Gk5Py72UB%fH;QW`|}kC1-3gCj9?4@18PKi_G0 zx`XwNw@$lXl14J|;jJ3o9vqi$0e-UypiGaZdFL@>f!>K=-I&CA_fz+HtIebY1?v05 zf0*rLa%49iXHmse$V zK?AWQ!*0(~oY|q*f<=Eq_RZR_g*K!3Pn9vN=G`|XG-m1K&v%TmP<4>hO*8B{ zx2OGX;k>v5-7|Vq?L~}wM7z{9+r=>Yx+gXXnVh~x*pAIcyyFKKo`L4H1OY+!yD;GL z_;j(SJ&e7YSfaO&gp{gxw_CHdk-To<;+0p8Aiu8eOmAk@E^~3k=lqzt}mt42r$Bu2)yzim6wf#cj;!byVos&Gu7oZ7OsntmfeqTgLE1 z(n{aN-xz#^!!V};+Se%_ zjClD5DRH*D)A#XX{kEzpaW)SmCu{FzAFX8InlE9+ov#`6^T*wi9bU1`;8gr&vE5=P zpif$dla0S@KEFKT|8nQ$9vleUH)i%Y3W|Pyx|?QU0Nu7p)pGecSpUE;QmiQjKYtvp zPu**VF?$T3-~Q1BwN4?Yl*A4|V_|{0paZ=Q3VmyZH=p!0$3`JjQE%>R&@kfft8=;* z8z=5RnRsS`zYlA&-II#I4l(h<{>}=B{9dn;9!uvxkYDjzkNBQZe@=g%_rB*K!_G<@ zyt?F;oxoP!M6Q!3^6XksouuQSKHS@eU0B(%@=nlD6{ieJK6_>I@9l-ID|vr@5`4W8(E;O!qLb6&sp>_ zapcstV)(;uxhzv2t`4}=_Kg2SF(YsPU6R{=UGo>@8;z_G>s|&e)3HL*;a8x$`N6SW z9dy2hE#^C1mt{mEd2>!^XYM1cVb(rV^S%H_)2{or?w^M{hW$l`z4Pyu#y0C0b_?~m z=Iu{0-gS4^ zm0)&{eQ4DZq`bkPb&*@4H)eEkWBB5E^ETONO?{obG;ezaE0bQrrGHyL^?gyUkamqe z4(T8XbNe&ucs)I1VuU^cY_{cedulST9!Vm&x?FgF(u8sRHm0{Xw9QC_y{SoMPhE&C z5?gNut}iDM9NdP+)_j_0XI*YP$9px%^W;o#bGzlizW|1>eto*(@hCRFkuw@~(m}GE z%Cggf^-#ldDE&b*qmGe%q+@Js_ZqCRl0UVaZ3>D88uG^~<&mtox%^{E2~eK2f9YHh z3*;M~SJSr2Fyy5?){3^2#gRj&y15dsVe|R>Qhl8S=v{Vo>%$jTFgeAhWE{bS6luew zdbl5RckG)uHQd0s|J8!xYq`U*c6aWs`mghNK9Y}LEX_T%2OF4UzLUg;q1M_qy|$nM zi#mT8?tR)iPapN~8jUx^EQPA3O)M5it>>N(^I~#W31lpLJG�E;qKd?V3*HZG`T} zuOIH^vxW9Myx;jZheBqwDb6Nb!4jJ_6;t(#Fgs50$x4Br~6>i09d?Xdnzj>ik>cpOeyQuzLz5(GDhNG!T&2x-C}D32L+ zVw_#a^7YmG|HV3vNR3QB$79S=UtKw5)Qf`$z<1)PIP~9i`!RHU49Uz(-<&Y{KJUB% zrv)b*c1&Qc>wB-fQzAGn|LV+G)eGp!6_U@l`~}91KX%(Hk#Y8lXXZvBMty>u@wH<0 zp&saSlsFi=x*qEUOd^R<^m-{abDS+Xw1i&A#S#9q8Ju~V*et!f)hB=OKjom~gGp{h z>q19HKNUG`|ITAw?{T6{Z9VfnSt4_~gQTWJPHw$Od*>(&&z zZ}awddVKih^W6;kV_69ldB!?*)5xmhug2*8FIZoeRuxsl^iT59=Pf1)az_O+{6b-L z`E3u+pZ|D#-_MY@jg!lfa-^*2oEQrxrs={9D6;9miAjcrMM@4C z50=BgqH=e#Y%+=92oCA;xk!F7We;1Mfh?S+h}wS`yomXyDgVy|q-1zXA79XkEn8YY z5j>cYocpVj|J+09<2sfwu~-B8Qr#bb{v^0;{J!IPI=$WtD0jOx0k;&$}O(t}@ndMnslx0^Xr{_c^eBI)>oei5-~@3VQ{#v1PZ< zQzH*Xy`_n~dqK5l9Eos}Q$2;P0yexaA=HYVo${QET+B;arpZ(SF%m(ZurUnH? z(EDK^=@rkFYUXp;d?`=2e?5I&aE@23+a2P^=*Kv@ZRC=^j1Aguvb_K1QzEoFALMcf zu4C}3566B87Y6*}b1T%@Ir6Sw1p1n=F_}{gitq5$PpI`^XHc&36$LXSYjhN^Q>0%X za*CU>gY-_qMAQ+{A*(M8eZ3AJRu`eKuR-aw+sede4LG2|Z!%Ls@Bf3r#+QBWp40Q> zAlrRiPts6^iBtC5y)x+dgj$*LgZg>PNQ6BmlLy@Ativx1pdn5?=#-PEs-9|Wt?q{)1>}OP<_$1b;G+2C-;=x%*eqUS5 zNz5fmJzTHLg|i*MJ0{swU~JE^e95Y(B!cR_bzuz>IR2$y>-;5tY+%A>r2U%fyT)`qq z)kiO-EpTZ4-7P#d^ZKTsE!b%M`i%x`x;bufWZ7OQ;=Rg!qo9l7XL#Gb3B*2T*rSC+ z$fjnxpTXD{Lq$3^i($e*MZJEC!7n{hhIU@ zOzacN${x&Ig%RtBb3BUa<-^2Z@$~B#khE2^9NM6=FZ1&AM=db?r`v=jh+db0nRm+8 zlTF!3;bqSV(WB!J+gEJ7R_Y>(zspydbgZ3V$g{v|#gbYXB;R6<2v<;LJeTip_S_nq z-;Y*$Ma?4S8l&H?FK{UNYtRoIK7YhVIl~&uR9~!(7~Y6v`KpOy53ONLvyXXgAHzRS zd8|(jkUbB5%{$kg1ie9C35%?& zv0SW#SM`PyHhim1>)06seFn=+v@aN7)0$^~`AlcAv8h<&i12*-YP66#SEG0l`cJaj z{^rre#?vL8gNxKKH&g%A;CXH6J0dk8XE?7f3!5FyHb~$|;GXw@(9!9Cj7!+OZ(Lyu$rkG_XH1R^euwTBqjaBB+y3$R`Ym7cb})-D z;*;m0T(({<=vjIk4(HIH0~R@PBrIEW4#!WR_tf@e{3)SR)qLYV)+-rwZ}ax0pWmx^ zC4yy)Uf)5hz2$G@6c_xXU#$tfkRPJWu=Bpn{;zCgxS;W3N9rT5;(wHz>=F8#TrP-B zcJV{+q}-ryayUT5m|jOk%6V&rcw3ttGqh~0ucL6;bH+WJ&sp@owSeKtJmpWUcEum6ML zr`#64YAjujwP-(bI(8R{Y0+`dvD5uXwmR>=Hq7TA&zn7SCGW#PHvKxa&V$J&0SVX` z-*uM!JrYWY)oc63>3!ljo0E8~%KcQb-%?w>vdw&W~*_%UL)x7mF^x@ z6#2*J(A+J1T_oKA;|?EC3y9_4}45Hm_=R`j(OcjVuRW|9Inz zjhjxW9waP*O1+i0SuU0kS(ZC@cHicKkf6-&off_@5G%Pi&YgnauDMG+y?7scxr>Qh z%MPHQp}>;IMYYi5CI7CH%@zl4L@fQ3NUtN22#+{o_c*`Crk!ccBR@JA`c~2`ms~al zeU$~8r3bb_gOg|1qNNVdHM=jCubHvFQFJWMeM-oLF-|Rg`LuHtmT|{OWUd^ZcmK2U zr->B{_5V>n4U>Vxu-FhAgDqU9m7{QWchnAtXhvULJ6}(j?IT9NaVc}+_L{ZyJ{4@r z&kDFIqK$36R^@|x-@uUj&E)tjM*l){^@r#S!w0Y>X36S%RV3&yS?qGnQVW}te(Lb$ zaAEUk?!ldfh<-nGB`bpOVY7?L&IJ$-1t-_)7(JcEqHU`!_pkTB*s?>`T;0w%FqTkz za2x$OLeUAIN4WyI(68+5cysF+><@T2C2}GZ{>TVVi(8A62s;LptwBtsoAS~=wJMX>l^|Vt~ zOb6(859H1JMVDM%hs~^}pGqyb|IPp0UcYbATIgjx?l)xzHSPZrpqniEGzS|9zq-oT%ehE5&G`0 z`Z6sajvYB#hs-#Xar%&B(~qzRSawj{S1KzX`nMd_b-Uoskhiq8`?tO_Hnkm%m)XR~ zJ3FctT-4x8z{sZhOKve1P^5fh&GuFOP!pjgEwZ#411ZL57mPIH?7@!ZSr*YS#xA~G zr27f=91!&k2)#vTHBsi63_S0z9ZO_M!VdIu@?JJ!9vQH_0> zpjhYS3OW2Y{RK)SkYZ`k-FeUtJ5t9!Ee*VZWk0sQPq044@C*LT83GF1p!BSXXv8}T zHYUg&+Z)M_O+vL}I>uL^uR->@&F!-oI+lO5pkNqFrUcJwo~}aFt?cu+y9O;`s_|*Q z`;fBP^f&8PMjqVWynn%Z=X2P!y^vD!By8SuA@}9&eZzH&VXrqfTHF{eLh><_5sD={ z!*Aq~RUF)xLyv#sJ@-E!q0`sPlf#d(PFNq3-E;dWFy`^lX5x8t9mz{mmN{uM)@L%E zwyddBCXomWY`Y@e4?*kVbhnf}q1co*dd}2_87airV`AI(VSQD+(*27=kRO_h@a<^>52xN?f+TeG-Qs=3%tVbJ`aIfqYd8` zOBnpO;wsaz7pv|g#V!8s^|=(c)!Xa5iD)d5MQTHby=xdWSWi^^jQ#?>=UwbQ((WR; z_7a)zaMl_9+yvzVb(y$pI8Rs7rD$kr z_!j&45HE@F!Eb{w`4m=qm71njeuoEkPvher)z0%9l;A5#Qi7>SaoR~vdV!Dwvf9;fek>v^R+nT>o+np#&eKE4dj#sG z>Lobm^2dQ&FSacUr{e@?>zVVvZS!H^k*BQckgpd~*4X8pk}!pyMU?_m?2NwI=_`hv z0~$#%U_aCGXfb{L8#=Xu`urxHVW|81RAbN}^h%oLOC4wAIb_#GOy|;6kisdRe=Uga z*GYuefffPEo1pK8YTV8lMx1O;jsKXc`<_0Y)VT7qIFixt(f;gn*M&kxzB|~p(=L0* zJ**{nT$L+23{#?8zGctlB@tp0Q9Zn7SntWV>c?6Ys4_d`^OVToRgKKK$2_l&kqFCK z>lI}hkW#WOw(KY?G;9>D6>4Sh?wMS#msbsYkaEW+HM{2&P7DimrJoI8=q;VJUj0)g zcsDHfW?*71GyDq261y`_ejk zzw{ix4>9G-VZ7(nuk$DX_-WIbB~Pq)13ep~A7njXtm7p%8j7qi z5r6?(zvUOb=EZTw^EJ;UXm!)`C1{lT;OClT1)&3p5^v9*XUHY9(4-@hUO&Pv?auL} zOC3x!oS&n?j1-zze*AQ35_m)_S)t%d&V-R{OYdtuQJup)3o;1d@Ei%oaL2~I+=57&i)4Zmk2RB+AkIfSJU|5A%HM!FVQpg(x5}woRDn!aM!U+6=w!*e<9} z?-z$Ylb;#pkI!S5O?Xc8%9RX!jF~SHx}*lVKVNP({p1Wiv8`FQGn~+)DRHOMn(>}* z+HmdzzNkO*^q&{(F>BCH=S9#^d2#967m{Eb-Cy#9Ay50IjfVTa z!r;jAWvnLi*Rh%oEE+uPkP73b68xLCMKZQaYpzJ_nJ-6Uc~JLcO(c>{2J+1F?_kG! zV@vC-B}ArMr%x9>zJ--{&)c3duY;kQ;Kf2$Y++!l#V%(j#(N~gnfmo-?3iIlzF^Bm zhcw3ZLWDnTPT2~57boku#&V!%EGE!iZxwW!nx4CDEXUvl7F9==f6O2e__v?GmG%T1 zmOJFdm%gs=CT7pY_UWj*)S`}F({dehB4 zVq`x-SD-(C0DBMAwSG?z=%|A}z8znym+L}rc=j0y-(Og6({S?O?)8w|v9r6c;T}>H zKc238nhC{0H!7A$*+8G?H!;yG^ZPB!1;1-%KCXnGi~TcWFPCA{Cso^XKb4UZpnNV| zju)oCF6mP8X@^b=v&1k}PpEtBwdj}n2y`DvHm;Cb4&9esS95sD!(gRcRSt0-bf~_H zRdJJN_(z}U9OcK=NN%#RQ7~8yZMr){9qntO>$H&pZ`&g1I}sno_pBLOpKegCh_{DH z``34B*Sv*U9fk2|!`;y7pQSCBXa`jz8cr@RnX!>6?T@(KH4@?1n$cJ4B1D#QIldtQ zZ|Ls!u^8ZB^f~os9~&9fa=^g-_HNg{$3TY$Q$uWx9ya{w;GZttjpVH-j%yX|!xo{< z+4`OpF!1^1g^`Db*klm>XuDfDPJbI6|1_7Uc3!edY&ld1UB}qZ47z@q!_)g`OLO<) z2yfHMftEW+*5qE3eM|znKkZ%iwQ~Ua>)mgQE}B=D$KntNhnN6U=xy7{Z(pv1%`a|B zSxFaS%a)yzr-)IE;|?C?yK(Fu^xgTzcV&4j12=hroA<`8gT4Uu7QfvrIBOFu-toD{k>htH2Kr!UJRU@x-12XKsal zY*;kKUqGDa-=Ozv;P*g(A8bB*p(Vsv;UDwtj{S1q&(zR)40vZ7O{vZ4v2I(*EzPz9 zy1Wer74-!VyP?m!dCi{1ujle<)c}WEd{A1V(de6;Lu8RVax!AkoW1n)e7pF8v2NEd zKe{McCk6UMMOCCZ^Z!YX`T3asSJN*;Pv^;7eS2Y#UAQDeuFh4T)r;wMGUzk5TYhU} z0XDB&e6Mx*6Amuxy!D#ffkZge^C3fp2jfSy2Q&9gK+nhxn{&3(3_07xMTsYfB!VzM zpW5;Z(0SxFIk4^zQrdZ)&W7_r@4%*AvG+GY-i8Ym%u)(-@!<5wq$5^HA^cgNyFM9y zZC>aX<+Tg)8w=Z)4A1LRgL;j(`hBj$G>^mF|)A^)fk*i;krw)Wc}B(D)oeo?pvDa*|t+>V<69f2Zmj^q8Y zbNRhNbz)rq6XJ^(&$8{?6{tV$ZNqfai@y+644Ps|~T=4uuBJ_U($-cQekbdzsGkSrGa# zrr2W49R?E)3+e3X#H(l8ANRMdou{u^9#@r zca)H=OMuS1u|CANqR=6}`*N!^8~yzL-SW$(l0}fbTbCIX&9Nyo;p__UjSODS*|Txy zwM?W$8;uh_F#5iW0#~ZNo_q&w&XmK;`@m$!&TeS_`e_0x73{Y{br91vTX#+|Yg}@cK~LXXqLDl2|gPj-@se zssmRJL7(8Kb<1t{<8b%p>>X=YWBcc~?>Oo{BY8bE<@UQ0S(okF|NKe})CoV;ZCGhV zB5Yc|%1a>^Iv-5Mj90zKc7sPXU~0?Y+28c@E;(%!N zcD5t0kkV@ZOERho+Dw-f1o@8QZ0k#H7BNPiJ8&lIUfb|tC?rR&Ir7j7o7R|lItDV< z*@kbWD&^vAP`ToP|AY8HFtvq8Wq;gfY_?7-H>tb;vsNrt zx3+#q@;;H~D~$DChJQno(v6zuq4ml5AzPD6;Cg7VXC(U>ktI$mZsf~5=v(8xOW;;F zy2@^_h`8y7Y# zFS!Y*6LOmPyyGy5pmfmd=lBj7DdDMRuh{|Vdry5VmKDIkrSb11ye?tmLH!+-oHFz5 zeY|IM{ap!BNMzfxT{`(J%(UqDcJl_pz{*zRoK{cBx@GI~%TENGg!$L}o>fBot>OH& zrJv{WV}2_h6C#mCi(94a&|KUWfBC_7tpsFj$#hfr!-SL-ANMOtU7q8CobBswuD~YA z=RcByqijJG3UF;gH8#-0dm%AN{-DTtylsch4g}44H_WquQc)&b?Sj0=>vS9m|x+nTYpr1QZ&>B?44=9^BM-(3GXCx>3zXCxYuFD$8QY& zGoaVMQTsyy7UwlZzBYL{?>>j>x4*GZV|inh!npYo1a*`hZ zN{6ZW`CWsyxEZ3%QEd8DZ0k)VPRM z^mqo{H%#<)7m|rA+zori%s&J9y{CLXoh`O!fXebT}Azd+El1Tvvp)H+d^rucVR)91bCRjvXYz z4ayIxV9^nB!a1H6^>g7eU^OJ?u3u^t@$`Z~b7~ zd+R9_vl^fUeE1vfq63E>amdeDHg-zkn z!9OEpp><2JoBlZqoE2JAl(>`5D~L?1FS3iYmqG6n|9*4DhcIxt+rLAOk&g}%lpecj zOrVR0`ZK4V)i9{?az-YG41KZYhW_kpp+Swb%ixeWB;VRncQZE;yOqziuQ~Jy8@3JB zjjYl8NB&amthVf+GB3n+N7e4o5rCm+?ShIJ#&<;e9h~2Y>=%K|;-8K0&hCQ92u#osZ9>zB&~2!rC9Vq7T%iA$hezX7$f`>s!!ga=XW!2-vzweb=GTEG#l_ zILx-CjWLh0JjX3wVd&T``Tb#P0%R5Fe%}949)}w!VGl)iW0z^A#b{_TOo(1-94JkP z-UZHEKXeD+RMCU$pZyqpmy<7_a6D3rhN-%MtqzBkp;wa7r1Zs;$dt125KEp2Y6{EC zte2hRXXQ^n$q6&Qf6$!D{aR|%8yH!Yx*_O_86)nWFBynG%s218DSCfI&%achx4m)I zi@kb2S}?q<3w{Y)hZ>EF)2SEMGk9X~@Uoq;qyPAvXM*?KSI|?Qw>>)W*zx4E16V#9 znD$i15IRnRAbJv^g@31H^j>-&9(H^R@wL{o1+wry!x;j-K8`suFPp4*zC(ty>C;<% zR*dJ(Ct#p-g5C#CWZ7QtoIfoCtrnV}?c)BezVl<2rP1s=wC}xUI>fnW-t+A~u=9w? z`-h)%NAvTwcc=e+;>pJ8M+rlf@9BDhu|_Y>NhK%fPuuyRR4N<>1OnYg zl8z!p?wd&IA7kiL_%13dY=bS#-P@NuvB54yP64;60;utx4kN9;4f$Q#=U=@y$LR0U zepW{r@9pQe27fBcTnW9~vhNC!Rv<+ym*qw&ohK6sI~JX8koUkQ?h+mDjzw6c6zzW{ zWg5wu3f~ffzCvy`d)X4Q5ENjj#MSZD&{@D#60Ko~&BnFqmyQx)sv|>aSpqYX*Y341 zK5`TWltNZIys?9h;_(i${w8fy{v2I!@c4F^5I?zL>HZ;Xtgv&rcz zxSse=&yC?Xg?|W*pOGfAWW9!!e}Z6m?LoPh*XZkpSp8?iu0n1`z0^B+kL87W6(Ffp z5;ZtCU|DPENSLfF3@i*)bc|z}KfnG_o{!ybFhbO>S5sqbZ#*osK*QxH46HCbm~e0n z&Zh2^54YYq@BA(mS7fwChKGnBKyut5eP= zkb)94=grfnwX6Gx#U-SC&m0T2vYfZS-!b0hDNg?9c5GYY_gH6-+kAZIWQu+kxI6Fo zsj!t1Ca;qrsr33%-Q`1!>;GZD=Cq&!%XQvpI5MBX?3{Kl=2J%W^Xes(s(0{=lL)*0 zTk5zN-^utf>c$}|B7;q8Nz6;0--M2Z8h?07=<90G@H)ip$k`=Cf>o07SG7T`6&nlP zbnYH~|HP$hvDOE48S#4H6=jF}7bGjn3|>tm!mp#NxnJmC!!C)dn^T?_ECf#odno2SoV;hTJ?hi)S<#2%35 z(jN~)O4hQ(U?Xg}S}(C?B%a8s#r^3|iZAs3N!%}5!myXFn5f|QPqq-5Ib0K0v^x@6 zu2t~Vb%{W?@2wo}9zLWLtTdW>Isf}|HDwvIpFWO2f5P4mtc{9LDL$%M%_|5WHrbZO z>|(5Av{gCH7Ti3=&`;9C#U-EqZJpI#X>~}sJkHd!@R`Nk!kWtFQ*qyjMCOsc#rN_V zc#AgPG@6vkid`2zcz2JtVY?`tMA-3mf1_I|k;Nk}J=szeX5Vi2 zKT-Y(de135kbLXSu$#cgOOp;$pf#Jz&xI=k+b$QWM+amv_Q%@Wy%gg|%B$kauxUnL z0i~vS_%53#x{pOYoxK=84+m9z>Y7s4Hqf#|+4kJF1=zLd%+c@aXXYK(eC5-H-nZYe z;n9bi-M^xtSN(FzgEM=vnRPf+q+&CcZSjFm(n0j&%RLE2Kl^l`C!KqXyNU%i@EH#+ zk?bWAR&RW$DK|hOI4$_)R`L|uclpY-MzNC!YMP-QgY@sE5ea+iI9h!Tu*YX*$Ta04 z^zUbn+!PmwLk)|)x(;0en`uGUB?}qff9Y2awwoS^Spk@fTS@i?3zTqV?_jcsqaT|8%du*^f^%4oI+R^{<+J}96p{N5k{hqPbtT=?TqAI#t1?o^Vw0rc`8%7`F%&(%N&lYx&s;cmy z)ocFuLsRRPntJ3`A^EP+t!>QpkS6ueuB7`P`#~9a>*`%Q{7eB_1^PezkUYS6E-k;p z$*;a>0azK;jndC(!>{W0BZ-Wkwueo@LtNCIjP%A~x2Kai7n>E!b5{LoUxW^2)K2%9BJOC$X%p!mR* zjBtf^A`9{E;+4HTNSsJLchK>V2ZRd@tQ!L7~Tw zOD|W-V#99sWTiy99YOuivh5;A>Gcn&xp>V_+*cgORBn9VtaTTP*o5wdInL>~tn*vj zdNybq+B6_*JxLFevVQyH-%6W6S?aW^WI;dNv1&G^M^bVa`V0y2bCW)uPUv0feFGg&obN};ozz9!bGEISZ98#TrcSv^nMf< zOk3AChqoJ{ufM;;rYqMr^>uy)GM8&VF^Qk9U)Bv031@a`fmhw?9T7J^&y%Z_BiZlV zln8ceUQOQ^$A%>imp4ji>ofdLswDI3#sB6%9_hJVTsxmv4je5IQZS^~F|nI-YrfbU zMqKOgAl#hgl_C)u9)~G1NkVt=)zVg;Ge|k>P#dMj4h=OAUe3xr!V-xwLXoQ&l7qJm zJKPk5oQl1+hdwa;ZlltPAI@85kgTzxguHxwTzo?(0sZ~VQfN)4RF zrZY?)x4IT!<8hux?}V0MvpYX?MBLo@>zhlz+P;LI^F150PDaB}Y{9pqXGWpJM{%}h z?KDGwybqvZn-aDi+jnSd+&Ro$%s%=ajcK&^k6QiLk*#GR}{wqpeSFLePb+$O`D{b z?3rN9le7Gnxlt%Tj58;I(XTY@@4o*-)L<=4FWBg}#@!Wq-X@RcmRy7WlIc+wOC2B| znHcX&UjYV3!&u4{!*N`_F8#dLcWf@z`=U|57Rlp*?Zf)D zTKTVJ81>^G>CtD82eyK{Ctux15mxLg@z*+aE))tvws>#zK8Dqqs{>EAiX-{mAuR)4 z`gabo!7Me=xH}qRSn4$viYzCxhzWdOCinzecTqa5&M~2-t+##79X6~jJg;^&<~BpV zUzY+~>o|xkiGuaby11W1Uvlt;M@Kaa!ro zsvSXhur7M=kLrF){G-uv;_Cr9Xw^w;ykAMLS73A0vZZ2Q80$sk(0sO6-x+;$BR*zv zUV^%o0&c} zLwmvS)I!F+2S|SRh|^@jICGm_Gi)SBnStK7$!K);pc&=<5SS zmY1P3cR!zipbua2{kry(2#3vk@~?G4$^G`FYt1Z31fueynWihyALakE-0C~@DeCR0 z()Yrq+Gnq~-qOcc*LP27EZKx@qsG_koX=wOO|1^YMU3w{^g2vlF;Oz6kB8S7I44Qn zfWCZn(u0lj;xLl$T=`fmVvoZc^FNas=>(Lse=TGvXc!3R{e^S(3obi9pe4p{% znE_2EgVUI} z-}QjnnO%odtS`b~z3u0su{!8K9@T4W<_mof_Fnls%~(g99X;QAFJ}fvr0}lBj#3y2 zKmEgh8{<1T%+X4x^%pVf1l0-tdE8fT!N^Sshw^=gpy#keevt>g-xANeZYG@nGq1h` zUBV~$TkOtY=@a|H7iXHVA^*GL78f(LL-FJ@D~ zxLc_MFD`ImTFQ2gDYaS5pWIpK{816|*AtaaG1(&dl*fg|L?!53acboGr7f5&wZ#0& zy9>}a+#R~|GNW&+ZSSl1v*l}`Ug+$55B)%B?N07p{+I*%eK&-9|1gGLrRUsx@`teH z(nl7zU*Diex~Q=tAp!coEKIP=JVqiou6lRTS_#NYPbyv>H^w3ui5r2_sG7>Vbn@NM0EK zOZpG}dBW78@2`s8mLchVzdCz0y+0fBe*3IB-1!}byJKGWt7Sr;14&+JcRqAG#|N^; zPs4!KR@-#=H||wwcEQ@%=%e{!_2BD1 zF0e9!UI&AIe`}dOj!7n^HBdv>EBs~U+OF3bBcSA@_VRr$LZHNeE5g^ z(Bw|&6ZB`w-MMleUMmC(p0HHpVv*C-X~_`A`uT{s<>|{e?Qzz~qieGu$|5 z_F(__yOJC82~a#faJBg~z8U-(eUe|%m&%NmHCE#~|v6HoIec13SOq87?=bz>y5o&G;Ylbq`dpcgr z74_Gfz{FXL$Llv_V}(IBzpauDQc6^s;@>giV57pzg=H*^@A>2`>b_H8`wC|@1~ug> zjONMRtF*OZC~y9DKX+|y3+Z`vD2AWDx<&1PwltD+Up>^)r1!-`8|m4kve9oCd((EV z_0dKcHKnn_=j~fNL?Qn;1NLXDjO&+pW;& z;dGHp`UZ|)uNXTLRDshzvO8ZITElqR@5k3FJs5GX>#TD0B_)h^_@q_&MGeM1R6p=8 z+=KQ>rWa4EZiSXVBY%zwav??AWtW(6B$6%VC)92+)~Skre|@{wo$(#MCIimBGCXF` z&v(*M^rFCA-kUTq$PtCTADDN!zZpYvwtTyHs&8oz%HO>|K_>F&Gu^t7Fqn8xll#4^tNs*4)-}LW{Ko$D~nMh}6 z7}(^h!_Dc5L6v=CD_R-vyX5!u{|Q?20UM9xweRK{fWGG0KWCVQq36_^pQH~=&~JYK zPfLdjq;-=nrJam{zFowc;krLq|FlSjXxEOEN$Y#s4$aUvBfMk1>yEj+-&I6$EEZ=j z3A_nxWrf-HFIsxCjQ*;A$8r+;eO06!7mw7CK8X#3a?ZU1i=hUl7p{`r4+E!-!bigc zNrX$AQ+$&cezqfkD9~syjo}3QExIbdanO0$2`8~d*pygRz;oIGqawAE7Swgl`M(37 zQaSvIERT%uoakh{r#K)S;xxLfjf>qmS&Isw;%v3ZJq)kcWu{>z0jm(ch?be%tDK z&=7E{b=ZZO`zto(ZLGhUL+4A_yklcWeYhspg=J?P%bCmXpPqR+o%RsPNvF?U&$q)S z#e2)lZZZ1lDB94{S$77Svn?j&UWOA{zFfQVN+|)0;(|JumCL-Ud(mvGMoP z;_WplnEUp2)~kD?Fg(1p@$DyBsEc~BT;$tDXyTRjyD=7o1KaDA_CBQhC9L&VmCV?Z zi?W+u-ag<>)WM%Xf;rhz9A`u0Gs-yMYT&Acw%oHI4PP8nKx~Y zJC@9=i;@WTYJ1q6zaV9Is!NKc7>ut|D9bt|06kw__Jkc{#4XAx%B?@z*BSOF?zR`_LnwL{`$x%k6(;_g8>mPo|9em zSbbYHNjXOUU-tIt)aP$`!S7KWEVIEMa)_W z|b2-u*lk$#ec;<=zIN4aMg%43mb$$EI`9(R4d#<x^Bl-wme;tLP-)q_|?%A-WS;9wST+)MaFjlnl`Q5 z!cMWmnUX6m#_gw&k`i0CdWzvkC<N0!nd0aGW>Y|E)kWimPc5*#_N^}F#6CM#-wGRy*ETk%?hWC{u9vuFuniXPz989 z7gWm_y};&7$Zz}14~@2pYo8so#iCP7qJPTI>EB0{_1c&viO@P);j)7W11BB|EGHMt z<%6?_l?4JAev>o-@~oNASm`2TugHc??8&#k`$|KUk{e4U#>mJpR(+fMX=H_v(V#iEu{m|-b zQqSPEL$|kI*lw@@+uXf2#5yv*ci6?{=;i(<8Oasd`likFzb}V%POZT&${BU1rl(H+ z4;Hy2C39H%!9m8lWP|Irq0cv7Vezde9kWIy&@(fzUbOr&6l{2U)*+Fxp5A-$QN-|V zJ0x2cJ0D6}RXB{4AMzU~-pF`-o|MWXBd~wzFyXJH8;-_tx_!y&4rhaVT#Aa27 z9a5&OPkScLVgYY?!wSkD=&u~>eaa^HpYt(}KRBqX68Uf6Gg+RNXSaD?9~zS1Z2a|l z#d_#t+YY5j!iEcJ%u**dLa%=35y5PC2F`;bvQLW4 z@Lubz<$3y^Bm$ThUbHWRzL70I-na$9Anx7!rfwW-28JFJmaIWau2910_v`R|aUA!k zr5Oxt%M8A+wha10!fLkf+6l#6yboo6Gu|&A(46H}h+K+URV7vg$^xioyQ((MFAKdZ z6E8R&eF5aurI+ocjg?`rcZGqMyMCR(p9m~4hNQ8^wOOC3ogZkd}?fo19P`Rk> zOj33*w(R9FEAyG-Q)}ecXD2Bjxol&BH1kQ!Sn@+?;e#>Ah-3X>Sz`o)Y75K-g`UCN zOyG_1N`btS%2$kk*de9Yqa%jr-aLKx_GgYfRPu$sI;Nh)5e*oT@Gxp%F2p9g$O{1q z2cU0Lhm2PQJzv4vIAos*wiNnULPqs(4Gyc9nSHaWW2Xr1`eS66G@q!toeCMlE zcw&!~#S`z}<$Q$B;+C>yTm9yEL0aIdom@~mapz|ksS5hL3MG3-ooU5JYg8}Iurmqv~(bt33=VYKgHti-KPZjOKMo9_% zL5t_m7rH{|Jj=ZAl0y5H&np8p8S50oYeOTZ^>;!?RlG-H9UrtmblSyEV$@4!@^|GC zV(0hqWEIa~P9f*?*sjE5R4i0JPSQFo+YEc zAEASM|3qASJW}i)%bTs~hVC)X0`AhE(DzAQx8xM#@4J#P&UJArBVX=#Vyw3)M-=`n zelDB3PzCxwda@gg&HwvT&1v~-x}Lb;Y;XFlbIadi>Ji67%f=b+!H_RhD|#@=kqG7r z4OzLv(7)AHG^e&0pATo39Qn!k`#%jlnl}Rvoq{&k_nXu<- zB(tBwN{h~tk*AFHtzp#+vEwTqLi4+->gRiJ!T5{nxEDJ&V`ZnGzzX75Xg{!dncTij z^X45~u^?BBiLt&sE7Gm)_WzLf-|n!6~XV}~MdjIk9cwSyU&&%um`ThRt(s6%09`}8l*Xxaw z8cim$K7@X6VMa#6CkcB5xE7wmi@FuV-4VN^+b%Wl+JsdlG);W!X3Bl}6iEhv6% ziTbgljzA~5Z5yyrfWh+m%Ax@hjLEB9$&1D61=pDF*Va(wLUmi;7aD#p z#x6Vkwf4!y_EGeo)=jscpw6-zN#kKfh@mLD%4;`En}}x+dDZ%x$;qoQy?T?ofvW%)>Pj1QH1oIxlPB0@v$tE}I5qU`4sNqs zwChhOsIk5y(xDp8q?=w4FO9y-JI<|)Cd#F>Vtx&Bo`dYz+V`k7cAaCEYAYiD&@2c$q=T3;Mg%YV6jbfOdu|oZ z!m*tt^{WhE^+m}w7Hx<2e}I;zKKY0h8ZdvZl*(N!1Pjh~AnLZnKZk}It<2Q4mevi` z9}g~ldVh&`62%+&eAw@g@rqE|&iH+$TjAI|MPdUKu)Ku9-kf7>=P|y@$&1BtatGW| zFZV$&0nToC`6*%0te+3fHt&1%#5w~A1v3d=r!e_U&VHJGT4Q19JH07I)S22crVN~JwbjYKb^)0TM#2H z+~lkFOQ@r6dY2Sf14Na!qPAOXXu)}FFRm;KbrfxW^;Wuh)O}$%SA%)FlIQWBs?kqppqBq!q#A&~q(+&zxK{sz)y@ zn1&jN929leg_{p*_SVikKeQZg-L_lHXO-?>D1URWLvF)rM4S|vHj(%GAN<^t3o-?W zRletve!>!;G>je2UKvwq1;jNOrXwx8P-G{U^ef|^!0fKySgjwp?EH0|FdgKvS=7_{ zMFvL0oS`GUg{-f!WbdGon@%{Xs%3H>+pi4QQ@+mLsN$u-b%oPdyk+L;>AizockygI zacT*tAEMa(#^KEZ_Nd`mjKARzjPHb;bZh0oAE`Kw!8HwPep(2%o>4Gje+HD__Y6JP zkg;~bmiy-SgaS*h`D!=H*el&v1~jP!v4RUaZBG^Z}D4!U*xDs=f>|4WgPMC8(MK3(N;z;NJg zzj?O?y)UX%nigA*->#Ed+d4n|9rX^He;|fgqtqf#%ikLNfvG?4%Bla4Jj+VK+N8 zf0Gv9GW-OU(Sn~a;;K+<{?k0Mmxr)@D(v!*o4XOu7IU0%%;qnm7@U!pyE_Rz0(?cx zFU#Vb|7v%1@lt>4dHOEed2b3m?UwvdIF6+0Z}E}=!00h^Y2Ue|XW-e6pD#+jv3UQG zKMD)79wTy9uB_ytWpVT6dT{rO@5vE3l;u>#D-?*`_r#olPxfpaP|-OPQR%T8hJWEz zBo~w+=EpUKqc4{1Dd@WrpT^%;fjTTdq&k*i@s%RAnm1Xs62QF2Vand>h)9n`FI)I7 z*#oGLA~QR5Tnys>)RYw0N}#cMf%K~#wm43~lQzVi+hOoXoWiLXe#DvzJDkG1ydHP| z=0!Qk_=$~Uyxvgk{8C;Aj*UI*2aBZ*9f{c^Q|vY{i@*OwJ3kP|sp%KN(_0Tr;fZSt zt1co!;;9E`NEL{@)voIR^B;AZNG=D3#VjD(;s}9$V5u$wBK7Y}x-tduS(ZamgQMt{kdnOedy$7XfwTJ1_S32ldVCNH>4PmxkP%3W3apdW|?v&N6SBW zq<>d%sThd?z1*-!LRt%&$GkJTy4BIw>U`yoP6AlJ&QPNT!aKlZtf(ba@d$R$%0$r~ z&u##^m}8~-L(G3hxcXHHuO_u@Ug?kQOl3YhBdX|D$)|k($deH~sCX$l5OoP2qM!ea z;T1_DJNf%W0>Rb#x%&IH5->e3cvmed8%EpvpV)1E33M0b)o*TNeu0$iZI5`Ie*o>1 z%8{#{Gf?Aqc5-*>7qpOhE&An=1)xdPwgwbpe99FeQBdcURoCqd?%j86Qh=1d`yl1J<9h_-XfH+HU_pI#BgyVHeue z2*WB1+@`^pKW1=VXfykjmxyBQrWltth2r@RWU3`${v>+tU`xju8PvN;R^!b$CU-|Z zDV3zPV}2TPYupLF(U;Kwc*76j>)V0Jo)mA}corBlch#B$7ZBxz)Uc&012JD`9!*Jb zK%G_J^V5DDMvTAn=|_nlu=TpE@5J6?fa5#}?auh?VJiTJD!F-la*zGVhN*+K1~;H%0wuASYVQd}-UQJK;$ya`rs zmQf=u!88{~0Cc1(TzTh0(b4+o0Og*_8R!=MmX6zssPPg@}886p!j)`dJSb z{>-b^gNW##yU%k|JR)nPZ#A^zMzLuPBg&iyfON}TB*Piw7wI&AR1>c>k=bpR^cQ z&THOtHiPMWW-0$3&(`Aj?WbQLbV$ukrpDo6c;I4xWo!eE(>_h2{8%4ief(y0YyJxi zO*s9aWkexL#=RqV3@i~jXK?z0hATu4P8fJJDI7^^ zfMF(VdS3K+LqN=-cncoCz|bkMTq&}(f_b$9MUW_za0`e1Y@7lzt}n{ zAacvPP6qk~eWy}>gnJLbkdy>D#x4WZv-f@KtIbB$nSEOFJtuMOdOiv1Rp~JLeml?C zPoB{G@xGd}O9Y~Y2W32zRYqi<(H-(NH}P!jM-H7CRYv_GSFWW0QHCk@fK1&Dim2<$ zqfw70GEhZtb07L_2z_|B$SvQjwFggebcyPtmcc8FQelW*j zJtF7FQ=8=!fcko${p5x!?3{`|c=TK~??0c%^dOb`j(w+@P@eg&-t~GZj-9Dt6n*#* zo?Uq{7?j^hs+Z0qTioZhnLNqMx+@3NA+o1h)@*mBwM*1 zI`SB2#c(Vy%mAr#6*{JYw-rxWtye*eF{};29yJYSi<>%9dq2Dj7H*{M9HBj@3 zzP%364eH<6iYMSW9Um$8uKI*(VuL-4R(8$O|afB!=As~JSuX8xsQ-zM<= znCu#8`HwhR$xWYNZ33cPoQ+qheu#(%*FCJ6l}24f=kF1BeSuN!_c1Bks)4ZQ6#k&K zGPHXexp#frhi6+6p_Z@k51xl(g$`dAtpr?<2H!!@O$kkr1fzgizRY&y&q4}Ed zll^RAz`&jOCh}bviXFGye9LbGJ@FKm0{)iG{ zZH(rS?qAAhL<6G(M~nXMgo*LoT{m1JQT^u&xn~(^K+{$j&RvVih3Hg5FHg^P)Y-b3 zq8;cB1G5(_-g__MQlWpRPVvB}orrW{-=4&+>4@avP2a`9_-Ct61#fWMqk(5ry!%M_ za2X<8v+=+31Iwc*`?IOy)?5aTbM4S~-ExGwM)%A-{rhX#z9mcNcUWUL<(Y;$>A}DTaIa|E6^>>Pc0Kdc2;%yp!U6<#aM)YP-DD{fP1Ol7y_C2o+d< z-k=^U>59s6)4c(INI|)_sw zSB{4*Kxy(C>5r79IyR{LC%;5+aPQ8lz17XcJy#nJItPecbn{{jh40*b~IybyjuO5Zp z5h}jD@ErOUET&q^-GS~)G5WyekB!HI6|H)ga`Vv>*tw)v!_+~;>g2{nd+^5Gu`4O>vG(fcaBJtc0$Fo-T4w7y zJlh5XN1d-2?`_`AFnJfcJ0hkXNP2bM6SEf*U60mp=S5A`A3uJMUgr15^su)l9VZ%E zQ0u(FfkbW$U%{zx`l9JuGnDl`OPqfitBc4|xbFO?6yqJEFhtCMFC-(%sMEW|J(wNF zs#K5~@y7UHNh2Nx%&fu1^KtrO*``R;bv5LjfY4gfA>8~$}1gL}5vF?UG*=kItne*V*;m#hBschtK> zJTe1^{tI&$}=6Juk3) z_F}%s+=K2vf$97~c-Q9_z$)fim-y@(Fv9Q144VEyU0vhK7VUjdvhs2Bxc_4H+|`YS?%Vtu{jIJeLZA56$d|vNVtlH$_Vgaq zYUoh1@5#kQJ8lo3op3rD{lzPE>{12#I{7Biaor4JIlR)3y0GL2gc6gFjS}M@fwVsE zfl>BaXju{Zqc5@=nA&_P^Fo~1KJ|3SwD$l^&ibicsSIlol41T2lBA^0l)|9RI z2v>owvoHAbs68SzFl0CW>_esI3!SHr#v-yYD&&(cSIbipdLQ@kNf-vUxe6=pQ!7|d_>&krG0}M6T)N;=b9oW?;cO@*fOB=M75V+ zJB8hE&prpS6y9O`{fwAzRA4A{W@&|87kZBfx3zD3x2^$FRq>&t$B5A5nal5S|2Fh9 zQeKQ`W4w`>Tr(wgD|*nt&)TnQ4Ue#Ufc3rQ*VO{l`d%bCE?WS4FL%x#ns^P}+xp*C z22BH#k66mb`w@|TLBO7@N5DvvH8beK_*RF)E^U!{Fo!IJ9%Q%koriACF{=P%2$MD= zH@veRz=W;cMC%1VH0BelzJk82J_?XnUlSASs6f8*tbKnwZ*%S}xq7ToA|nY^h)NFkb{2LT||_rFUszXrL%# zNBZMyIQETyAF>+vB8qv$vo6<89GhJ5(bv&4h)F9LRurvAjI}|VtahG;G225DIoxiD zwB6?7w#V0j$;&lyz{DAbGQP%FEb2p@#@YToc^EI=fZ#8l-&9PG{`p$6i^8|q-nZsa z-9e1MaAtZ05pafa6pu_q=D>MZr8z80uUN4kjg7_T$0&krL%#@0TZC_w$LyCoYNW>A-_eEQza zSUw}8Y$qYXdRd)vM0&@5q^I^Sl-y~IZ?k_0LlsSGa|f2=(Ax$&hMC5Tf_XB!|P$9HS*c{!7Hfg^vOHAv=X2{8jlrm^??@J$fmVnA^yr3MrNs!u{a(?5xWmG%>}4DvO0>Uz7$tQU8@K;leluAKT_1@lif029hlzl z(a9l?@x4xH*|d~u>jN>{MP)(sGa^$DJ)$&JE#9B?Bu&vgL^&^`$mo%#E5 zoCy<~XQQz=bosu$O6EVza2&jsqc)qb1;(D#lUK@{5jkl6#5>jdFhGp7zkGigFDZ1? zOBiH~9zqlcDc?iSmvE>!j?sHv0z?@o`a*O$FWR1GP=#bWLlOM{Dd&tB6;jQ@!t<0YTmHGmj5LN+U( z2mpqn+%qkP9imv<(=>xW!04LI(Vv*Di)UYGw(<76J|BW(>rG>&NnoE2a!24^vxL=xDLggCCiWYt=MvkrGrgWS8{ z6CR~NxAb3OaPcS15@Tc+aD&juysLFGO%|BUeop7}4v2E~_E3t}FhtZJ-4fr4`CV&^ zGWZ%c9z`rx{P?BCb*;6$`}Qd>mOnSZ*y_9{!EkAOV|N;2ub(`JP$Vt(fU6~{rb{VZ zKVJdWlDnMnjcPbfU+2^L3wAips$bEZC3lwXOHzCrTL>aRwS~|__dx<&f zJGqLu<xgnOqtJal0sJX=tSBn`99%*MxSPCPnGL8B zH+Db8@glNKZpN9~>_xkM(#+fF6q?>sEvL<82h_-Y@)vBmP#iwg-wK1_$5+sEWfh>+`{)OC%A(GmQO1fGIjA#fsdP*_gi4=_ zdDrdT1a&cgPHjsOM`U%e-_2^T@N6Ao2Lf-EG47$w>TJ!xGMcc(xZgg@F?XfcawPxT7VO zKh-Bm{MEf;0nMG#&*-Jt8r;izzuS*$BNBsH4XvoH$TW z!N^(S!3KgD%P3Ri9?+lGguRzehThDmx}iOpFn0Idngzp1G&p#_L*Y5re)f)o$D6Ij zh(USxBLDUrMDP8*t?vUnnjyXn&$Ug0{;l)zAITCh_o24G+~g`Wyd6Kuc2)~2sN?sY z^FmOkhy2Xy@lrJOa_Xw>E)2glq!2%}(|#%c97vo06wB&i=Z4HDF_;mF;p~Q9ZXeis z<}%hk>$t)iMkD;ihh8VVB*283SFTNWDSW&p(DIc+g|RH;@y4zNXe}cAwzvL5S^A@o zN+BAkGlXkzxzr#`RETEfb+MyqT~#v!(IbeZ&w6#^#$_1cmUrLm#e-Ns+p6CCw=UYp zjN{WS2~d$8c$eEj8%=sgy_C+k!Lvza)Hsg?AQIkT_mkoXL_9(sKW8C?NM?E?F=L`Y z=-d{?9b5i|@^uT#-*sYdk3LrLpnxPaNUkTh5L1m4~dPuub*4z-?~ zpspTS2d$EC`6)-Y0zEA@v_mQpwZ2`sA?IoZ>c`>kZ+nN;|6~peyE83?5wTHXOQpv) zMDkNg9$h<);-cMVrdU0w(|Rjs^-D7}eDId?pWNqX~T=HZ1{K{}DN$U}&rc>NNW z11BO2EA2?F|3_ScEO}Ky1c!dluZs$~<_4q96PbCdlo8pu>SARM#wXTGEHT=NzY5fP z*B=C{R$v)*nMx=<16t21iBa2kh-jvo>l9^(og)e;ytn~E@Aubgp5Y>r+!4=sOUE~?GLphN}E`#me<)Nd8}qn z9=Zoi!Qww!ulEBpCaQpEtQ8G2#=N(y+(A(-1znt8O7WBV3*Cqa6_Q$z68Qfb8 z_jqT$`=7(Vuz6?I|FnMWmI?(qJ?AIBE!th>myGe7m-aL2+_raJK+u83eko;dPgnqY zvya7OO#q%vh~p*UKpD{Y?~H4wI}qvu}PlqXH)(hmE#w+a z$l7(Ny*OjoDqdA}Q}ajL3;T-zW4FY&35<&U1@F zow4j5nr|x%3Eg?rJ$-Z0p8jpV)!-6JzO;JFQ0M@nk~8Am?Op?un@G6ugNDezbc}`j zu>98^K@G9T=i-2|irvy6xCqDj^LBrC#0wx1MhQDOh2c-GcW{czEO!407$T1k! ze6nVXU=<>?9C@Fgip3M?^1BrLJ2#<`0#?&n9n7y*etX3|Pk93R)-RYW{c8pWy#f#M z8)13qT{uq*w}519eTM&#MIQ^n)PvI}NhS}NS6%Vj_QB0-Mm-?;kD0XkMY3&9l z9Op%^H_twEEb58UTRY2-166zOj#AJ%MDaFoTCITPiFc~yBA&fwsIzP6gV&+Az|eb( zTbw#Xd2u3jzy5(`_gxJ4%41ezUch_}H}B+Nd8?P(tYr3Wa>lcr-4u6W;|r``C)D?v z^mxIT@-xA^W8A1ZjXQ>0S_IWxOfA`%l8ea4yB;+w&jE9RTKh2P8X|O9Cw##8-!Z?K zbYoL)I>ysAzTGFc-exK8gt|DNb6T9l_>`CtJ2j51$MRFs!ufKf4r6>(bwst6y23{| z&Ir{&8OA^IowHi9S{SXs(7hqge|Q#=h{KV)zp0|J?MZMcfrjJwc}B>v!O#&f`sdeylxvd^0f1whZRXVfr4e@fX{= z9xQ)@Ua*4aUQh)hpWS!Q-UH*kd?C9{I?&)dkR!S$+SM#kmv;SxQC|U6=5%X??~n(k zMfK-WO9`OIuiM98R}PHS_U)2ZO^AHzkMPA$%0SE0x8V3C4n6+#&~q{ZGX(_8E} z*VYP$R_L=1*|+e5iAd9uf8&G!YxjLm(>OPV-z5bf$cs7955;E+jf(x3_@-f&=kM^B zwVv31<32Z(IQAS(o>sMddlah!TRYNfL{Y<|#2|CQi@PzqzBif^9UIq%j<>x*QshUA z_v7vi({AO9wH7)U)9!*$pi3iiIfXLlJJZ~v;=3wv@g zUU*j5ShiDE&3+7se^`Z0>+Ps z*Y7nkzG7xAW9QXvSl+>q(L*jTH%u=gOlz+Rl)sP7dsY4W-S4l$NQaz;EKv^iGQ-P0 zuiJnq$LyNddtiS4UQ@4z+>jI)QMCGDeH)Yibn62OySsaULHgZ4(7s|(-rb1k{J0D+ zjAFD;R(U1&poJ+>Uc1k~fcAaW_ca$)QP+mBM$uhFVA@s(74xPc(y-%D{Y=2}bu~$| zar0P|Lp#EDOu70Zvf|3ahrBWVgX&}4Q^|@;xDp_RwnrcJz~YD`o;KxUYz=re^^I>L z4=>~6SndbdXhLH-mO)1lt5r=>5b9<*ooz%k5zGDMwWk~wXtcAqN)`lwu|naRk-8ij zmEr#QaCaBB4g!*2WA*nT!mfrbz5Vxa9B6x@Yd{*JJbia#Zy7I4SzbBz!88bU?eekh zUDV%M9?aOTKc36R&0uK^u3x(>KMEL~Hf$fpy_eciV|_Vh{WhWwbDVMe|*X{pqmb9w;UKAyySL_UAzb9=c9D>u3@0Le6iw&i zU)1kO3Nw2{enD@^!Sx+(Sx_s#l2?&}-KS$w*+)AQFnQDG|6pE$v>E7U&aUnbaECu% zW9-$wiNHvd%}cu%nB1mWSdh`&q z^#L&+UkH1swFZ%;x8bu3tD*dEc8$p{>>Pb~e8pAxry4MrUp>x7GjLo2jgcQ`Msb|` zL=LSDTM5iIb4$iB!m|rK6Eflqz_AIeUbv*&2nFrUiuOk&VDQhUvqQ==KseuW^W@X# zXv)Z#Q*9lFr)POSenli<`K+}Tn?3d9_9Ak*SMI>hB{_>q}vJ9c`j@wo(sb%%xQG4o4FE@%p3)k!nJ{<{#(SYVd8}>QCw6sfvN9yX;MHLmhQQTL z$7~VV@(ntNn*q9ezocV&5uOeI*z=VsmhUpkcYK?zt}gV`H#}1kmIp$@%na>u1F9Fd zCCXpgjK&PrGIo=XBZ958!TPTkq4?L2HYctw=*s6SP?Xz(I=84LtnAi+KIun?RvkG7 zb=;%pK2-rCA^o!sS?*B29Z#75E`vz#qbJLEh9cIhj^~0_T!?fvb#+Y5L!bs+e@57z z2xVN)1{NH=5oP$r=*3N^5oyf^<#TVayyu#NBHY!%HBh}n6rW&s4Rv~dTN^;!gP8kM z`PH>EVDugLIq!Yvf!1x@%Ec9pdLp`d1%Afk*t1vYpc(`W#UGd3B(}lm^|L1W@s|+A zN_`tI_t>)aXL);2f=n#nyMesYE>&Yx7fs6=nCC;}-efJ@{-D_RW(JV?`CV$<_Pao0p3V5PhX;Dpx8|SD`~Wo*Cdxs~KKR08 z5uVwD>ERt4nM&)wCn2)abM|vBOZbEpc~%9?GyV*~5#7Insh&3mQN_ zt5|t6e*^{x!uQ4`rvhz*me=%cET4rLW%pKgqy>hq?t68&JR5byW)GQu{RbcCffui$ zjl!W<_{@jGSWHhCIuMwjC)x`meWy)djLgE|_sB+fwlav^rS2eiMoH@B|A_)N-vCbbA&S>oBiam*A(+P_|nNDaob zsuwXHc48s3deuz~2hygf;;m`27lu<^CX8dSINa<`?da1}-r&HmahYa1gE~b#$~XhR zqZ)3?%X&M^9xp4D(u@^?dxJ~t71?p^qD;^P3i*`$JUkYh-28E^kUd0)PulSYj z{3MR3Jp51CnwEGgQ0Fd(%&poOzc;Oe!cvpK@Tp{jVLDIMEi`oYg)>M5Ali4U(3P7p z|2-jZ>a_po#ruU|RrvDPW$ZbAn=cvnz5v7dhK^GU<~L_Z6oz)JyoGACuV1dx#P|^Y zJUJ=k;EBn1#4(0}Cq_V>i*CEs-$gmmdv_>tTHA{8vF+0J&d}rXHU|yg{H%Da; z3!@II7V(sQ9IAcX@g8)R^6;UO?b`K;CXA1%(=zV8MgLnw3XxE-qi%zA- zIIIzqxrfL~?#Suv;%IBTOG&B?&tjRX@0jjjndKQif z{k)9z;W1111RCdiaE5da%QIm$nXI56+l=apibZ}pVz?59lBS7A+X?8^xnA?3aV2(s z*L+{M;=CcAO|@FZ?a19_`)|CWd*!#%CI3D2+Yv4$r|g6NFO}b?G_d^Sg!HnS%>7QN zJ2atfLb)55H#t^KubV>DiNn@|{+kf%j_akJ*?nkA=Ez1>oehX}uAp^&%MYk3%sLp@`Od8dUtoDC=EC0 zX|=cB&M5&5a+*78i@3-w=x=zV&MU`)ekYMKCCS%| zc1v}iao6%XeB||4j!Kldz$mX-*=n_$h&=XVNA(HpeUx9JgiD>Wz&!NsXmn@`wEf9V zFWrdM39fg1*kk7gXe_t>j%CZvMY~A%%-dHOzM3JsUL}zw3-oUBdl?~osB_01{(Gz4 zaqO*}FXlI|!EvpyDwj~bvnWR&sp++dpoN#Y5gDE%Xu)~hH_)sQ$Z>l4SJsix)X7j~ z)lE0hf~&Yt(8DX}m%y0P7s_+QdapEp;}JhFS3c26ywZ)vr}bCO=jx-bk^LAoq;=Uboqrq+3I^wEj0h4+6RaH4lU}DgfP8Bcl8(Vo7r2IJumeTr98Xi zBF6Iz>4Lg{UQFvEa`yf`nU2k9-UiJR8=32rtL+?vC_}6$1Dm%k{Oi-2w zCgNmW^)%LjB3?q%n*>(7sj;+vAdeB!-5M3RL%xw`QBSH*xDKJ-0iMi6-u2lr^F zVfngLr8YC2?-*}5fnVa5Nb8b68a03SsDIM+4}96$cDuS6YeYWs44i$Sq$f#bVg=tXAj29+3vh?+$)cu$`Z@YZxR7rTS>dFB0So){qHi@F*%|Ra(Qd=OKeMg*o zoh?xRIOrRB909tG=O8&l7)W+PAa^|u=*OZ8ko`(>co<5sC|71 zpR+B7(^~knNBhZ$2BbK*-zlEFgcj01g>Y+Bpq7@d&ACAdz#y%a;5qRO7`A&9^n7Zd z`_j|eIOEf(H0gJuT;*0^Do*OgPZ5^h*DLI1Fz}=eXhypS;@3Jrh2zc%hO#1R2#@W3 z`iTwGQXE!!TUo zsD5I=2ZjtH=VylBwtk1>uodJ5;~i+!QO@bgN>$V)=)3mxPa4oiZqy&G+6Yr;Y9h=k zH{sd*e6=c*TVeQUPIxr;JdC?^uMfWN0>ix4GScN*zyzD^n&;J!xy&`xE3qFD>grzp znB9h2GsfgP?)5{baz%oYBpG$|@>w2N*Fj80^G|Dtcd+|6^Z1_r9jeetnVcG3w+)fb zGj3d{Rf0b~xrH82?SYC`d-Z4)!QVeJr)|Et;mm77A0rShz`pf<1(C!~rqF|5V z$tSFBUN`BE!tf1KYu{X1Xlv)O(+|VuyItgv*vU3;NUfJUn89uc^xnc`8=k^NyUTCr zd5STPvuTY?mTD;w65lW(?`dtOdOZ#cyz(WA23q& z=lXfURq%(J{~G6w0|vWFQNNWh2o|f%@Akv^5xQOu)W0CV0J1zGD5H8cBD|>)>e=dw zYNv}&niSX|(pPHtsE-t?{VKiT6;%P4pPgQ}h`oUhX|=mD03p<(=;WQ=qpYUozuA_c!YQy#1`K0R83z?A4PN^()e5lZZ_pFgG^=2 zq5k4~Mr+pZ!1Pf@!ckm<$_hk{`X>7M#tp1}oq=riOAY9m(z?2RFLvI(lu>w^Zef0{ zW(~h9Sw=))mOXRO-XaX6cAG8hPTN84DdQ*Uhc00EljXBiYYGd-Qk*;O+c5tVQC5xn z))oasxnR(1mxt*=G}Gg!-+Fr^mTyp~anNfR@{pC-ZNLLHYsNI^IRKiY%|oNMBE;^S4FhXyWIhh#`jPQlamz6NSyL#5idluD= zhQk4Q^iIV5Nwr7boh)Rt0QxVZA`uNtU!ZMR{qln?KXh!I_sQ85hFH;_ z-FH7@c%lvwb*cRpn-Rh36kn664A4t2W&6&~A&T=JnTshlDCO68i!}05K0YjPsN|{b zdyXi^NBm!ZeGGKfSlY@qEI)B-#af;gmOJV|X4B_KB7jPKaN?v2W!XKt@X>@{S$3IS zx~yHl*CxqP3qB}1ALC=FpMcHV{QypZuxQktnq1G3u+NWdq z&DKqC&;3$6fM?^)cxfq*gW_wizn;DG54>_q2Ubhtasz${zzj zZAvSiYpX|$S*7a8erxo{`|cK(3yXTUus~o-%c5Ra6fO1A_BUe5$%4>7^yfh-bm!fj zEkJ*-v}NqeOJLeP`@M<24yuB^y2qwMf$~k~h`!Dcd{b3AKp%gI2m?aJHSy+)^C|00 zH?76w_KLMP49;)@nM2)^B&xct5J}&nmZX63JXcY; z^isH}h*D>paOyTUVukH{`qcreZ_FxcsUCLKK{I?aJ4UWybsh+1zR8bUe(Rwi(@X zpWFa+hF;$V{Y+qb-n%+^>p3Fr{4*Lwd<_F@vIbM1uR-}-1D(+U+7qwlYb*23uS)2~z6yMf^qT-UqN z6ZMECt}C4Og8XaK+RlQYx_yG53SOmWwdI;e}hlaS23V4BvUZH3wB!uO3(^7yV z1g7@fc(ZUo8p>18W=9^z@Dj|IC3Y6!^4Pf}5rZ#tJ0d2%dHc1WTTtJow)n!BG$3y0 zLC=Cqfu2X~)58nH*B(EMwIAGs)L$Jr2jYTChW3Vk_$Zvhm`lViwSkdtUCk+={KMP4IVh#lNU)CoMdA zC6L zOZDpzl}E}m@o6g1r>192p8bWPM{iTTzU@M#Ki{4Ez_FAMiYndoj6S-p1)9Qhv+uG5 zJew%rz0znpqHqWVh`V1~nvc~S46&9w(9&taTwT^rq4OH0?jte zPntPm{6fqH8?)XgPH13;py_;eDQ=Bt+gtzfxpNSrWPH(=TZ6@K#~gDHT~f%$v+X~# z`Q+$0^v_4$-W@auvpu&D@kLfLXdz(MuTf zPY$H+Pk+*m<@*h6?|c9I%N7`s+n36(`wfOweTJU4dqL)u$!`!4&AaN%mo$*sW zYGsI+9p4-b#ohz;?B!cq1r9^U?)IvOq~(0jBaa-o^7NPDN{CYOjJ_gl6w2$wguc1x z0QJ^2md5a+9dGiJOX`{;4A%H|7@A_dU$kns{eIp$KrGVAeZ#s4w92cE>B38REYx)- z{js!&JWBSu8?D#w57YXk7ar>YnyT3AV%neq14nLiob2gDK6YlT+yk%hY+;!=fga4h z>BO&4_~s^q2vWu8vgU&jsa3yE`Bea-1iu*hk+2$#T1paLhuEXeo?S`Tr$Z6t&aDF> zQp^2AZ8fv^$}Ml9zH|HUuMKoWMES&#%`=jyaj={mKP~|bHg8_b;wz;Fc_(oybGKhTnp-5&C|G1S`LD0Qzcmfln?! zptSN%$YyyrU<^!{T02A_!igqrokQnP!|q?TX=&Mr_+idz@9#uxeKQv#ZPsJ?Wu$W{ z*Ct;#0Ym;Ykz$GQV=~H*4ZE!V1P${iedcW!<@0o^*+&5r=($Y2LH&y1C1x#a2-<(B zh(&E~J#c&iNJl=cEsc7MXY&=yrxa))YBaC!gD}iat;)F*GoW}J#awGm?6PeC$MTdv?k@HI=!gjQC^^AE)xbvl`_wB z^8vj~Lc#b{4A2h!&JCBsaLx1^Pj;+q#pESTT=kX1K`ibv7sy<9_6G-Q96#Gh#ib## z#;k1}9mAJVM|E2mhYq67By#b9`YV_zPUsBbl|qz#Qm3tAu=8CbwDP^F$WKJcR8)#| z3k2G?fuxh?m_Tpd@4_*?9JiOeBi-Vq&;JEScfZ;#Z{YT#9$Q~;p2rUKc<(I>2^jw- zrSpxu$$czui*h?6fAZG>=vzn85NI<7%G&8bd!&ZC@XZ1z6|lU5zO(tsb_G~I9Yr*| zI`LdRB2V(Ew!Auws!J8M-hX$-`r)tk#-szQ>qyG7=b85NL|p+6Q+{2XK--fMZ~J%< zDB5YR#H-jn*(;IZ<5OFUD6X3{P7;^&864aA?BAw>C3%UyTiDpha$@_G#nmo&iAMwI zI!TAkHq_!cdwhRiJ)aMyot8&p)tB*x0%NDNc4HjI%SYE4VDzVA_YA4s?$+DY_K0O4 zRk2qU!!c1p-#xo_{V^hHuOz7VCL+3aPvMo}C3_w93UJE{X?;WFyoWD^KTX4k0cdt>fiHdcSV z=K1WyYg5-y@AFf!Tke@I8xLK#cE9}5t3dl%vyhUs1{g{fYNxEdmYoY?(7U#zyVrrq zb4Ba>u|<8X&*>Fqa0=z0S;H6Z9}B%J|Mr{j!s-`~)d;!}j*b8$PTfwp3)|Q9f?*j^ist z{9(MovxL8WFN|2}Sm`x|1GE0ZLeCNGyf)0XL_{36L$hfnCi}IlfyvLizmT4b<5YTj zO?vVybRTj)SdxDWCQGk$eXxBDvGFgrowyK<8dLlfI=MN3LBHC++cXmxhSO~CYK0)+ zeDS#hC!|nQtz{USVKKHp56l0gIK6|mh6hz!e`G*IN%`iQxZQ|!BmIn#7nWZ~it6l3 zf*9}K#G&!=5t}ofBuf?Cb#gjP^DfT0dt09I6l1-_4@C7 z=K3TAk!mJoc0E!>L^dIl_^tv(n&SNSTY3Xb-ae$^q3D7}uLmePX?;Rnxf#Q~s!KdB zK(=e$(RpwqYKp#m=JCrGph~~rdOc(qk;Tr5eEQT5q{_X2ar!|(JDAzZHjC9!V47c~ zzKP$A$U6=8J>$FudxS2})3CU&H$5PJ=1PRwOJ55fQOlrjZ;C)k=6ZdX_~|9@NRELsL47UjH0hSbSU7}cP>UVj6tM@nls(|Sd<19j&5hL@S? zWBLex-1dFFbSPc@GdT5;IaFk>A${E+g~+SjqvejPqq5E99vgN{ziaq>fv@Yy_MG!L&Rs)eP4)c5o6S5MWg&V=#H3Mx21VEo^88d+*}KB>3hR#ZyU#` zdcmZm(zDrv0f;4^?dA2`AM(|`PDgof#Mb@Co{2Z6m_KE~hDxP=dIde=^_QN~1`%N* z*71n@QeF^{rkFK+$Knv(TtQ)5G2i0;kK@^H$AUpK-zVe^d6>usy+}u_!%YrB$@&pJHM^;z;$v*`k*V*>Uvp;6!gSetSGO_g$ZHFy=N4b zXUQ-u%8WDaL>wp@oe@tq*MMJ117m{#k!M?sLMe* zAz*BY-vc#OhMNCud5mKZeEZ}>C=JghRrZ|OUxylAwnp46P6m2_Jo5@c6nch^r%kVU zk9zulZ@ul93$(_P$m=b6sNwnxLRt+I7#)6(Lh*($lKoXIIcEXLb&uU7?y3Uy`WUsSl=$-4pK z%k-;dHe+$rTKi9@{fe&lwA zE8Brp6#lE{0w*w!o96~`$RqOcvIf=RQq(zJpz2+Z;eb2SpBn6HdIx0|`ex!i+kttZ z@$`nTCvcpuY3i+GOL@g;;Iy1!SdBfZ@e*jaI@N>i4>6z4sLR$s5tf#XvHb_Gb)#ew zx8~wKF2N^Rd5I0r#y999a9RS9+gy8iLrF07{o=vT_FsWgsAS)kC4`7?x<8N0IY7qS z=T%!|{)cbr%t^7F2Z(ayM2_{|VW1nCSb1m(ll@{FQjNb+`mq zRg9m4F?1)Jl*U~8|47Sf9orKnI4N2I%<6eqGeRF7V7oZqs%er}nK}tKJu1jeh+q*0@BqG zn!auxZURm;d}ej_X#F%$J){ zwv}9k$OjLeH+YZbxfSf!9(>`r30cR_(cPW|V(Y$9T5x<{{GuMR^O3%=+M*t0|4A`P z1BMt{QuESVpy1L0ReSZ-z&t;ud%Cj&s9ozlgKc;bK~#yTq))`Y&r{h^$+i`q*sK`* z-h%N$6ZxkU_ylTDKkc`O>!m-y3_qQsmg54$LqC0^T@``3^^`^e;T15r!*UNQ{f0X3 zFF)vCZvvfDE#;nf4KO-Sy&Lv1hg$Zz={x&+5#?Zs*tuitU~qf$VIJZGJe#U((AfT;oA15m0h(W$|EpaRP_fcY?r4%O)ETd->g&{m;p9Vpza(6t`+UV6 zL#ri!4kEC6UAQ!=VM0=KB=s^sw3mljT0Trg3*)sDIa2M2RsAAy@7cir$p=0tKwF<4?f}r6QWagJ zjL@*)ShK187DPI1a+=2_7HY$2tYEVqLUxA`J7xEadc2yCm^?um%8Z|f5L ze#2nNGnw_%d5Cdp?!06<1Ch&KHCt&uUh0Rh=;sJ0$6z$t^Gr&z6~jOHp8h2D{JHyo z?sGhw%%6yk`;tJrNLZC`tb|BEu5!wYjzBkO+|b_lvrrBx4=!xuM`WHg0V+FZ;J$3V zxbPc(Y5)5y^Ifomh}ANMH1X!YKeYKwCEJ44e#y&0tn zcxGNKI-kNby5On*``9p*t%kl+#T{Kzjc|(_{4b8AUIxOmHiu%&; ztWd>a_Hxy@&AA6ZDL_1&NlI}f0P*d6^JzCaj`vKf3gMo>Pro zb%Ft&4aBz|bz90qLFKlty$k9D#QNBy>(HAFeFGP+;&_ihZ^@@5S>;Elqj)k&RdCt5 z1M}*&c2|K_Fx1}2xK_Am$N8#>HOij`X3&#Jf;`6eK6x{t*vtXrE$FPS`spr*@eWO> zH~Y5VPQ%`NN#|hWt7={+?>PF;$Zplqvi*(-T!7WI&z9C?-*Vl{m1 ztJR8xxr%S=%I_aWU1LpKM&G)lS>_9k3w`XU^IHW+{~OG1Z}nUsYkCsH-&9IB_J-Q- zfZwWu4jYD-&Lv{)7HJ!7JAg7{pIiyrx!m7F&uHH;g2(XNogX$nG6>39dcKPi#v!#| z2I?%HRK0V@{4{kjRzrM-?}6sot5@T{RPPa}=kqKZirO86FHwN)LuMP3Wc2R$B#2TlT$ zhhqoJcgbEtzXpfg9+Hm%gS|ua*yrOw8d)gMQOQB9JBn@%?=apy^80&XpQz^$+1gdk zp70S7e_gWb7hFK@&EpR(9X6r1{elNy9ydj8tu5!fr^ivx{{02zXF3sKyV=T%Yn`Ce zW#!;=``bV>saGOQWBgeWr}sqm?lFUw`LF8&4GIy#ION;>R!g{REuSnbl89P;!zW74 z4kG%G6$N(dFq|6AA=Ar({S2PXxA6X%4pDeH^FHEbm=a1df{1CYlZX`kSw45?1vKiH z78DgPk4R!?$LEF5LW^cWDmUd5P~G<%MxNtA9;d=?W-|wY>HR1&cwq5f5(B<3r+*;g zlSO@p`!t$zI}^>XQH1)(_BN#LU#k0tV^{a-8OhuTeb?*e>>Lp=VsKKDlNiqfbKad82yeUt(!;~AfXY!HJzoTCpN?6P@(qm6IflUg~Zj(Kcp&w z5R%k&>5CssuG2DP4pKgP35K5X5-8N&sS2pzv@k$4J9 zf0LM-RoQ_#-Fc_++E!qi3)Z^~bK%*dB9!N@awEd6TE5EUn}{5EML%>84QiOfhwHa( zMxA=b+`^|Eq3Xu=t%By(hP&r=I-ueQQ^zB?dS2-Q<;t78PZM4PbMx1j z(3Ug6G&nAPz3Mc~il1*#*F6HXB%ZziJSr;@yXW56j~0HXS2~L>_tUp$ z5^l5K>4Ral7sC&S_aoK`9)91`HxQB4?6jBTHzJ0#OFy{1fEEIxemzX40HfOXx3dM7 ze?UYLUTz!KVfRG-mRgnO5g1dq{_)fui&M8!c7FT5Z2+@#HV4#}IaHx2J)3gBHJv~^ zb9TpF>Y|>V^Sg+ART#&meu8wca1;^T-pqI>Z-U{vS4WTbQ=LcGSeP&=U{tL|_nV))p?xTV zS6t^dYHJ%+$^U}!-SwGoJ2lw40kKRDXc}8B-eVdoxBS`B2eSfI{jyj7;Mp{NPu87l zfKgoym(D?F|JEC70vd=g z^=md`4Tk5Uf0o$0zp@RH0}{X6Zpuc@_MY}Z{XvLChE1|7`w(HgVpTa8#=BfUbXVn8 z$aggM;q;NPXPnTsD2p3X!f+g#AxG3iuzZLa>*Ob_YIjucDRFj)PylqUgKw;pu=;Z4 zojb3m7-ZqN&Q3^#?lcFQ{MD37!4&8}cU=BtPXc0$HER6+q6RdV(bLIq$`I@AGuO61 z#KrlDdA{pZCXRF3sFA@HjwYo^;U}auaGcuojR&Va@NB9NZ(Hov1Lok4ZtsRGP^WV5 z)`E)@w65vV6s2KycAxx)w_eg&Fnm~i(+lSu9A}|=4vD`Ih()8564m24F0Eh_DIP0C zI&t)5rwSRE10cEk205?O41Y-is(IxuH_7RiQNV z6FBx+QvMJ3795wGq0)n`@xa(M88Ezc(N0`ftV})Jj=oCnSH4lr1$842KKw{f#O{lp zv}0Y8*MOW~^5$d8V?^d$V?)bLM4br&L|t)JD0NOBFYtYfCL>y?CRb1k`p$*@di5|gC-wnj``PN%*pv?)=j?uJO`Mb zmO*n}H|Gt7hU@OFey0KStBwrqx1}(8pe`jdP8jHacdiW;QbAob-E4aeSD0oyWvcza z7wX@C->v+E3_14;NE=oTBBIkl1xpDTmap<@QoSMY!tcFs)r zBVQI)<2cfu4xrPoaGXw|ze$Bd(6{zZEB}Kfd>4wA&Iry8S+tXCpZhwtnZjtPpvvA* z-bK5-eCS!oDEvILiucQc8IDWj_>=iuWjx#dADn2`{YRaqYdD!f`Xsh4r$ma^9>n;%%YTJ^k&R!{KcVB{ zANjfy-(Xa{e}ubC72EH>o%gY?S;|X5OtJZIe6WoF85Q1r{YpN?3D0)(UaWwx0Wj7c zpACK@if8j{Ft1eqjoKX|%7&^^VEosW*gr=x9H?wk@WnE^7>?t6VNPfY{onQj^oVZ? zGx`UA;N*JRiFr#*FX+EFeZh-!6jfZg=6*-EdU2f(%Nt+8aL66MR=)elUXJ2V>`&C} zU&1#*{05ZSsrUUmi8oqVjik=QYf)NV$H`K0vP>sNws*MW*Xw=k%u9x9u&p zFmXHb!hA>!>ba`U=Dl($9|IB6`stzqH<#Lf&{B6Ia!La&)WN=g0?V7A=0v2p%I^T} z-K8-$t-b}QJiXnDaol-WT#DZ15V5i`6cGZ?_9c)o zUJ>dmt&G_nb})KM(EL(4mLEud_WQ4*7?wBHLF?AfR2xRr3(6A;+qMGh^^*^e9k-!B zfz*@UNqo?y`P8Z=c3C}vrSD9QCFD>?j1Yl;et&-|mKQXXy7tU!;eW)BL`~NfdtHJ7 zLy91K*9s_AvhY4aS;8MeVPr#`&WY=b_BuOfoEwIZqFc^+rHi~=b{@!GpF50({{VwJ zH66Y$1ob|j(W=nGctT0WpL1rUmiTvo8I~=1NoNaCC)8Fs2vL9@qphkT4a>$es(s!RQ;K`i84kdwd7^BlE0z7E$u5Ct^%l+ru(k-+2-&0fD2 zkK?%h#&GwF0$@7V`^~v6;hoUJ+)-zRwQTSgTpHXOtf9uPt*i}B03mzp+l9q*({V9* zC4D*!7LHd%p;R}(tVf|d_JMJ2iwciv)H|o0FWHwOp~#D2f~Msgr@i&o`1sq9qY?`jhXYCzu^U-EmUT?idw%-KB-D zPdh@rQOcN?#2k+E_Wo+QhBz3KQk|prZh@}x8~n-XGr%CV#ptcxhos~MvL7-lYQ%*WlJ$UUdxRf1GyGn=m>6$p*CT7ROZKg)y`6CfRN-WPG2w((z^cKUTf-t@!*O z7$5PwSpGu`darNywQQAwBL5HP+Jr5jtXsn(L9Pf5&)**$eRgfx^W`tQ2Lx&`+|(2e z7rObqCNRdl_g$M@!u3Nf`=3oKLTaIgVlW`kRRjIM2cBD-Vfj1@Wm|Wa{j7!2cB$lB zghhKQcPw2-Y&nk?@k#SWR=~++{fOV7?EmHK44&=0@GqgV?TEQxa>*(Q%YP}8b}>Gf zr3zII+CDY4i}Pl5f_X7~4{A@A*mdSr9@IYY@y#7b1LifZkd{;1!DBE({DW&1Ff{60 ztE$9-IUavNCCvjZyj(YK-6#S52N5TgGa8s)G4h4o+Gu*qWeaiXEx@4tSeyM6%VT6% zakgdNOkUPsMtPIKt?B$ldpM8Q{dfWqXKtSB9xwrVFV|^asu+ws71MKn6^PokUf$A7 zD+grq+M;WH6433pX^@>15$m~3FYCMzs{8eF;bMO_R4ZqM&Za4&+E$sa(EAqHxObSU z5^vsy5#c^#tFkFzv<(e~7)b$RlU&h_cu|;aZR>bK5(ef$n%?XWZq&N(hmNEJ9kJ4s z4!+d;f{3#9?6MBdK+jVwKQVp|3Tnb-K0UYqBM0=)T$AgBp^{5E){e_@t4zFF#lku*}W{LtqAlK zN)ud>!s7f!i6-vF>cl7d8rR^|m+AvR!>t$XZHjml`fBw6cDmeV%*+2Hr@6gm- z`iVd^_W1Sei#^YPE_(EApqD9X@f7l>$C{$xEme6>nXhnMc?UbUl`N}&^8f3x!dg=A)iv>qNgYz-~-Z`hOh{Fta+|qz_^|njNl6awqfJ|FwOuOYfN` zhc*B8z3DH<_C(BL^|Gip%?i8-tw76F{mxSB{D1A|@^g2+S~6?+*P@;`L{;S<|F84P zN;augfbIYPe)qrcj|QSf(jz+OC*V1p__-6$0ewbI@>E`m+|LeaxAcMW43hNvfszHbsyk+nF54_VPh2m zOJKI&mv9bFKy3>xo9uSkFKa)D*Lp2tJPy4-4!$wW`vpS}YM#&Mv_X#z=iP?$YRlSF z8jq`g*=>bbhI*=(+p@9uTspM%q~J0+vZ(he`MR$Ewx|bKslPpW5X!w@n#gm@ENe$F zV7FrSI$-k{_jI7i2dfjx(w*aTYdD2ijKg2cUz~*D6Pgk?7(aobryF0eYadR|iC zY#)wG;k8h*7>4sKcbazD##IdcU8S;5Yzly4^2y(jLPccD`JWTcU$AwRsog#0h~XvL z`VuR5e_e*#gTAC!uUTu$Q13mhO@w?cRCg_G%al(Zj(x=0|5GX%Exb|=4&A*}4+6F1 zQPx`jS*r7cT8+4muWjl98moQU)R&28UsH5t&3n=4&RbD05-ngbMLP1Kw)e7i>OCYZc9SavIw}OFjE1vN7sZ)9^=T;78cnXt z`F04}e)K2|yG}yNZCPL68qA)}_x$Si{$4E1dD@ofTy#eR=T*W>{Pj?~aP8*UUTs9| zlUk>>W+@Jd1{#Gb4}2~IrrPSmtS2W>eT3@O@4NS-;%a;QH`3Vo>*}4T4jyzw9eu;6 zd$gkwy_1-PvwjW3*WM1SGW-eS!^UeI26h8|k~21phaGiZ+<94D<1%6#scyVrv2{^T z^}ld)Vg@x>UD;$b*8y3DJnfc`CeegzqEhgVCHxU0KXCGicyb5z3>YdC-$=?KI(@hzM!- zGb^8oqpsbjS^{&zp!-oMgTo5rS!LY#mb_p4H6p#s%hNwbGGqW{f#+b7+Nh*P^8L}Nlx4K;X*xD)gDOm;Y`+;CV2R44PC zbNF;1X-a7I)vN>><>M5mMixq_YwI)r6oNXR2C`ij!O`whk zkB*!*;r`EgUwS{WLb6)$sMFH8A~iyPtZ>6{W>s7KLd9`^VdD0}d>@Yh>Ky;pv_O+a zq!l+rZ`WfuR@Rru$$+lwn0#z1C`jFp>5Efr>K;mkR*3P6lPYw5S$^@-_lV8hLn!?! z2Ya&+v``_=BIO`n`M5KkyU6F!VJ}h&c>C^igG5plTE2Xxq zmu_f!P3(@ls!R7LsyiEc^}%C}|JILx{r)7^6_abc%j%sjYuC0R&-S#n`f~X(!F*j8 zKecS0x(~VO^dDTg?02TlLEVfDUp$*a>*o89G5#6yJIjJmrE7@A+@qiQ<=_3)oe2U9 z)pCZ*A(8)O4oJt|A-L6Z=UXdp0+jA(D>UId-cmu)fqHuk;1WCn1VAyjW<} zKooO>`cSVdD4o3-=>9$Uzs^VBMitTaB#eI}nZ@9Z%#8To&O6X6TjE7_V0ca1_}<#_ zEsqe%Ql(6u#|mnh`3}Oim|wGMR7i04HC7+7^Mupa?N3{luM=>-&r-?!uir;apGU&4 zzei-cRy?Ui4Uwo>+~XIwqh7rO3YwvS$O8q|<6aq1Cvkf<&vpYCswG&wxqJaJyr!kN zIQJk*n&{sXMwnlO5+9mZ${zz%lE~54>g>s(gFL*Zg z{AZbrCPcbBtd&#H2_27#Yf)q@>Tir}zS1p<=ox~?Uga(2y`tgAZ)|RwXCbC<$g`(h z($H^NIVL504PEy4+U#%2gnEAyhn!E#^d;$h*rDwgzub<%af*JppLg&h(3oBa$Ap&df9N+M$r{{QR;L^) z4?g7(&t~k23Y4BjSdp z9nD>IG}|zp`I9dh4O2|x3~1F*6~()$jJ+Iaha6W5OFl)GENd*|arT907l+p~=UDvW=d|7G(Nd(z?2 zu;x)9tjBR4(O8e|%cs)v7tal&OwI$>w-A>6?1&b;b@KY}B|8@pZClo^G@*ydfVR3Y5}sIBQYnE3%agYxPaVSAojLbc-69<&j`1Mj*PBrnyXz;9 z-|NvY)cW(n4y=w^y@rqXx8M{+FQEsYSicSW-L5JR-;74YB)g3_ZSjbFs&I3n>vLc| zh`7DI!yMbsTYH@=cRz#LPI$R~V(Zd(NAn~;UBVT?2e$FeCuNq-GoEdi_W%ZtyjdOO>&KoW3mh5fRZeq@BsQLyqayxo%{H=nGlRHLd>luAKo6IfUOD5ZZ zD!(H1*J(b~k##Ikp?VFFgX70J7O?nd$+gRW>Oz;vBh-HJslo8xF|0jhjAmT1yX z0z-Mgx+!3p{|OkIN*vW1$*A{kt6@v}-v9P})aiDrdeAl(YLX@A1Z#^Bp-je(?X?EB zK8i!VEWThgF6S`ZW)zBAc%(>9B0~TAzAhXu(MDAo&B^Af!RMuT0h-!!7iWHPS>6ho z->5NPyY|Sx&QI<7)XWCEa767edoM|OfLO0Ql6G6>pasLNOvkWKFnrSd^`gZNjBe5G{>l_al-=Y zaxz-ebjKCdNQc^T^cn(bmET)4%QDpCSN-`cZ3L*#pKAmg2cnWk!a|x{VavwZXcXB* zYh9+V{@3+ycPlu2A)y571e-~cUr!)%rN8f09WB&x&G(BcYgv5=p!e|c9Q&%Xs8^jk z+Ewrm-YIU2gi~*H|8*UU{+<}9Iwk*V5&2;Rr;)Z=)N=`&{ts2X!#MptUm+sWiFlN z+YyOv^wxvuI@FPIwanmr3HqD=Csp~AIFuXwzIf|J8?@it{_?#%LWIulFFSh{qwu*k7Lb2A<7EtO7;0RJ?{v>B)vrYMxjXLfRmS|9HMuXNT}!bz2Ce9^zTfDQ zp9@G=ujsr!qy)qepPZUA@rYFLWx{vOv_41p^=*5DZC{E{9%M-{D8) zyTB^piH-bYj0cnGA@lx=$!8c{_PiN4)mhEDQ4bMk%kY9GW3&=QC#?JN%;kC;zHw*Dok`OZ9j@yL0L9tv`-C4t&-@om?f+B_bg(+`IR@ zO!a@Q=L7_?%A(1~SpGOA$SKU5O$=svkEjOB+ynxzUGvp~IYiE+3MRx&~%0ju8M|Msb+_x?{+*bXr*M9_O2 zWJfVXUCvfldpZ~G$MJcEGrw-ZhoW2WJv|EWY~t>B=Z^(0Teo^8;gCyTcOud*{-~Z0 zPoc-b;qVCK7fg=MXR~=LqWTux&PJpM%w3j~w1HY^*V;Iy+~$wTJGQAC-(E?h8jFZs zV|N6B_Si8Z(*d&sM!T;*Z2O-&ODi4oDJHKGd1`;c56v=Qd_xgjI^X~6zN&NT;u^s4 zIRDy?_~~fn)%8Ls_|ukEPi--|!Bn01c%SnRzDM5`d@Zc%e|ry}?PLFzr&;WXl~${f zzHeDx*V6a$LUU7Z3^~G3&OyFcUoo8IB-Pn=O8_QsnacUs6a#qBw~4DWJDx7xUx?r)z5XTy0{yh64CB%_L~4AyR%%-~Hm+D>&687Z(DXzzu>)`Yul-?h_dBnn zaV-76KUXbRs&pMtqZG1k$@;@kIHUBuU_Fp=Wl@G-oe=YodPBwvjNiOsKgV8nTy7`?nCT;GGr)7`fNiaT!bk}QWol7c-zIwV}7L}!pME+9!sF@-L^)6>l^yrnD#4) zQSq<+B(>MG4{>cr9U4bej6VO@xIHHS-^s~n=l-i7{9m7|y{nw%X}1n$l4J1tM}Gpr z=Yj5B`6a(7B5z(8BK*MSn^{P9C*)%FIGLRDXHP1<`QLsIQ8px_pFgnsm$y0Y0Dd5XhhIGtv9vdF`RvzO_WsuP zlLNP{qWn=sC^pZjZb$tgl(lWaNTc?la@f~C#Ab&Gc zr>lsfS{|!xJqG44Azx{8jFDSbZ*}>;O-teNV+D`+qC_JY)TK@o*m3Hc7(LSJB8cVL1)7+f6E?5JoP_yVRVjN37c3)ma>Q z%f>}`wk}BgEFCfy4)1xY*?}kq-y1}VouH57&E?xYSbllYQPZQJNDDxEUAWCRBKlwF zuXc_&p`pn_MBJHOImOyQ70Qw(5%nQosY2T&PZlO;bI#d%dLmZT+|ZqVG7x`g7n+-5 z@x}#5YZ+^=rML&`c97MJ@7{&_SC59uZdXJkb!jI%&t>_Hh$!)4c)jij>U2Ex!FMNy zYpxgk zT^7g3?t8O_uOZEjh*;}AF|4S8T4qmjJ~7cl?V{Q?7bKVMLsX@JzplLF5h6M6vGR&I zi%87p{#mm2sKueBq(`w6Th}%Hr@SLNp|bfC>+XUGiaBn0_MuT9P)w>4_V!`%;?4_? zPD>Ro`Iq6lN`J1*xg&^mA?QrCRPqlHUy?d(E0ptoOD z7)Zo$^R+ks2nq+_fySZm$2Ve$j}DEzeCueoa2a*9IoxSGBaZC@Z}*{9x=VGFP|ujT zw9(#sh;$@0&gJON|GGXiZsz&YN^LL^;N%kc{WDs)nKit>`YsfGRdO#h`VI9&Go96^ z>|owo<@LstWJK*y?=mRF!{6@#m%n@uKnoL)#g#gXm?t);-qR-I*}jRrm0h(Aw~an{ z1Q}hC^#Ib#^&RpCaX^=c65$DHlyR;jWqS}7k7n+iGG@DU4?f*lUH(}`8cHubAI*^b zhy6OdS3~nfBmC9te6B^0hB4Q^bG;nP>wmIJ?k3%i zPoDp}L`6hG+lWM!EYN#uFNd1@psLKLg}0OjfkqUg`xswBWZ{$n=E4maHgunNpkV$1 zlCIvzgu5r;k^9|{(t;73I79ezOvdZ2*s+~Bq2=TOb1i}r3^5zxzC zc^#Ewd<(RB$82xjc*MHBAbWMd7xkGI%%6KR4?}pHn`^TC@oX+BU-Nh|-hkfK?yaB1 zT%pf2^pjg^4Pw0#e;)erBpT2DbS6kO5tyrEc0`I`xE0c^NJo>nSooCrC|^Nt5GZju zDTFm1Xv#n!EX`w4pMUFj(Qflo)OKp#`9|;nkk{Tl_xQwiV4Tr(ifX%$h)&|^7k5U( z97TYR+oA&uuQ|_9W9O!Gar^8#AJLd6I4&MfhF5O0+0~kMW0S=$pwPh15 zs63UB(BMGT)+S!nHCUa=S>LkDz4x*D?bJ7FcRt`9@of4{EmI!KXw-wSW}OL!mn6Tg zb&G1ZhTgtMo8_xUm+i0ad!N)=USm8Y?FpgB=IoZ`D?*MoZ1-fikA87{@$c>DED)G(l_3p5N47od`j6L zn}^S_JO!(PKg1%&txxe=nio`K5;#^}h~X!C z9>2C*v1SN0w%Db=bE5#Adv{CE(VeJ9&E$%41Xf4B@3XN@!&>Y)TEWeI2?0&0Gk>p7 zmfR!6(odP#IkJ>Dg5xyT^Y0mr0`iJ`JIGE;`y5Jlz02tg)<&$2*Y>Q`;YDPyywDlC z2kYO37+*7qDMTXfb1*u80eTKx7`d*r0kz%gm>Kx99TEL_J;pS!d%4qU=ru*u17&{N zA_LA{Dj< z2FmaLGT4WGpS6EMcaLKx%x(JUB7RF9&$dVG>-6g(`1@oPUFI5wLnh6XZ09B{*`qM5 zbpF6&4J$+%wX;tBaSbtEH^;u@+kpr!wO!^<{bBgsuGzhgOK~(j8~=AF%fLl_*5i1a z{VpsoZ`hp8%<}d+7<#a;xdG$x`LFxAT<%+n?kbE=y!PDsPG&OYzdeU2g~bDvYn)+r z^*hfhk{K`_)cuY{E-?GY%j;`u7%+~VJg{GL$xjNC#}lCHgaAyo;QY_`;keFf z2>6RXfa-;IcUkRa`Cm}(^gOA*vwzv|{q`M{=AM7-4_EqY)m)6noyq6mXt?nORP%Cu z`rY$)**M5PFU52(FZcftulZZ^@;Rd>+2iNbDtDoZD{E(BbHdS_n8rX+Q1|kFd-ExuOhUmHH-zH5m`2k3_ub$_D^|8<@E&$^4t zHGYGkV~s^u9j@WndzCs5-snVq7aj9FQ@@~wf*Lk?wkpKJ_unoY=K)$MujlOVtN;4G zfjRAK$@pdZFwo@iaq*OaWxsb4?$odIzmF0oGb4jOdcu^iTG8zmJ=CMuzu(~{CJ$Jw zbMZ5U$APY^ptsn~W%EkX;M^{-`(N=FQ2S)hY1M7}P)$vKEvu{J-`XLH?X4+?$J0Pp ziH={7z5v<3c8#6Eic%@ocKv_1iQso?l7{PfUKta=UFfG-#WzYaF~jXML6HOb2RU)qunwcSQZ@ z%BOda_&o$!`IV=^%|ks(%ys>f$j^!GEW zVbFqHE?oqM>%r4cjbfmdJ}lJL()?fd1Hm=>#A>H}7(Jb6kzl%E+24nDNQ}KWM28x| z%3a$}cA<2So5x&q2zWM7^ikr`f?{1G$Nesmh`B=iwSOvBpC-OV)X@#Yc|u=`@+ONr zaoGBKWv8FAPela90@FvXgTQ=bU9Ii9l;?<|8eOgiQzxK4K(^n{GXsbdmiDf0g{X^H z9&hR)0tJ!=lFXsN3{-Otu~N8DMtauUHatu@i%C~T<`%~e=!=E9jzHV6@~fNEv< zs8Z=HV#QXTloikcn)3Y4bN?S_Zypb2^!^Wzr4m9a5v7F|Wl5!w&a|LXN~M$tZKzaA zi!32RDoYEcB0_{@$=Ws9_kCYxwTxwE?0e>!?;p?Hy*~5n`FwxRU%gECa_)1kbDj6K zy|2Uw?E6~R;g;ZDL5-(c7JFnqOzi!*$HNe-1McWu_gv=%PyGtdK8jH?-8PYf@y)St zZ!<Bca~D<7 z->ML~b7-AN`7uxc_4-ErRR+1d5NX;sTqCFp7WIoPznS@PdmCKj_9>19G?2%O3A$)c8r| z*Q|3Us(u}FEmiRdbbPQ@i)6VW);;kcOBD_Hbn&k4Vn#4b+>j6YaMlpmIt_Vt9&V5l zJxZa+dZ9u`37gX?X=vb-q{y}7zQEdaSzgBQCK~LU_u0}v{-5g@Hm!YL*gY1!cgUH2 zV}F98*yM@Mt7bs@m>m-G*`DY6pKMf7(1hi|7Mm0J?;Z!bRo|Aoz4Jn#~wM z-X-ljWHydyCGShs6c3`-Tz0>&-#cgtW5_fd5rk2@xS@@@7+*hm>n=;Q@fd0iF1k8f z?gxW`QpMVL)JpmX z;r~4UfikiBlm@o1FD&2g1t7zJ6r7Sy}B)v z$L|khcLV&!tXd2Qq)E^Px(k4!dhgI-`KQ3%c-Es1Fux30>@dG=E$3egbMcDxDQGN% zR+bTn<&Tqd68B!}BmsH%Hml>7hk>kGzTV?xCxi-ow%@bs6439ItI<7ZK-T$stnpF| zkOCbiywZPQ_Y3wiiU-wE^GYFy@n(#lfi2?PnZK1|m(I&MvvlcSd40kGItq70fclI+ zvLs%LfOFGo+P_K{Xbv$S$h|!A6;%AE^y~TgnLt@$zA5v)EB4+2|FQ->Y3zNyt49_W zaqVcRl3la)o02pTjc%XpvG`|xG_dVrl<~gi|4lv!e1dKC#6cc=1~L81x<9fmbD!63 zFZ}8hq?dzWY29jlh(G-KqTjx90k_c{=bgeh&6%W;C4tb827{@-}6 z_e|5T?4gU;_s2}Fvth7-+?HVWGD8Qdb=wsg3srz!r()5VQi;CY3J(1k#>?Z<&T#hL z;If$K_la69_9=G~P_OiWWg|B)1xUJb_tqa+%`>k!p|KF{A8+w|EPDT?0~=Ac6>WZT zD3)*6p_sd(vj^+*NW4#NQ2)3TMRe_+Uh=vID1G_W6Bc1G8l+}rKPLinVUGg7q;{j$ zz$bR6-io7M*@MO}^qJ5g@Zwg(AXe{>t{ite-PMl`ZSPjvv)b@ltIZR-@rXmLOl1etK}CvlA57r(9HDu zf=D2y{SXtB!|DgBmpb~;`nh_M!1qDXdiujDAR7&7wg!Yj^ZWbVkB1H;=EJh_sxHQV zJ|8cjX|J(xSP3cu>;fVdqyTyRN=dKDF-Tfo7+rJsB(w!V=6*vlY@XwGG^gJC1nk_H zLXka5(DUl2%XVLJNR6)?8HgnSbwYww7xW&PC>XEV^SByFV(n6ea`FV6^w`OZ8P*Vx z@P1(4W(8>Q4)eL#@dx|Pxj6h$8#`dsTpd!5c?Rr}-=10RR;c=<^XS$CKcIc|raT4+vo2?2}4QkHn|TR48=xxUN0jW2uK{~AXE<&b#k6K|gW^uV5E z7@KZOHJ~02;yIT+_ zGCR%}xBZnjgd3~*oe&2U$@&8$$H>4|@7-+0w+Bt`1=YO%zs{MOuTc_*PoQq2{Ic^0 z2MD+YX)72tsqo{N;KiSdLV=>Y#NdfkA7R+i1#5s)2_ zUp!ID6~wIiKmsmUz0P4T9lMTcG+%$MBT$;WJ{P~(kIlQWezjvp4v^1jhyGUj%Wt;v z)=Xye1p@B*mS>Cme?v7pMDyYp#(Tl4ZY)g3`0RnbWwm|-9#6oT-&40dGYa30#_TS9 zfAr7!CzY{&O)Vw@>B{Z)II(eH)sD~zAD8^=dS3jrx3|wExeG?d+A>9VW4MeY6hXZF zObENK`iF?(G7apzF18(O*?I@a?ezIh$^W#wIQzP0$7I}NU{?t3$dmPl&b4`dle?vO z<;d>brVY*<`+75V$&!H#0!~&j$RRcn*!!2?^!(ut!^5lh?U2CuirCXd>hF6n`=9FT zaeK$J5SSZm|4Q7o5qcMY@7r5ffz7jj)RDn+7YR7AZuvj$51}r`d5=jkb{-Qgsq=5& zUj&rou*190ErqcAYe(8#R|D74Nsl8`wc)cAGCo9)5)Gh%j!u3=A)1I4;XBOve!%-I%JZx`bF z0ikx~p0WIHFX$!@ZL?o?0$YzWN4MO)IUg!s9Ujm5!J9A5?tK`b)sTfb& zxXb8)p;HY&RXClsS`d?CO6ow!LDe7dO)1;Tx^WcMx*k4|#Kipil-=E9PNF%$rXLgT zfPFy9xO;s~x(4b;mRDC!cbedTB`k%HufAuUWwm*-lDXdtMK_UiR36fHSTT z?psmT{@nv8n%2d?-zEWB_jkE&DW=~>-#VX&S%vXgcRbl7e(@a-e-f{}Upzos7S@QG zjvOQZHbt;fXY!CXrJq*?F#szCMj5YXfuy}is%Cw4n9G7s3JmUwbuSk zP&d*T-ra@qos%oCJlPul3P>B;j|SyoyzCQWE}^@IF&^`0hxX;I#hVatHCCGQ27Us? z|H3=dT{nRo5ngkB0UPR$IxWA+;No1ws_#}j;BxFg$3wtr-FVdBh2iZE@4+Q6Utc5O zPHaD%X66pWmG(!@6mk79Fjz-*IzGU~SI{eJ_h6%e8PEL9+@ZhX6R-xdr2<~jmI>^y zKdHO#WBk-3y~6^JPCP~hvful6Zha3m@%Qf8KcS&w4_}+|4O`&n)w)Yfgi;{3x)@5} zrC?@Vkmc^r7@ur?^zpsR2C=&74*X1xNkAu%er=UOyH269sESXlC>+oHw`m;k5@r0a z^ZI~jrhcJr%LE!u^GmN#ZQ|e^k$&OUp#Sat1e_)PLiq9^ARDZg+M~!hua~QK+bKMU z%+sq#Xgv4d@AYr~w)HVLFBwQ12eQ6g;K3Q_BrCakTRN6kLaaO_RXv9Jn;eU-ejRop zqxjm|Eqx8bKoo3${#ya_>rh{xxyY)PgzA1@F+z|90e9Kw&6&!-`~|+zLP^%u$mKJdVA+T+p_*4uq4Wi+g;S4gOS;&xGfOK_YU4L`YDPA({_vmu6sbh$zK#c8i4gx zuUkmgzMH!hN>zfZ=IKX4$Bgpwn7P|9MUZ@@x`C(802(e=557S(AkJynOSNtX)(LI% zLzjONaN-R65$!uLxkT~U?+_jy0AT+Xda`ZpJRm9Y#|%Hj>fP&}G(_7QuZQF$GXLzJ zd~AJU9rhkr!h)V|PkD&OqfrUM)|OmTAn&f%d1+LJ$=jvnZB23^z7J$}+OKm)CFP{Z-{R%I(U3~}aJ-?o8bm7KL5V@;+qxbA0sF|moCRDJO_qpGG zQSaQu^2Vr({Xc)d{sq(Pd+kXpolAl3o_AHC^cGMa+;P=uvGpElc%>Xrr>-cHC9)B zs+a= z+=epJ__M3kiKyd?&vA^*fm8W+};C!x^Ivp-+nGX&a+=(oN$@<3wet{n}OLpd6! zLo)#wFom&x8Gs^ROMETp=7V&Q|u60*R z%4#!^F8+FdMoA0U?UzPF_TypngJrP(ESFbl|%0o!ZbvTMnCp5G}X)o&dO`hk@HdgAc0Y8bf) z48u*?sD<=%v-WT_2j5nThl-KWAC`;p{%#rQw|gFW#eE;PPh4YZo!Le})vNNESI?~@ z0-~<*Ru3u;O`Y7RH>JW8F9g~LyccyXHvgnU)oz=fVs(BM5ySZ;U*0;(neL+{v6a!- z`y;oQP2W1hS?>+Lje-~-dv!#%jIi=1AgSz>mHUR_x3tRPTNjwv{v$~r)GW1pfj&RE zlyidq*t``EjX2)n<)>mkRn{FfpXT|UtSu^a{0$uv)k)hgc$Nc^`Y^BM z4<8Uey$VhZB%z_CdnYUBaq9?xF|#6sU*{wGEwn?2#?QgC;%CmOsr+TP;w&UhT&iTr}e>0Kw*Se*99zXKi2beCB zztyZFpj-UV6Gi_gK-QS3Bjx~)x%Vr$&mD*U%DKeRmvr<&?9Nu-=|XHCUtga(u}~j~k=_%E6)&Jm>t}X^ z{(t2iL=$VzFV08Ucd=7?!jn~&0h=hf3IDaA-A@8#xE z0I|A7bJ@Tu=Dr7?$t(o!T z29={9DrQg5{gr3COR>k3C?5(m6F>+mG0v2Xo*0&cO^J1sl~&0NRvk1nqUvO>nC1CLHZy-<1Po{v0zuX&!| z(Q4OlVP!HR3m%;Al)-QaX_ZGzVOA)5SS)sZqsBo3E>YWSMCT!pf=1Yvqt#JG;sy6~ zgTMR{>2?VoIT+s?Oel6Vj|0?lacM`tu=s#*40Gs6Ac(?NU7!C+SJFJZCHA{WB1KaL%ywx+zj-NZv>OATd{LlGG-MN=4JIc*RLF5O_ zeIl!KfVS($)e|$JK&~cvE^os09I?Tst{kI_k!cV=~?L7Xfe~z1sJ0mdk{ICA(yALXa zbkl*{$BNu-MFTd0+8h5giC5kl8^%uc9|a1{6d$b1t>=OvsrMh(Yt4fhSEn?2kwKuW zyp;0$fHpS%qjGZPI!A%JgFj{4xBE~hGOuLh`WR5{YpEWlSl<00T_)Zs{teIqSB(s6 z)k5N3=C<34(?A+^%}lv}7}ZGq=@7HVa1kZv;@aCv^MP%==WJE|Cz#djUb^M>D0Dc6 zZD1a$0m{DASIMrIg6iR{aMQ*E!1j*w6@L5?o0rWI4{wWIg&N<`!Hm=K z1e{)F<+~Vce|;%=C2_N>3h1fKlRw{-L3?Jr$%}9dM<;DlBP&H?c@+b{9FF|Hg2iia z8=M4Q3QoY_t5a7`9o_NY)(H(f$fY{fkbye){`=mBrx5qs%X;cDhHuFUd-iX7E06qK zWhBJvsu1z)Z{N6A`S4n8)$bJ3KIkKu^)2~h3{j3Us}`Lf1`1h$T4#07|gT zxn|)UpzaiqepX0Bgtu-)b(9C)JAw`Tv^jFrKsBq9$%O8qKGT#x`_c51jf)lYC>(q9 zdO($RB(TKSIhqY10?wAQuV@jLrUw^{Ji=|b~mC0TpD zFhpaDq_@an{Uhit=O36$`14#}BegGK?u6!lTi<_USK0D_8sf@M$&kva$|%SUy&3A zI#h+C|8K`fzzNSp_bvT}^0mHE+Q5!y{GG+}Z`N*}hH2unnpfdi-Fc~jYx0+cYA_;Z zom!N@%b(ltty*_jI*p@mx4+PJ-HGS#xltAJeKSx(PCa`|Q-_Yj&R>KNenR~gl|GA? zUPBEcvSVfk_Y-jTKey(5GRO1p&hfkVMu(S-R0v zt91eD-xE^s&29x^X562&emsr_H{QW{I@RNX+2O> z16*29_#*o8en*kcDj0ldxciRQF(7Z;Xi6{tD}GU^9IpT28;~m6^nMytLWlGBuT3kS z!`ofe77}&;OLClPotgMa_~?wmr`)( zPASk1)Msd=V)^wfbAq4C@w1T0xO`95?Ko7gc=3Lr?J?BdbB=AF>j6chHJucp5LEKo zgg@)H1nN<4BbYZ90l5R$d1t|0RArKK=k|;H=r=v<+C7OrRHx~W-%VXfz-8??@8(Pf zip`~89;V!UPH68mXdl_+fV!44)Z*jLqvE`+(`l*_sDN}|<6g`T0*(|z5}Ek|lz^CP zPS>-2w6=Tco2)MOJFMkNT2B+U`jGc2n4y+YMMbG)& zfSi1Lg@@!zp7mu{EZ6Zf`2|#Sx|v(rX6T)$H~AUKtq+93ozq@+(-`kpSN74RB{#EB zlSHlI)A^l#sZ%QiOw**F64UykWd(zbR!t@&78W$|;nBgeW5;oYn;noZ!GSMx^a_m+4=*Xq+d z5+y|N0^h>-o>^xPbo(9aXCx1xfg)>9U*QkPN9J<*(YQ;99(_3qQ6|f&C zt`iMxMfH5h&<(J8B@v=mj`O!T19K!a;Se$Psxob1M{3)f&CNTfwu8i_Srdm zL_99MOCuVqN9}&lIR_xUGxq0K<-+0UUK zpuTtbqCbqqb;wO3mWwPd04e#Ht1S8_&ne74p}VQGo^$MjZ`;22Ve3IGp{dzvc*9J4 zS97{KPd+!W6*=dbqZRKtwKZO|qcWf8_m1nCVy`4F|7#rozt8QCQ#tt#W>}oGdP9wj zK#?1ap0l>u{38Qinb)82eRdisdA|Zi_zX}_QT@ctGAjJh7u~fox)Yl(aotW8&(lD@ zJF#zM+XY}}wREN|#Ozsi>|k#9*fpR|995IOW(bs>gOq0rgn>A^nw%_&fG$^F%aY24 z$!M#?$4H#(wlt({JJ`y>i!XgQ=_?X&8{a4v=!gNudFzuKj=zDTR@3Y-vI{5*ZGX&~ zC1BL`xsSr&O`sm$7Es;49O}v(WCjWU>%8s%_g}oguN_@KCwb}w{`)y-lXb4JZW`E= zl8d%HowW zx-8aLk2z`h?A<^O&-2)}zcjD^=GsfB$EuR)(V`ANkG58wpTqo6gXyygMR=Y#CJZj~ zvWu|i$u|V@{?vk26ET$cf)=kMn~AzLt`#mOiX#SeR6g&(_@K!0N8Q@#7f|)DENhV} z7Amcs21xdO&3Dfe5tAlex+}H`D0Z*TjoU{<&r_(QhOGorTIY#csAs* zTLr2fejWX&Vh$)Ts$P9+zXz1&G?8wDS>$&Q=5LHn1nMV4UzI(gs5W9p;Xd_ez-}Kq zIGnA*v4fQ65AM!FjIU3-DaF@-Qlw-2rfC8%@N`qxjp7NQDQ1fZdt{)F@|jzZl>oGT zO_sj2H$ZuQ^x4V-M-e$`-Mf4b1?UWx9YJj;IeMJYmAz6Qu|y=Me|4t;rL@F1crO+Y z`MiS^@hHiffRnvA(0J+<8Wo{iyA03~Eq`2NE}Q_A6GfXZ`)2^JiM?l zYtU5ehx9>Dm+JS%ph0q>O0ex6APTu_s?5{@G2viJ$fmnMPItLnyVec@k8i$ty2u}h zom#IGeKmooT|a+WhZ&G99;fejCITtCC{|ZYO3z*f=9n(Gl>&ar@d^vaMS*Ah0JU0&2yPvW0hNZ@xg@QP6qz)3+~?oj~5q*#G2^5Y+NNOz}E%AJ|NM z_s)VL)LI;7bM5*`6n;w9ab2q@XW#EUp?3ZOkg3C7-7S-dwJJqqM}ZLxUfjCi*dq-z zkw&|7^Og`|{q{I`bypz)hdWDCyc-C0Q+Km>Mn1+1SXU%`-M;~`l$;~x`NtsUi{NKx z*OUWQtlCfHstr(`j;%O#2Fs@*?NyZukPEb!2(_0^Bhw7wF5K&EYQBVt3 zheH1SdZKr|A#};f*34=-LFXsub6dxG>*|;>^H1FK1fa!J5Bp~q0_FP3xDPBLsFx5i z9Ftmw7Z`9g;q$=qKdCRD{d{neH}9zSq%J|XjGMO!M59&9=ck84m*&g$Z6A)K!E=x6 zQaog!bLC*4klSuxd+qsXyz@KHbE=L>ZoYou7v~)6j=mnm&5uNt9=}xvH6}UdX>@F` zu?G-aMKqPaj05|e0W?R95^z!#cS_cjq8_QT;v3Anz<%TR`qyASbbQ>a{vt>gC==A^ z1H&Ysh&)(m8S)l>EZbRm>{t$>RX^Fk=FwX;xc;!i7r$GGxh>fL;o?m&mt9I5E!hZU zliMpP4az_hLVNh+Fxtj_Z5g!#`~^oY5(VS-0KHKD{mWR(&0O4d3Uuv*5|!a>U2Vs&|cJ< zt0TEMf`=ChHQ~pi^8Fv6!QUqjr^^4Y{VXsTe$*&Db{I%FCa&gWUk5S0@$N##%u3T3R#}iut);%{U9_sBrlL2)J#U zW+QgnP@ytIK4gav)Qc0hdxbG$%eO+? zhsgSNb6ν(3c762Gi~eRI)gE9>phy%dd3e(-}%$^_rjts~Ic$TT#R-2}ZstPO<9 zZq(X$fLS8?5;ZNp{Nc61W}fHKkEN#`5AOt0TZ~2eXLrOn7Nb8+RqZcssi6LXxuuP6OMys{e=Od|v4_N;gx{B21te+3v~#}`fi~=L zCqV3@7NGI6!@tH1N)*Wm@qQxDW|^a8o* z+W1v@do=4{@wRop3$O`WEbU|E2{@sMY@Ej{7<}e)DDKKX`Q~`_Hh^pk;~Nbhaa$cNZ4_pdLIFGJH0J2RXR?c7{Y z%M%IRQA$(sFL`;Q+Gm~_&I`c$8?yw8&g?Yc*yYD=dtUYas~@Y)&aAr&^@04tOQh&w z3(RHpzO{7b+LMU9C=_p)e3pRQ{Nnf9Cu)f0H%+|dh3S2g;TF|S_EOGyYoxvF1=g3V z%i42E#Ogz6YLWNqMV%PG6j>yUe`M1Gp7D|$=5p>F!TN1>{KzYyS_cC;E<12rtQzzR zOso9Djwn-$8l&zn?#RwE#%v3%5Rge+x!@ALkru9l^#GxkuRM$@YKC+XdpR zo|76-?u3oHjs1k7pKB)aq-j7kTyK#lID^QPQ;C}!N)h8}-a_-=1RyV8aos5FCs4*J z_^xMKmK^<9c=<)MboU0)vR zsoVj|@_Ped;Y`34lb-65?-Fo}yx+dRy%tF1tWt9RQlK1Yu~yl_v4`W(%Dd~;I6D$7vv)wy4fe{uBmeYZzpdI)Al_$XGP^MK;OXQmaq3@F_*Mjt+6 zd?q!UGm6fOcK|#4%#6E5H1t{Q6<3&?L;bpCPyhIH`#Ylo%T8CyDsJ8h>R;Kzm+xDO zx=gAia0=?EC8@{1D@GR$?U+6wdcqM#=uH`MX3kJ?Lwi7OH9~!UR&pBlAAyYCY3FW} zg+4Rc(6k|0Ai3Y)@O{GrJYU0BTx<%~4}o@3DANA)Y1DnmSnr~qAL=_es3=ZdjQS%N zRylmoLjAtiB}~y1H1N9OmkiSn^(bBIZpnLvdXC^1HZ)1WpEr_QQi6Jb-0S`)p_h#Y zldtTSZnQ)by*5YoNXnr8^uc-2EkTG#&mT?77US64f7FbcoZnNz95Wa>DZaN$X4dY>u|V=g|V+wFa1fq-BkPOJW8Qi00$! z-Af+As7qhAkH;6pWcalPQI7%@S24CqZ3n>!t2Kz6ya~&Duns%$WOCw7~WNe z&Vg-*Ed)G)oVIT7yWPcz=&9s%`T>cfr>?H7NlXS>Wb)(YKXyRsUbuL~W*9XQ_iGB- zVLVafTB=w;5eILOs19D3zXI4(7ScC$qkt^Jvi3-D1NKAbZw1p9Kn@w$7J2mu8tBuR zx6~{hbyPIMUPEgFpTt_){a~z~qQ>v3xX6AFpmYeV6+ZYACRg4~sJML}+6Thz#6_!s zbpPGn(ZTCyW^ub$!1bvTwp7&PdUKjuL@+)F}vnTq)cfG9hiPUs1#qUzbzm-`xRq0UW3QrBe$RsIl_ z_^@sqIz^a&CN~(s=uyd>^y@F6LSl`qCLIuYt>T8EgD;`SVJB;C4izYOooH8^D-cse z-F0`6BzHX3d|i*SYFbd|P20zEZJcw>^!mZCw|797_(JAwTa4#4yP>1vrLaCw;raDP z@7;m==WS9e{|CgpAhz(4kSNe<%Nz-tuL3oGO@ZU#2sH5UV9U7>aiGSIDAe!Ui~2vk zcoeOui5Np~x|d~f&eMJKOf+;%fU@}(ZQq}jP^CU!#MaCkdM|5_Mb<1uea5;$$M5U{ z@*BGnKcmk;D}OV$VCy)lOg$~;zMBV+@Ztqs+GQ_H(|diW+G|-amQ$ z+5H8;l#7*K*AoOJQoFv|^5K8V)hMH5_tg*yp5GftMP-*VJ^t(XNecEtQjh-0OaH%? zZ}J!Vvs-FEfwoRapmWJdpiMO`npk9nh<2@~pWN_)o~_m=j)&u*J!Y@1%>M~rtKIoY zSX%^R`y&@sjavWb{}XV>r0iBC@a88Fcbe7vEx`H(k+t6HD%NB{V7R&aR^Zy-P`j-& z_Grc$=v=V(+R9vMG$gW`tS{RP)SQmk`&Y3(O;KXRFM(JeJ-ookf!jjcC@}nUKMGmO z<&T0$+n2}Il&1qFm^l*g1Is6<(e10&>%#hvlyU_85@RjTd@-iN*A|CmKu0`t$84@3 z(05pzYWfui2|9<9PY>QhuTs4UK?3Wc@{uj|jfx@x_Zkml3pYaDc*iA$kMm)qrhdV* z2{!iLrhOO7EnR`eC%I#2Ed$64_S`Mm#6b01H+LqqC?SHnBs2HhWuTU(ACIgtg^7T+ z;ya%jp?*hRE3sZ+^ZH>|zz^|xh)%pTuSQ4- zrs&5G&bPaRD153NJ^?Oh2tQjsaM_H2+qCacYxNzV&wBP}9WerSOEa0q;ztzWhb6b3 zvQe*b<4J-0*MKT=Ia^g)A8Z3Hwy#;5kMdRZDc8;%LtRm;H_$fo^uglJ+oX=!w-d{_ z@gT&sO_OeI!TOL5OB+pnXJT;>ice&&tK$X4thlQF39vjodwmwg^|3MPpC2PLr2GG| zKitN@vuCVWdpyO}{Dx z*oXo-!f+o6V;>$5xBV;w3gk|P+wkx$ zK#$%Rmj`d25^(z|e5G;Mff-~JywwH6M>Wh<;|eW~K<@cUYX4ISHQw*#mEL0fUOCz% z!5SSeOkP}f7ke}?VN(2lzqtDk7_C0K@^(1a&Oqdo(<|vaLZNPcazlU`1K1k#)B7cO z{3G0bD0u13(TavHGr)}Ra!+zN zPn`*OyuH@lD{k#70BT&%)3Ot`Kdb;FKclYX#N0rg+p>>qY*_#$uCDQGhQxrqn&u$Gegj?W_7=%XQc$5m zo2zt>BkBxwOjS6k58v0M-@3{_ijpE@pPAe^4XK%l$7vx)&}gUcWTfQ=VBWCaeV8VM zPM!A98kxr7`7bMEAmKSQ(Ro=ldDs~*SmxR59^#9JHL908@NY&eR>i=+bH`EFd+~6^ z<^Vd3+;Y$$LQ^SW02&@^ZHk#>apZiZ z?`fw@=>6UQ?uvdSH2CP>`Z&7;=riTvcJg5;!gg#?W)0ULgGTo{WedzV+{d_=o39NMpJlXabuix10aLP=2>?|F~9p%mXjL=Yr0uZMS{`vFgJ*bIpT@b?eU6Mc3n?>U7SjCTCuL z((<>*2U$;aVKjY*TS&MmXPh@(+jB!urhcK;y3iP?+3Rpk^P0kfMD&CI<+Dw#< z3mVWs=DE$PhJBDF-I4lq^cT3RZxZSL{vNd~GN*20U_1=88$HJkezZeOvAwsOew6^- z{>6nuN@1v_mN@Ut{CoeCkJ0~L=?ZGuvPS7T>ka|uTpum8+r4~)Evr02?(0@;aVAL4ltsN?*M6INDGlm9Ht z;Kc#x)L*NA<%}OpOsBuIx~&c^4?oYH)-^$u3mgI#S*qaqi=CeewF&{{d#z&mUQa}> zw4A|xuYRUtplog_{EhHXVKsRxr`;l6ri10xU&AL7WBONw%mVXH<09QE!2b} zc;DyUU^>;o^&6rJ_o1VY3^2bY>*(jNf4YWv?xPd+=PR6f2((F&D@x5+y+DWlOcs^@ zE+QMTKfk-P7}fPmjo$vLhKRe`aFor@foi)>we$ei?}r>dEtaLe7m*`VpY<&|ifo^o z>?nE@jH)z$_8%!Y2h=0W6baW(L*0vMFK3IbsI>HmgJ{b#6fyoSF8TRmL^?ijon_Mw zy}zn1nEI|jdDd4dksltlvej%d2c+hITruseEDK@2g!iH3;#3PC4sdty{XA=oFPAUE3&_~q5j@J%2Ms0!w<|a;G=iZ=cWvsnBm%q9x95EJ z|ACK1_f%8_bYp*?9q>|>;Hfi3{ds1t*Olo&dun-MAc^b82R3Q>iI?-`f%rVMWnH=o zu#47us`+bkaKWB?Ay>x-j0 zn@12cKX@KU7eYnXh--PkHy~ZH?)oiq0KV7dls#nT!zfWC{*ZhhP)w@kGX_4x)FM8i zSUYZh6OdgNyUqng0aes=gVbg&PXG$oXnaMlcL_|WKaq};`3qlFsI!jU!u%t(hi`5o zzI8{{Cv`%=Km*vd9gVX4Fuyl_!PKrkN370?y75gz_=|NM`SvxN@0J9e@ngEZnjAgt zzw5li)>hQ>BXFnmf_9+HofZ(AANXJQVTWYvWazF^j{JvKXazKAL8qX%9kn zXC=1}CUe&P&|CR|UKgZctSn<7%GFn>@^*kp+I)=nzp_f7&!JlaQB6BM53rY_z@jbV zOZdLy1x^ONR(1IZ#1{bxj`r?|UK^vdsGeR>I6u52%{$3d4o&K$@Nq)_pdP%1&>w z7GC8DRF~Ts%^eXyuDZGX&c+M^4sJ;ft-OP}cXU48@>&7N3p>O+qj~D-|EpY%OOcIA zmt%YhUz%>sAAPzV&u4y>OuCNYu-~OU8D%GVcqQ-x+k)(yCNZA$85yl18@s32zFI5m zDdOFPD1s};$qVkG@r1o>?{F+Xlo=#Cm{DU36luQh@XA3vZf%%WifSN`pAY0vO`Q?L zyrMnvi5y-aF|^%NG6TAQt4C*TjD|mAmov}ZjR2zU4WEU#Tp^{o$nK3P8x0IIS38xJ zqal5|zx!K5=s zKc=ZX&Aa>rRkyu6@L{b2P>xUp#3b~Aa`K|-B~vc%6Y5|)E=fP-kEYvSNxgWw64(ir zX__yGfEImoWndfEj}N2Uo9{FRmg5Dt4f(qluLV-5vjJ{_J#@)CD})Cw1j_R+1;ukg zh?T3=J)$U&sy45E7~&_3#|iJWynW^tl$~5o7o7S5?0(a`=O6v$Pe04xKm7VLfsfDe z&%N?9Ks-x2uv(c9$k7M$t{3J(UF0jH6|eMxUi7SoA^j0@avxiZq%#p&c<1JmqT&SH z0X%=s%T}nn5SnP!{tQT)nl5cFDn~VY;>15`&`^bWMo=s`quyTAv*|i9XmGW9sQbw` zsN1I6_9XurpyXVA)^ic#d7#dQ{&Hh*=B?7Oajw7xSkI~N*t#8bKh z44Dm&?vSd-3(QH~a?@LY7pR{9G1Mp-FR<}l)s45U@I`)AlhmUF1RUR=(dTVuKvU{0 zID5$j$fs=C>L+>lrGRd<&^^e$2!YK#vP*&w>bZZ#!DT!Jbu6{0zNl7;7zri?_M-o^ zhcUT$D6zkk%+n8=fZOm<^J|9-uz$V~y|h~AIpuXaWgRT!$p@Ii-3Rt}XP}xq>m!x9`_p z`oJ8>D&8@VpCtk{-0u1L%Tt8rZ0P_e6yq46X06ym+7rYFW-%C+BhV zgQ2SLVZ!PR%s)jMd^lTphwF#r*l}4i%hqB381mBjZMS+jcCY&~wwZ?;HZEEtX&$>B zF{w9>pW(lSeOHCyo4euG1l-#I^Z3FlAbS;U3%hy^2H(c)NwP7*3mjt^yi2`@t^d30 zOZ4Wc0PCCJ)Vy)7e-qC?XEym&2?1rOiv(PHFm;c1?{G@j?P_oNpH zseFc(zNg>SM&t>&dFQ0>)9*m_z=exa1%p6Z=JG)H!&U71ggO)15pih$;J#v{WE|RS z+s=^_EuqzMy40!aJWyW0%#l&q2XlKAtJm$`LBQR&^Bg(CImi7%S3#x$D*K+ZH;MiU zMjh*9x6w)PC-bz?%KBmKdsj`&Y`CBgB!~5{HFUk;Q-KmLke~CNU(d-YpXKov0VT`` zS0nEPllz*pI_5tt zm;IBp#CI2v6W6uQn(cwUTKm_w%0CFWMX~nwJD&s5@kE_p9L8rvx13wuw+!QzA~mq4 zPBD6V3lQZ-{o_*rg8()qtFID&nFK1v(zh#Yn8%0fSP%78xGK z@(;*H*+Cm}w*tkY_0);m|BT;3+klU8%KY1?@uOMc`|@lcX72BL`YIP&r>F4|Ms93i z=~=;L;YrW`)WaKC zME+MC2Pqc(TRsM1Jlg$Tk*tjFIUqgTDq>-F0^S}!`|h%+FJzSVhGufsyIbG#4|+7O zRj4NQ19j<)b z3s)zm=A!|XCxz=Wb%7nItG-ddAIN7eZ;+DLCE!;31eB%k0`}NgOm{3d-iBC(UnD(b z1hH{=h(=}5^awbOMy6yqRwq^;q$JlflMW*nqjwt5!{SWrisX|YZe{~JcA<{)^RGa@ zv#r_Bb3gPr`YOoTl4h$AR6` zdB3HfYlowz^>=E`L#H|RqD#WzOP(Cu@>KLf>VBX#rmso$wm`q<#|$L*|5Zy5nF}Y!8Ex(}G2hxb=Jl+=>yA-Ye(+DVOAD>v|s;yn!K=RPj$SEzrS;XGBj( z|L1iC+;XxCBP#|m_n+@-_F0Mg%uYxY-Ti^-S-i&jj|mpgcJIUcq%ULoA3R zmxfWZOC^B3_@c&-w*^2+4i{XI;s*WoZKmhgPf*Wg(Q~HzOM%f)*Bp|MouhLl>svkt zaOb(9Z~1)p@g!!~lUGkOW^6fIdtCL_um$9-$#6O?BKop*FZjE(w*n}h3CG>xY1|j7H=^<7!2#wUB2S^nygf$ zzuWR$KS$W?#J_qO5I+@bi9h&fA7r4k1O>YBc>)P%@O+sk*5@;C_$srNn1e{T0>l<7$hqMdPUU-1J$&ymCVmuryn8kMJ2%F2v8?d zyfcsR^bte_M(w9JDClr-w^?3@bu}uppDXRk!}dwd(B_CIZkRngDy7|i87+anyxDpQ z!}X|RtINx_CtQ7lDz~WA7IkC!pEYSNwjcXtvF{JEZE}&=4D2&5lmT}MnEa$4F1eDk zpQi(?MtUuvmru4+BcvDFr$>Ubr|+Y#*Xv>mUJ?knqr1C=Gk-(Riq7*kx#tnf&p+_f ztCWAvGx^H1NBdGS9v^CXnStpY4sJj3c0>G=8&Lga%e&qi7(XYqmh?UM%L`x^?NwZU zDwX>^U4FMupFPaWb6MB*V&CGHJ1DAssZquyZayagx4`vhyiOw;zHxM@C@}&R32S`R z)wTjst*PSc<$6HgO>DoBWx+dMb7wv!rN8QBPaF!dU~S-8FV>dfhIeXM{nfCSa&pMV zY3L0d4ALK(gGq|~zTK&R@%|##5W0Ar@=hN#jR`;B?L5Ud^Asc$4O~ ztoZym4Jhq7r-trbLqSX4J<(6b>PLvECrU}2gX_N1qH4TW0>$BJZ|5M^4?f(>iQHh$ z%XdZJqgb={lNxIO6dDh&v2!`*2+74amky+K+Y>H6=b?LZV~D`=F`$SG4x5k`WAnK~ zBkS}sFVr92efNw7_Z&vuLeu<3z7vS47`%JJiQBgYs>N>8L%YvHTWW@5Me}tSB>dQN zeJyVsk1eo=-m|I%o9|hP<%gMTVN!6Nb}5|+WD05U`!cN0BD)9WZ(kO}IY-VL8>EW; zOYV5r^lsQyvi@cuuO=Us^pv1lTL7DIewv0RFc=||x>k@rOu&h`2i_f+1lm#m zq;uz7vFD3-$17WZoA=*I`Zo-EZq@sp#kBrk6sIaL7^tpie8y+s6^`4?!2MF4{Itz9k}xk1220RDlV8Gd(L?Qv`F3r^3(b9x_KPjLZn+dFvdB! zq)m`{g`1azT1NtR+crJrnUAs1kJ2h7f~aNV;|XCitp6o>as3sOTCRQt%5$R)R%fw3 z0ey?hQy(qK$K-_?u)O&^hQo-4eoZ0L8wt4FGuA$ewH&*`sWU4%ih~cQtf-J4>4?5A(c-@|()HLXOKz zju_Gk3#7bqpgrL1`8BL%96RjG9?|i(|GEz-XCtnyuGQiGO)z`Yr7XJ_D7HF>GEQ^# zD;kJBy2^AAvk!YkyJ@a#v3_5pvi0c^x*R-xE+k#=_FsIf_bH2S&`tpP*VRxT+%h1~ zr!>mOdQqc8s-s2IR>LYr3p>`Vj)v|9rZf@(v(U?5n0*2Y@O!_0wDL zpXWHHkM%Equ`d%x{q^b{4`cu_T0d-0*87MmLaeXsJa`U`#Qpb0z_& z_OtNftE)g>{_10d09Joq>+sm?$q`+SeWx&!RqFZAd1QM>i~bbn=BuFg>E`}7X}ob! zijhK4KsUxGAMzs7*~i8aCPE7hJCgqZ#m3V$D9s7TW6;*7Bm@=fTc>VG ze#&9|nzfO3cZz>;`@kTkLH5~^NNXTF{hlpQQw54}YoODHc%TeI5BEgf%BQnD z{l++U(}_f%A~Td1cl?{{M+0mgorJC0&E^Biy7$8Fhx>rV6sDyoVS0gNF7j4^{s-9a zwhKBqr~$?2k;$Lamw?<@YT{k=9N4Y`3aQTJaI(PY+@c z1l*p>UtFw(P+OzyE(u5MJYsd4^{cQ>z@+rta*`9)XN-}Zw&`v>0P)lxvlD`)K-sV! zG0LZbBF0BoO4 z?vz!y9t3q$Sv7Zl-9>FdX|=Ts&i)c5_`N@S0|plfygTNm1QXAjvKTwdIr~m~Qn8mC z&x8&m_F!84Zq)ZP_;^9Q0_0$Enud`n?aNbw_y~jxzl7yb- zdMCKT;DM+1WhXHIR?koY{eIeYU|0(8>xCpI`RL zfhk#uZ_84#{=S0?8S*W^Y_a<`7(JPv5D0Tl-!I$@NWk;C&t9!OR}XCp>+v_&@$?ts z;Ezz(CyCrVSLmI1qouzCbP7HP>$7KKD?Kw-zo>^kUb{0 ztp<8Yhf?oyRRS()dEz@nJ-B;SXTu8_P;O z!(nnzbmrw%!++NI5^&m%d~CO1sH5-L+E6tGbH@d*d0Sz;n$@BvDB6P;$I$#|1m~{) zGtUypZNKKHHemet)NSh=8xHFO`=!|Hb4`P&JALW>e$zV~Jgs>;^!h?VG(+1T&d`{!9d{28z1iqTn8ff2{{%PngQD%l{3O%J1)N`u71jYu|gVk9UE* z%J|mvr83w&2-%j&+hr4QdQFpcW_=vHCIq#yJ%L?#rOdLQx4!|sPjJcctysMzCCcSq!D_58M?ofa=D>@qKvW97Q~xmyI^4eR8I1Oa+LMe^w$vP` zzsv~a+pYq{Jwofh`N#n?&iPM(yeANgqWHW0im~y|&!{oAE5!@^m?`)1Z$_++%p^vwl$8&J*}!!d z0?V;Hklvq(La&Wtfn2oc?8c9&KpA}D@uOCRV?R~k<|O1$kz{v&+V;Qtd>C$$O0D9Y z<3)KiNr+Lspq`RCGkPyYAxp)+)tk=%MYO_F0f4a5e z8xQY0k6vMh9jCp|k^}Z0ovDnTTiA0K-0&6K&cjE7oqunyor#<*z@Bq`O_1MH1xTHn zcbcp{482RT8W!&7#bv`Q&41)pT>>^Pdau6MPM{QhiZnYv3PmPC+v5DOz7A{!(W=~t zPN=Wi;e9Rn3y_c1JQf{d!{qw+A=&&Guf$MD>e4gz7;oyJk-Z)zlw(haUU1qR^9=3r z`SVTMj{uazukNc;bb+`F7xkwD>xUdPKf@|k)d(m`;_B_ISD=y?L8IQ&dQfAl{WGfF z5VF?scTyjj0d4ov4>eJ%5&h6R=P$*|ob}va6}0{dB3T^%BowQO`kJG3?7nOVqRqp7 zCQUIMT(vN%A;biTQ$AXsq7>l|j^;=%#`uECZi)3v!XolQJ(9AI~A0SA93*->ahu& za{JA*KY;u}uUUk)5>31j4ZCA;2ZoeIwW=R`bIzCOU!#ksfP67E2cT34{TsH(^8bXE{CR`pds1Or8l4kFU zy49SEs*x@X)@moW%=QE2Yo&6+3n!lEv5S-a%DT6~NL27_V*nuL@X&9?iUp{&a?S0H zTIq<=_1f>xMpKxPUA3P5I2K}m1h;g=<)8`0ZM{2|VEc{|9kn!Y;297_c6Ex4YeR|l zBUMS0yUg?nQ~6YlT2PdP zm=iyyrw$WTKc#v7>wcc^&DXKDyErU~jl0EZxc$)pbO!7&I+>}7#-gJG6Uhfqbpo#c}WG~`6JsF9I)^}wDaM?mUtO7aO`J`SHmkb?&SATwaW$dqa#*V!ylj)ZCn1L z1Tn-?f91VVt{;uBmwiTHoPdU>Z!b6GY(@;4U|tg613aIx@9SMnd}v7cf?<&lSD)Z< z^he)y>b@ejvh4N7YlV0L8Cz69DaP~DFQx`k-@)kHeV;@G*JIyZrL*Go*GZJ?(lWRz zP!wj*U&x&G#`>idyfg8&$(lgq>&%^pZZ3k+Ltn)}>wn}(}A*a&oTbB z_t2Qmp1mg(s__CJyQd_Lub}4gskCj17(O6=+g-iRj;BupbdOZK=6+j_6p;an|Xa$f;4(Z6ajxx-bAuf9cMs6Xg1%fOc-^qR!PpG`NYiXdXXS z|G>o8r^-CTzK@uDclQYw-aKvMpS+nypGGvOJtp;{rJ8f!mM~9o)ql!oZAHM8ZI&9E zI~;CJ7=MXM%}o+k?!xL)$?Cn;B1bX2R+HW59NB=?Y5%Iw$P7POf(C7V%i-R&p<2VF zAbgS&0k<~(6Y*;t&@TFr_b(4dj0fWxy34V=nqTo?Wn9QrJ9y3EFVwkQ-+AQyb?> zqjzKZAV0`mZ>S1?vYI!>C)@X6qhX5mdP`Me^jm zMARQbs?zr8f#R6U>H|A`fMsPEL1fP3*bR45P3UTfWr{O?^C%dxsBr@^7GxloTqCUw zY=v5jl?$pXO@UpJZ~mu%E~DQo#x{vZM_HS@u<=tn>n z=*#P>HsjcFsm3Sw#G&85LJEn+9DBxTea+2~MW{>Q&iEQ4)LjmtTvT3yA^Dq?e_8FwdDoS^=;9UxKTk=;>}LG8)k@-M{w zAL4+LL=Lh#8jr{QxTdMfScMm8uW~IT6t>*E3@o3s~{*PPq%RtWT+%RV@idq=m zxm5l&K;F_;H5tZ}$A^YL9GyS%tr87X4CJ(A38H>xR@1;echnU*Ih|@J0iCO=@BiuH z^8BN&@w-cv_U}S1cg@uG4r-zSiCxv7l>JbdWms5sN+4nwuRbETp%f@}?i)ikg<$Jd zK3($jtPoJ1rukMUPNEUbCUcpcRfzFL!h3fJoq$^^W7BbM6!oY1=Shw|BH)yS1ZrP~ zqG_ocW(jwB_x-@dvtD*S?!5ZIcR9pUIW%vs~UMlZ+ zA>b3L8J{I)>uv_}yIuIV2UXEX-tqL)1O)<4`B$1^Llq|HMF~+(%n-yfG)<7NS^?xA z30EIaDg!-PAUJ-*RWzu&QaI$<3Lpt?5m6JZ#m4o3_)26A8+wZMBed>dxWmoucI<-Y z>!?z0$!T;6iysn?I`+9s^Tgq>^X|ZW+>Lcch;BS5v-CAj9PEFscj=|3>YnC+hGmn| ztzx)&n5gurd3a7)<9}Wc6U~`3s<}qcxlrZwm#%Ld`OHp{dV<-_osvxgzeS!vp`5X? zrIR`8SCl#;vYMCQp?OPWO!U2GM60@g@a_NL_AWx`^%N{$va{KAU8nOJAg-r(*HSKU zpC2XUe&D*+0o13Xow;x)PoDID{oX;k`#yPLEdP(D+|lY>B8(VM4t{?B74x%Dml>Eo z+K_<;&DRdBxag0V#|~bs2rxj51H-Ql701J~b=YK~|2C|si;`kS2(#t3Rq<)0jy z{M8IEU{uoDwCguw?cX49Fy=X8RBqcGaO((Ih@~!mv{VZX8amgMy(a@KT1 zuU2)h08l9Q@5OJV0{QY!x@Ix~72o=JE;#K1q$+;@bxOb+D1r`rl1UgZ0M%K^Y@*c) z$jcTmC6(IHq)Nw`yZcWPaN0qhgO-^<>Fb?Z?t2{d$dbk`Xl;Xz^53T#-6Vl>{e)HY z_*RGu`gl2V$rRLIkB>aK;}B}=zOmOaUkC<|*pECSTB6y_XS|GzAHhiA*89P+5AXt3 z^V)xO{8f+U_+zDZ`y?WI{S@C6c7wA%vp<&JTl;Uh(e+aLuw4?!*N)l_?d9=*p}C^Lo*$J>0-=GTBVw&>!8r`tLEo|&8I z@eb%wMa}$TV-T5QYpI&&j2AqAY~S>X4%CZ^5`0U!{z&Ms*!GM*a0m?q4Xj)i&l8sh zX0XK0Rbe;bjf^<8eEA?$Yut_-d2$|wy)dZidCARFLIbytZ0Jz#LyUu`R*&m`Mjb;> zvU1!$yB;cR6a8K|%5(FcmPpfQ|uEMm!3JZ?>{XxXk! zsI)*|_`-a1s8eYiy-cM-i=V@K)lKVB4Qq;VDRLWPsZ|M@FSUgl?^W;O&MijMAJ%`_ z)4*eop*CN|o49Rm|6ITBlin4*_QH^LQBT99^aW~@Ex4vMq>sh}ZLe-h_=55#KHGZQ zTR|T?s>Jz&IHZ~p6k7;-K%G3PSZZwtiJvIKYc0}{d$^wX!%dZV0kPX(*JzBOl8Sdl z%F;hjFJpU@qAa(+4t&4j7CP_tW7PRmA}osi3NK(7+)EYZ<^e%EZlT_2$QGcSJnOi8 z1%+cr2W|hz!P`BDs9G_SG!8EF@ya;F!8_6Y()&fX@~js{Y0m+?^DUqW=mGAy8pjS$ ze4*Vb0hHt;FPW-D)GqQVC?r+|$SukeYk%B@&U^S4`py_&=MH}IoOc;0_BZ*;E6X8K zP1}FT8-5_JS-&RJ3DdJAw}&1@S7Tr>sOP5lvqHpr=kdu7d>=it=VMUF2q zUI1pgZNZyoSeQ?lrMOoX2mldAKs4teFhh@^b@S-x5@9%p*DzIM?8JWfWt;KOe|px?#Qmiy@uaOuW} zs|^Mb-M-wGl=&N|_r-%A>zx6*t4-Uz-QB2<(P2IRfd&c^usC)}gsW{r1N z3{d>+!$m!XLTTBf$MSbU6JzU1vT?|o=;PT8LI$P+K{sLTGdjWbYH z>)*%8kq~pQTIZ77NC;y+IJ7sPg6dbU{WW!jjb@mS9~1|O0L4M-+KZNNsN3Pjwb$CY zC^ty=k7?iWGk6VRt@NU-A_S<^`P1Kr}t zh>Rx*NR#g?JA(oBDVj&jcUywm&OE!vQY(Q@IrHhkk=Kav`so>y&pf;g*g2x5Yg#B1 zi^j%NQ&>Slt{7)Q#m}T?l{kqct zD5$y8X|6_mV|C*VXU+(D}YqA9Wn@)xR|dsH;Y)N+NFt&-Z69pST^yQ${WL z#=5+q7|6McGS0X@K>H`nL_MZ{|Kgnk7yJ$94w9b}{Gm(tW#?R*fZwcp_C&8~lb zFDvfSLv1~Kyg;G;u5T9;Q2&9vA(vgrsK!k>ymSPsFY0-VdI|?65VJ3Jd-j`X&iI#Z z{PdOUX9gPm!l0)&m!}_(yHR>o_O=HE+1|QxT;?w7+9-Tz_eCy$D@uv%%<*P#MI#>` zm34+0|L3>}IR0OHru^;LIO9|85AFXe&tdt2g6A#E3AjBsqz1Q3p;E_B(FX(u;PDUJ zbA)1STy1=gat9Tj1Ldx4sjAdEG-~yIsnq#BK-ructo-`}P~A=iUh$hljH~RU@#Ya6 zynnQYA;Ch7E6awKkFDg`4H>J%PKBbf+S78!bjWzz^X<KoobF8NquS*RtB;< z6wUUy0#WRmXSp0!ht4)IN?X~u6#bFhZoR&OgHuztd>alp1eEE7Zo9Bb)RuqVYyZ49 zR5>kPS+>iOfO~HpE7$QB4Qf>+e)GVdLsipM6j*QpvHs+|9b54OsC@Ucq|*{%X104n z%cJ*b;I|C_>^eJ0n6}f2WNbt|=QT~&Br5_%uUf-Va)M)LPCPra1M4&2^IFhC9VZLK ztg<(=E7}nacj;D%tONmPYiC8SSdB_dmp=EY`i0>l<;`+nI*CGhG-QQN;U0H?(gE#mr`1A}_+mybyU*-wPE-W22!TUg5xFP=R zn%_`AxnzRwbQC6A@%%mgR;Vv-!q(JK5ZHE^De)cUhzS1532z#qFec)H4DG#m z{*>46jOao@?i+l$XO4%D88Nps`37&{iG!huMY3)Jjru@SsLy%*G7~Y5@9eq#&;l_e z&VPUYau>?gnHXK-whyuHPV zisxn-h!SOgS7r*!H>RI&*OOW|f<6xq_$6M&>X%2F$WDQJB|wQJQ%3%rMeRS=Mo2aG zLi<8t!lk5oJfB`>(7q2^FroiRy+y_Y^%S}7xv!HBls9U|td!S4Q8r$C&=})eX6lyg zk~)O(^w5vQyH)PJOu)^v9~iKB4&=HU{7OQ`i0O54!opI3Ge4@kTIb(F{TW~Ny0Uk2 z$5kyKYG1_LUyl3#C80-ZA19pv^1LU_qp$uMZ{b;2TDP)uPZkz$EFK8e$U!`DDMXB3 z8%{d7g@9YPZ<7RB5NZ>T)IHn%0<|-npA)u-Ax4bTolQzKj{PQM()R8N3Vv01Y)4cg z8kLrmP!v(e3)nE{z8K}9k|UxrsVh3*%M!_-rI*^#ESng%O%lt09SjvqIDHnJ-sp$u|B`p+|eEnSwGrIqFI>#C$qJdRen|th<;jYe(Ukn z$KVAnm&NyKZ{t~qKm5UW7i47Tob0$~w0*(^?`84L9hzB*8(bnA?%ix;qCtpF7Gr zJlKYM-*_%alfa?3l#}Fn^@fmsd0@Oz`XFY%V!Pn0^Hn z#NKr6ILqUgLP^IT@d;OP>%$;y`I-aTzqocHkS?75HehrL4awRc8#8H!l7nxF#JQW$ z_exy%T%`c&F1ejA-1h)glwPW)+T}seFCWVNje3ZYEoFQ^vWaJ%5=WeIn^X6p5$z}Z zju)@_Vz7F99j>mN@TE#ogB#)(bhLu>q zr171leOg8G>A*&zzPlO0KwVZC_toJ&RBu(;{ClSx8Y_)>eQR+Ks>C1jS3j^3 zF>5alUD)*y>SI*b#47@@4X1SERQ3TSG;W{rQY`+`^~FzQO*5tsXv!KV!9Nsze^oM~ z8ZsY9rcT>8E#H7f=9EfgTlPYC*pg%WPAEXqyvIkVT}z;BlaMfZlPM~uotE2cIst4Q z=2AY>)sXeM(K4lYBh;Jdn`g2L;gx}NbK86_ZxqmdJ1QQZjDzmx@dOPqu04S!RJ}Uu z;(x<%*L`(KTmw+kLU!L4{!YNH#_`v%IrwnuQ2zLyUC{C5ed9{6TZnn$+m-_bngrbU zmG^fN#o>E6{=u?FKg6(`l{7f^5y%eD$F&Qhpu0~~zsO$|$V0@ZGh$nS81a!`e8dZ= z9>)`GUM(c|6|YXD*F;w2tW-2lb+X@!OpngM?6faqhfbPuMW| zQ~LML7k~LJza)gPOus-)&^a^WfIKQV`iwojJQXqZ7r?!5@8Q?2s&p+nZyzwSvMsr7 zT@~tg+e^azk%qxF>)iR{KM`#mpC5P9kPOCE3zL3N_>qefi$mqkadeU?G=xsLvyZ&ro>@VuayU z+Dw&!c6*z{>rV^NzyhbL!?#Fiz+ZUViXcTaux8m@I*LVYSFn%Q&lod3705nasf2M&)f{hDVX74y&U?ECzt%iHlCP=^GUNWZv*STURZ zUk+pbfjW_yeH{Y5@ZH?vUhPe+zcoe6Q|my|CXO8@_V`^R*2iri!^+mqv;gQ4o!vJg zdeEOa$2(iEVs+)U74=@T%LbrXGJZ}hkH;^Bos&PMXMP&<)RE$Ge(yam_+xQQYF%4d zcIscau(9*b=4ILh96C7s{w4zrw6z>KA6*C3U4lj7pblg|-2hwPN}x7(?@;QeL*t44 zyT=n#&_w^5r-z8usChWyd+g`UKw%wO)Jeg32`EqEN4mcrg^o{CTijhNA#;6D?Drb1 zj%xJP^PBWK{(pWCi$)M#`v!m_opEl>=gfay$BS1qHs>?7<}i6T@Y#1&g%5hA!bxAX zgRt@KTe5PkZ8}h|MptzlUyRKMC3*k51qXnnxG_Iah@afj;qKcgVEb-vL!j;i2tb>Ohew z)88?~!O=Z_TkA@Vp{iAdon^2S7!N(H_dPTKO0M<^)1QHu9K4ihe28BMY)2Ju@nSKc zUVD4r%&`)pmXb(EyoG_ukO|z}&EppW^3kK4_C5AO6T%&zI@{MFhEh}hX$`C%p=EP5 zrS8ZQo_XsvsG6C1{2g20#m5so!$W~N?@PR&I9IMw(Gs(@yQ9sZeJ)_0jjkh_Q`ejQ z4HpsP&xJv6yIi23l8X14DFUMS!tF~$?g2R``AOSJd+=UT+w}ZH5ghQfA8nAuLG72z zJ+(o+_GACEpbOXi#ZhZYMu5QnG`yhFY2C>KQ7Ce|+db>>l|Zw5{*Wx64dnA1kIc?{ z2$WLCrYAQ%fm}BJxZq|!V%;*Z+;OrKFHqU6e8tKLv77{(XO1P|1!$M}xBuZojM9ZL zCF};T^~b)yv>M0c(lt@kZ`d5DpF?#UZf^xPd7x^OXgwO6t9ta}lQZhPy-d8e)dSU} z2A*TC zl6VNHr}D1H&2a0fP;3%~K<%X5F?`tylutWT`+u$l;-KLNhs73XI{S_4SD8AX zO?odn{^~3-+b8Z$VSOwI?-X{%Txx`w*^#Jo z8ry+ldRyS~*Y*F}r$A0v%C|`P8fy1`;%W>MES- z9y0d-4}W+3T$+GB|C(`ndD(}96Sl-oD{BHxHPz6M{1GvelPeP%UjorX?bl*_H<0CK zBSo#~K(y^yRye1Gy7Uqpc5D2*e+2VHQjLrl4h`!3igEDWi$*l1NA{cbp>Zn%^}M>W<+?s(}|-o z7`#AeYgsJ+Cm4PXX$x7ZiyQJML@d0?dl5>P#x$au0{nwo`{?I z{M{W5I{tV!=v|0dL7Q4`6l9=|+DFO}Bn$L=rl0m?)DW2czxs#U6QMfr)DPu>DCiJs zPL+Q16{t(fYRvWt0VPVb{CSTM4AkrD+4dbogCy%>$D8)3^06&j)|v_A;o_;}NN3bl zWc+bf*90$+U+^g`pL;$4Wwv-jNpd%;Of4;$%@@Vy*ME(=i`FmHnZ5S8x!Q8>^$Ba6 zQ(a>ykYiEF=vadBk5Uwa)Lv@71G1W|M;)O7b=j%<-egb17m>Ks=i4uF&WBT7*GT1@ zb&Gji_iHm^{O;O2PgVn{5-sOX?!)R^sPPLYWA>|}NwLD9GYkVn7o%PN={L=h+lleD z5%FlG?905P$2EY$_|bS~^KQfpG@jFGWfS;h+S3kPw*;#Bnb1gbEKDvGX>vbjiC9AG ztk3E!17i3ss&T^%u$p6=?+^F^i&pfM|HMfGPHTF|jmj+hyqkGt3ZRiCGI@Q!;F097>Mg~|CC0`7ES`XuEPP_`DVneefIxsS}>Kc-gz z#kfJ%Do6uj31-n=P8z6tS**5i(qb5t^1S$AjH5TcHEx>J83QUyZFol-AA0`d+(P-0 zH5@&fo`$y7LyO*m-wh&DaPOY!y?Nz!Kz?Z|urk09vHJ1?;>EH4ITZ6Jq9xlh(RhWL zrfBp4&@yD_jVaTiLqgzvi@{o;)@)VU_|q6oJFe({xULN-t)B*?no5D@Zf<$F`3H9% zI<&kjc35-$7rsS zOUkXl@&!9_{gf?hE+YDZ1syg%Jbi)?!^~&r?lUYzev$q9W4S2Q#C9fx9cTlZRmPgL z_I{|@dY{eR!di|!6zh~gwE{|usn3J+96VQER-hcj(;phxk8f?SYy5?Jj^QMw9KNHj zVDzCzk8`dm2<<9J&4$SzUCf-MT`1ejfv@SpFVrufkktQiE$UQ1xasz{S?F9V+ji86 z8#h3-o8G2X&f!3b6FBN>OoIyR%V%j*g3$C?%8bx)9D0?`u9_F443v(>KCr3f;7FNg zm9|kpYK$5H=W z>;a*cQN&tFntsrh0rg6+w+!nbpguogFiz&+H-*YmOGLHN@KC3?=n3w=Ms;sxE}Xb+ z1hHMtUoOgE{w?yo)}>5A3>Olekf4@653dc+dK5P;8!U(Ef3Ev4I^Q zkS$~vv_WFA5LX)?nKRr708h)=* z>KhP=2GYizwG-usmg&Vs&b?Zy9u4G|xJ^j$>~<5T0obUVz4DXSQ4}S%a#7j$U%{p}VqXD%YUv+=11Mw~kfh@W{w%++n7Q(H|ixzttMS3Q8k^S$2N z5=c9Ts^9W;gf5(h&QD9*Z!8K!<1<>>=bv!b89Lc^x00qYK3E!6IDL2vNB{ni&ds)s zK@|p7pKh1zM9h;WHx8-Y{?|OAff%1GgKrCgrfasZ>x?mKUA@Ik*qZYk-0K^TS7gvY zwpECv(0(AlS)XK07bW1foGLl^8OzJ3u9$pnVLpSHucTU*Wt>BE9v42hRyjlPx38k# zk6UB&y)>mQ@X$BJ`dpKkCb1Q@dq`a)-SLO&Lml~JxJbl^pQqRSU=B$4t&&m;RUpqJ z^lHZYK}3?RX`~Wy3+8})^VC%8i7jX#Y1grUfXhJJx3o-9#0JP6-*At8{h(StSg>cMGHUni zT+I7wuj_Apk`U!NSrW61?;h-_91`y5BC_G%i3*5eVewYee~nu z)#J`n>YuOr&fmCkNSHh=p||F4I--9|+P8ZC2++uv4yOrZBC3?d1c|r{FJMFcZKiLB zm^S_|HJvziaMHP%$LE&-Rj0M;iu_MBVEO!1f%`w}G%!7qOJ19LeGu41Znk{oOMp^4 z`*7#3(=a=2{(O5(HEMsn)9bspJ$Jr-{953*?}IgBtTe0GwSgO_K&vI$=-=jWy zhr6izXVizjr)B#n04?HC>r0OQ>}0gi);eH3>5O;Eg!tKDG~92qRwwBKVrV}$DPLoW z7r3yyHp5~K8abkqp4+Ly`Tms`l7{Ya?2J#UXZt#VEav-)M!@*%``Z+Y`ChrAkzzyH zKbHM?{tuQ_hjEQS?A{e{LHhy9cwAln(e^7E^~^|QUFd|S>hevIo0+IBlQb4EgVo)9 zI~?+4nE}>Umb&_IZntv<^y3+-8xB;U?!-@dXSK=L_|usnJ#JK#Q$SN>2rvKY{SPXA%c z#=WTDw%suZX``+-0T+!LjIWTkujsP5tkY8Yb4PZ@I2S{ga(R&?vk(nz zH+q~uvXt{4&p@H162y`k+`)F=4CME7>O%__0_DuW&LCMeM3GjI9iZ{(1DI6q6C+6S z@_df9*)3xCWBy)7U&AUdYCk6D3=0XNsXidL^^m?z@baF~E;XrkiQ`bm(~6t-%X0Ao z-apch>^4N@>FF2uRc9gkYq_0^{Qg@fD9%WtASTpfN5%qa=y*^{~8WH_d zJ>EUHMU5w;%dysk3)( zVtj(s?lL|-+mD>@v)s7u{2jc&sx4naRnmY`>{9xC_AVM++7KqFV9UWTt{xT--=Kcq z^gX0&*E#qlwm@9D0XBys4eBe-Qe;^wL21^g~XyBxPc zJzrY$*Y8Y5<6%Y7mgY)8HT7`$F3Z75R!h}qkBFnL6`KdvI9ns8@$2VRo9^NH4ewM{ z;lHB7`Wc_aPx4_xR$sRu$qfzr>e+>--2;-j*Q9{}#+OL_P!sYg_$!d=AFU?dK7s12 z`ZpWx4@PahlkL7)BvnL?#bDX4WvI4Yy)NvpG z5`mbL52|GJETF?F`Py{n8|c$r`sI)uhU+U1K!k&$H85VAPh^~AqKPSrOLy4|Aore| z`!p~gF;AR}n7O?XC`zgobx&>rYk1aA?aCF@p|gH^BI7aCn(kVZ(tHJ|De^&a30e@Z z_4Q9+odX&O*3eh-B|*FOyf~w;TYe;Q6@^!^RwrjUf9Qc;E!8iWQFoja+JN;GN^Ni&L)ArwWEAyJ5sgeJ;Za&GfH z&+{J4<2KLy-7lVfeUINhI==V)+|P^ut9|UfuWMavt!u4wo#W?&6wW7N7bfQa-+MDe zMmQZQ8?1QjRP(V+epsdD<~Qg!Id`N`(E_Q30sXu;6p*^wN3dpo5LD|QV4ZRs%sa`= zI+wN}rL5{>7o`olJ9To~?-ao3*hPjM$4>ku4yOe_5OHg&&%=!xm->*Q{q0c9(f_bd zg;sazHT}hk(qi}Ng=$Dyeg1R*q!LCgSQfuZZadP=YeRGM{sSi|;=JFTGZRQtpbCem zltAAf`A_>f(vfD44sb!!%7<(=HO+IFU$2+gjVDQ&G8|FO=Ckux;~DZd5k#^=v#+S00(#^WAj zwyizhCNhhmPfYY<9_MPHth^M!HPOMc4{6eW+0Xpi37HQh)ozbi!_fXy!^eewW2|&c zhQn$Lq%4|KdXv_Jl}#FT!gqAB(fj&IElwiOhR&hka;1f^*JGvY`Cjwyg^Y`D8@!Fo zq34>(gZot~(0xj6gyW$M_^2!k`F&3VhVtK==Zd-@O_qM|$HsOnTl!_K$x8yiueEsP z>{8PZ=+BB5iu}3?$BGn>EAG|30_kbD zQW9sbOZ4Vy1svW;T5Pm?6Payc-hrI|6({o>t};xqhR2yMo{4DE7z_%| z-nUN~diVD9K2#Qfz7)0{A?Jy>-K3=Uk4@|ONSwyOciFl}p^dw;U`V75hPAZ{-~SPW zHraUYc28NPWegzc|+-<(f+$r!#>gLs#3os_G~UUkwcD56pgfCrsk3T@}E_ zH^1L9+kT$+NJcupj=YoJf7qwaHYpoC#yy>nV{tv#oT@t=~FV4$@#jP5Cx0&`N zHs5q~t66V_)E3p;qBMZ4{p=QNp3Uo3Nos0C(Ign-ZQW@@o?nm2JWrM0w*AlfVYbuD zdwqH+ot1Z}6_i4a37%dOr`I3h6CYW42wCR6ajV^FrAd3FFOg`DtqWvnxAyvd*CZnX zFQsj9hu(+LzfezE!P|Q12#GUOeN#JED)j2az6#cIL~4oXz9^xCFyy_FW3k&PQgcP` zA3IC%j?_`)OeA*B+m~9C@jk%-Jqa@!@45_;cH`@c3+0EQ-Q=9nrAjrV-rmuo6Q2of zZ^GA{m?=e?RqE5cW}j3cbq7hwpE3!^-O;27QrhvBEaH!nacpdPfwZ(}k>{HTCA4 z-xgJ*yUZP_75@N3b*a_8p#(1hjXk?S!%vOG>94-2QEOh`BAq)O7$ypRzY9$wn{=>4 zlda6_<@~&dF6ii8%T8i14JTeV%Q~ zj?kTU>f5+=z`T8Pvguw*8F8Leo4(rF9gi(}dom*OmJ{ne{`;%Z@+Ro(I_The%^6~n z#XT%mCXl#}|K#af{7>8!q>^~@M>I^3ru1#q>4JJ}Dm4>a!YdC=qGgojkp#|e?V_MN z3Hj@gCf>WlNqhv;mBo#>x`~u8Ty0Tdlk?+H-yq~VfByk7M!tE_~dl_NX zffF+?yM7`~>}&140X@vY){;PL* zQ!jB2_J@9M{*w&#+guMYUM?o&zjWYmTT+z|aqIyKg*tNPB%_+Z+u8XE z-pJk*^Xu*f55vrh41r?bz}#0xUwyaeVcUC~^0lL1q3;~JvJ>^< zFut)-M6f9XyZ7po%r_L1IQ0+Rn9(Ris*Hf?>JoM4buIn9Id(e^VOi9lqij>pu#(=E zZa2!zXTa7_5jn%ipJcYjT3WZJQjpdzxw_Z##Qbb>sO`u)m{iG$`ts}% zHtC9ppuRTv?M?8IBK`Aqkr7@Qn4Nm!<|0Uj{>2N0PRyu6{}-WP zNo89S$D@96HK&yS=RTyHOsmJ7xQ9)@bTi~d<&bK;I*)zp1*9d?D&Ovzw?jA4wRV0^bx){pSU77?=;|2SQwTkgnl56LC5g@WN(gU3)X6E`vUs1g`|k|?tt z1nzfkW~H*o1|Ot&h22myuz`L$@5BY}r%07_{oB;O5$nkxH~qa-2fcD>QO9FnB2`po zmR$HBafYtCk2c5Dux*(_e$c%$VeudJx_|Ar3$)Ikn>P>rzT)=_>JA5s?at*Par!7G zahNTJ-cu*foZ{w%)>5}I!5e}YLgroRB|mQ$vx%+MdyxTy!-Xo3LnJWy6>UoV+5Gsr zvy^XyGxZZPo3^@9W$i%{XEUkkAiElgW68K`bJbI{{B=MFR^NiYohrZgsw+WH@ynWp z{U83P9ANgpB=%(Pi~Y{nb~e}QlwL1V?wG{3Tx7v9hVH&i>t;r;&bRx%Pd;)ID13eg zR#`_AvTm6O|l#xrrzAWdVE^3`l$-ro8UzL0$s zKW$g)LsGbhJ5uCr>d2hkkH{Xhs3S*>XzCC>P zHNo@K!E0a78C?c_-4`d@&d8BC=eF@0?%s{mfW!N5ZkR<{O!vj^$ocoXASa6UQXMJg zN~@^x=14nybEHal6Z9o`2|wlBfo1`1TP1|%?H0@N_)Yr#7(HfN@ZgdvP~POc?H*{t z<}%~ji><$r7I>0s#C8(L%(gKqB7Koo?%-Ol`yaegCX%){n z$+2qj*oqY82Z{^qmmu}#h-m+@6HtdGro6|PIM2W!8Gh_8oZo-%3^&((x4?3HlcjoT zIoLb8C!kS^@DCO!2~vl?i4bz;gxuNb%_p$g%Xj#Dxec_sbY`<{nSY>EtFGO4rxhKC-kz3dbEn^omP(P$t2G2W1523Y)PC`UmsJZ z2wVqBXZexZFHfOwwzz-c#WrZ!wf})vo*(oD1m^2YJ3^n~;BL9*XgfgifkI#{9NSV3mr2GCHQloA@pcyA4 zP2f_(2wf1H-|Oxa`p|^5wXb?|9Je6#rR9~$*Hok#CE11fY(dI8csyRx0sYs0sqq=D z!{6(>19bKTvFsPW-6W%231<8lX`E}1hV8k7^aV3J*9u;m*T+W>e4pJmzi+m`y>gA< zZKT{$c3I#=;Ca#>aneKUnddIF?wR;%&>%Fg2Y1f-83-WFp^nn&O6Yr2{W8vh0`qoR zydL|ad;(91&QU=vQT0GtvmNJ-=n1I(($c69Gy;8LDiOIqt?=i*a`A!Njevu7c6~C- zq3_lYE0s_&U@UK3-aJ?WUF#}6RaVX0cd-;q?12b$1C>zz! z4Gte(jip2)oYpeTTO5tq5L9SadmvCSw2NtiLw@*8^pUzKP z1^pjN&ucDuLE`+?xPfdsft1GM8;TsmkeW3%c_fp_SLr`w9W1nXKhmz%^xuE?h|Ff; zTs-BSL1Mq_`s99WDK_u5z49&e2C-k*C)M=B4?^p~vu%3}UqKz?y^qApHAr0$_q{VG z183M(ZsrIRxDd3}Mv~4qzaw>#+9AdIdAX$^CA09ryd9S`F65+V4wF6k zr<#LaBjw8ZFO`kD&}(a)?e~%dy*1P$X1?h-8|!?z_PH;aZRfJni|yH9nk3SF^;<49 zm@qCbPT4kZAFoc{CgcxU28*`+sv+t&sN$5`hFW4hHQQ?c*>XpT)ta~7OW)Sk!T}NW2{p9bE7oT`SWd|NwEEe zdAX{&;i>js{J7uLf8TCzq*NSvaYbF2#F^bYZBzFcsgKt`63T3Y_lA0@!ObKrt&{h3 zKJ*{{2Pa-KIubl+G@50@>(8}VF`({ss(vAqEBffW9?zLSM+>g-L~X~W1ur^ZcyGa` zZ>`(S4ehb4`DzxQ>{4P}rhOt?pAm7__Tx|U&(WG8bCuUtj{GQSY;O*>zB2^*(kI|_ zrVX}e*xd?@slcYlpWmiOJfZj2pr!X;#8PiTw>yRN`{~>Who|39=+V!i+HNHHqk3EJT>Wrp7U}Ds$=Zbc2e0#=oDEY8 zA3%Y~dtuV-M(EAUoC}xMg^IJiq5&o^k*2bs%>EKf{UNj6u>%E9cm7mD%A%9HcjO4X z*N!MrOTDJ(=5%%}cn} zsCLRjf6a%pDwGHF`uud9YMTzyl5M;f@&AP}Ier2A)Ce504_u-53Xp!*t4(JqfeT5! z_xn-s;~*H*J(Mr0Z~*(sw@vTN?SRp6rHe0*5%{!UMWnA(l%+#A|E`eLJ!2TNa7}+P zozNF^K63IOIwMbF+mvPVeys&E9H*p481XjrV!m~8v}8iSe4|k^%kM0_v?!5jrw=9>_f^5 zija_I?LX&tT#T;!oR{F0%6%TEe4A?!`hR*m#cOUvYM^M%O63yhw_2UO-e4otC;DHK zkxRz%z6)|G^4dr_Yv7vV`w*!d9eXyPX~3qb42!e7!l3cI%FoP;&Y^4xmA#1VghXww;#t$*KJkTGKcJP*e z1JZcw@3~BF#RRUmZ@-qyV6&$?b!z1`Y;|BuNQ_|WlQ?=>X$RLeW*sPyxf~KIy3Rwt z{iu}I935#lmx_c0?}8!wlV0XuyfK>pT3Pd_?=Uza8_B7Dg3Kn}9%r=g9#(MZZk9Yb ziD(EQ6dw)c5Uy*XX0Hvy{u=%;FF^3bDg4k-ttMH_k_LiUu{1sa*~GnALxj0!pj{PX!0t9^tMlrNCj|Ed`A z*%IgdTLByGjX~j9wBo(p`PCKBw`x@NyU7daue3cremW879DdyULHk2yyWry(bD>~< zpG6y zJ9@7kVPLu7?~4Jl+c7oP{tP_QhAKO?sLh^l=IuVl-WT0jFzT``Wm?yc#Q8RSo$*$F z=-+T7p+SntF96lcc+?E89Kaavp@3W6sz^!p%5Wcjg_IL^MY8FC$!z&|5)~zrU~oa5 z@B4LBXdN(ZwBJVHc=TOS4XU~|4Q+2_Ub5N!g2s-72PTm{_u>ORgo;X1JL{VV~~#dDd<-))+AT&WcmNyS%0;|t7DhSxqf`j5ESbo2$^pH|Rz?7Z{3x&Pp4Ut_oBa|bU{Z>?9hxq21)KN_Td zaG_wcPry2w>2qj4vB_FL`VvwvTA3({sUhVYHC_aXb=|giC_!7#1cy&lYxuOB{$Ku4 zGIwYsmg*UIr~2~)Mdyb?pSm-Qb^a|36v>{q=V&VzND}q5rTiCM;%a|DAI)mjg-?#? zSLBmD{b>g@vF9999NvXAaaplVGkqk^u$0wObJyqX5HSJqm*Fs#?y*Bsv>(PV7{8Iu zDS(byYU#=I#H#Y`NdZv0?A(ziJahQ1`ecjQ6W2^v|yL zNI#GRZ3#8oEUxs-%a5v=futxHiqU^(_-zL1y=tfQ96TT+jQsj}`7}@tnXNGCNrY}w zYg6NcVo33rEMF0P2wEdK4k&A0gr4w)pRy_JNGZAR{W+G!o@9=>CMyn$tta^IM7FA~?0YiU1RDJ0IdTh!)qmtvHCCr3h%HI_~;AuD-| zL3h`gC#MTo;$Xyk)&9L%tUm&&r#Y5?*v#Y)frg#(GZMF*Nt_aA)>LcOK=w`PD+7DV zppWfc7&+oP(x2zNF3u}}UOn|a3_e3_*#6dB=0+`2IDA|@?Nv(lq8=7pJjg3ukuaMF_$q&7p zjUaLUrk>18dW!T<3nvytnj>YGn%NxnV8h;jxV!1;jTX%yf4!<^t zdswLs{X){2B4!y#nN#~vP<;@JHZE;mxqK7&t-BdzKdFM>J>N0R#nPbg6n~-fdm=xg zF+y-mx0;D_h3(fVmQtQqk*ZN0@^qG{OQ=~gH`5&CIWONLm+47u$F|g6x!VRzv54yC z^IMCEgHb>7FWcB63Pmozk2&QmLdRRSn|c zy11?+0!EKr*R5W(pUk$tGj<>&8k%|54R75k4r5K*t=?WVgi#}oW9%Atkj9tFuEwDa zZB(f}E?;&c3Yn{}A_YT#?a|YBZ7!paocosvcA_*-cA(Jy(UPoAy@{ZV2R0?@a?Uv=^ZQQu4PbUq&cpdUX=t)Dew>w)@ns;W^9Wmsvu;zUx2 zHk9ykEtqsA;;-#P9lNh++CsC#fk%xrJ*3tYlz*{}$0p^E5f;}4i2aiE{q*GSAZW3R zXn30b0s8M|M1e4x}}!xfs`{VWy>zKL*JT5GCe+HNRyK@r8N{o zPqJRqxknjDJM#JW#w-5V+(gx_ee@E>oZI5$d)|=P6qa7e~th8k!thpMa37Yc8)o=dl|awJ0*~+zA-- z+WkpkyA6x{s9LbzN8pDOrsyah^JkxgOF{bO#cIWr?N=eTWp|)ATK4O0hv}e7}(m6`-LCu}No)aSB0J#PS zhAyREf5H{QK?CMb^--YaFA0at>!rdzPIU zc@B)+1Doah3!$ICkM5n(k0qJ-?3CD692>BDbB}EY^q323o7_4FO(y2`Tf{mJ+iiC~Ccn&xG&-VXmm$!e6yiiv|_?;>j$rQJa;l|0VX$Dcdd{hE7<=OLGLWAWD{dv%c6UxvGUBLXg{mO(FHQJA)jG4%Sx-`4!*0DTiX z7o45Djnu)w<*oumzE|QNE^;;dQW$>`v!kJRF*F`7O}${Ejpg*st^VpI(C1i~?0iHU z`a66b@1!o7w<{;v9(r!YajL=4$Gs9ruM{1ucii;f=7YqU?b7KimJV&>`dz!(gPHR* zIHCXWbaWZg4+nI9N*ICG*BcjY{Mv-g33_{lzDU8y(skdp`DBpN93v&GZip=!cQ^G5 zpJTNT296at7M+ICVCPix&Zk&%^U=yfhNUd~mU@F$y=Tyy%oeh~VflMqq^;xXIL=AL zjZ<$ura!ur1r5>(Gkv=W-k!3}#j>js)|1$HE^k)5vY%LAnF1-x4=5A+Vg3I2RXjhC zYHJdCcI8RTU$OU|b~j-4-WS3jH%dUfnf$qm&fQ3jkUf8qdKEfueyHlanu63auFVmq zLBNRgc%`J2hmDkPexX!7R=dzx=bUV_2eHo3QRms+)7ZY4M>6ok66~^ScU00bMoNF2 z#Gm8qOC$~Mjf@PN?O3rrOCgknU6_2?`^O5SXPCFyZpJg00qbVKN1CfNik$C*A z=nXXz=jz4&vo$ASI4EiG!`W+4YxP?4)^O9j9(+5yFOP`(G~dYT^5PH!>VeiDd z-~Di-m2wUEza6@)^JNGrU28+#R?N?D@ub&jKTa!~ZNdI# zGq#?cQb;$O3&p!jkScIDzCn+Il*;4Zkp8)wXw4aBNg zaG(kO)>OaxyWHLXx=w1gL_53MGwX?A;QTVrJImx@uqep&Z_;;`>ltqM3Q8VQaaywP z!_v3kaPC)rRN$vmIJYNsh4jln=-cV%b-8-#-|I`jYQU328L3g7k0Ygu-R``_( zQf1coMs-a5>%MtAU+etB_gdKGp#SdkVQt7fW=?jT3kSOH&-45J=H+9E%G>i-j$yNR zQf*oov#taCOC%TTKm}6Tvdp`~$B^>p!k>`7NSq0IcPk)L78{~yotLdy_@beDF3NIb zaW$F!qu`C%VGj~#DwX@29^r>8@`-;kuy_IS9#hoj%+iLTyS!eIva<~Oj}^bsF7v_G z1##yX=WFNfkNv_fL4(Ze`)@6kw!>L6o7y#b2SqwkLm%W1$*W<-=7+(B?m^J|dX(F5 z{Q_WYRoHv&067SbX@u@;4$ahmZf; z5Ziyd?*CsuktQ@6=AUncokio~ylHzN$0|6|lp9et?LaXaMi1qpMMl#u?O3^ zv)eTL(l9Z?{DwP+q`B?NG5|?MRV4$E1na$y|uo1z>jf|VJzrN|2BdvYe_m*K6 z`O7MA6WV0YIkVvYBF)P!`d#u3WbCqd*Dn`|3G3?H^|~$pYkMUY=iKvGi~sZg3Jx{O zQ41D70|Z*E+Y;}n0oi-Wj4O+kaYAK4QFMB6UXQV~vJ7MLZeq)^fIa$BdvLC7*;=pt zEWGi|_NW{yU-6N);E3eULUChPY}L>=Tw#BX%w`?vZUlRAg!A&miVQt6+qOuyjJ_N) zTTSEi%! zw*!`wK8jbwVhFXA_j>9Xmi}~Y)Q9$<5@;^+daW0Ci^Nev9aZ*c*6~BWX|7&h&2g6d znxFHYE%y74lyia)Z|0|AuYl;sWzHOsEBCp1^-Fi8F%Em454nSbn)cTc^q6`*&aG2o z7;RXDa}HK_?RG{Zty-)%v*(|A_r&|!+S9nhGHvE7eU%No_qYWt9$9@#1# zORW3$&r9}MtFg!%GFzllb-{b0zLOd^PEsmniO*oX*G*}GjZ;XU4DpW(9>)aKEbY1& zhqP6a9aO7cXxJjwcu-s&DXtz;zep)iRNBwkcJ4Ng+Zh|Tx0kV=$3J1TPEr(_mhHaE zL$!g~(ZT`qucJuwT_xl0`HT;ou^6m1LhcFsFtL}D=!1JJL-MRHs?<9`gjEK|G?!bX(R{G>m&!Fe@ zsyM?EYZ9l^o}=GY2z{FBKfNB4S0c68|E;!C4pP2nuN*9R&HSG-w`FfqgacA8_vK%7 zR3ft-IlWx2@h(zh{2MN1aY2*xz>4c~m#}>GV`(E{79K_zTe0-rz$O+u9Xlv%xRd2$ ziE%xuYp&QPi)D^8UFXyYJ#303S$N=%Bg_4iMW1dpE+52>x@zk&j$GpRsz`%71^fQ% zJU6Rst8Cdv=#QylwO6kn}?_U4Uc3K10-VS@8g>#j{)J+f1 zBSrG2fVJoomGEMUQV_}~46rfq}PY^CbP+Xub5pYfz4iijAZ$=`F*Hb zAYK;Eyl${4UbwQ8@O%AQ`O|M%U=WO}c+`yszlFTOYwG&(gnroRld2~i-t>Pu&)6z8 z^y1pWtt`KfEgfEv`jb9o)_4kZh0 z&gB^7-`j@N9#ZVP+Cgj#+`H%XH49=Mzs%{pTKJje`Lpj@C3|a``G?rx%4KO{-~qjx z?oVq!{fM(3I)`pQUj%*aZYRCo6F4LF7HoH8H~d!|V5ENzU%&hy3r^cV-&?hR-RqXs zJpaAj``6CR5IDCaju0{VAcF>EgoRl+RI=zBSnFMspugh)Ft+vAxLa<(Vfl)W-JDl| z?wz_&`SThQNB=dpx5vZDY$}{P+<)t0bLY0aqz|?@G$Z+}?kY>(+5EaX;H+4?M}J-+v4hMe;+a;?&4E)ghPy8(U4rQdFHSdyk0j3g(e+dA6k@*lW@RZ` zr2lW_%zyhNv-AC-$ICJCUtwVBcw(5+KY2s`>w6mXnA7^kxCQ<1Z}iFE6(ez4bDr(b%D}9CP@6SS)%f|4+(=7QWSmJtR&Df`-|EKeg z3DXYYZ|`eBFK-DSV`&!jCT4N$HBN*6(_c)>g$dr*5}B(F<(CDKcFZsG$xRmh5a-^; zLDY_;*lb;?^30fol=RD+9g{PWcHASr{4bH`JM`M?8p%-(+HGB>_8OIwxTI$HNB`!6 z{xwglQvgRmxWCvGpr3EIj+7-Yxaq5(Pm{Cb%B=U4jIy~-axZ_u#DVy$4LbGIJxF!F-Fl-W%?4$_gHkmA!F%vxVgIjgud{tPSM zHtm!3MTvKpk^W(K+{tIl(Ldg_ObE0|oQ%uP{0o&yTr0#}m2XBOWwdNrT=OTaPmwij zoh2hR|BhK%X*~3e7t0l{`GQoZN{v2Go_V{gjM~-u0-GW}xH9}^fbpkoT5n!|@6Yua zl)G0=;-IPT-Ke*k#Mv*fB*D)OoA-{H9uU_AhSb+bH!K^0k^1LeWbj@RSJIB&yUQK0 zsaM}y?C@TsCMP_2BkIE+d`Q27YpG_b&=6L7ZwC@_`>E}f zRh>|X9Ve6bEvqcV;m00ZptKcaeFSef^Kx{v-XCl^Z__--QP#l#v~ zVpR1roO@a4wz-A19+6=i$~F++jdQaaj@vx5kn!&Q8M4uGXb;$|nW@fNkKQ@aVcz2T zkj(bN?Vjh!|EM<$T(X?Mp%iJ-L9s#yH((2!h(Of3GsJsuue#bBT85O|6J|5{1TPXT z{j>b}{~IcsC27?$)Ms>mkJ+mEK792U94jK*=T zi9bYKyw^r%SL)OV2#-WbF9vUvSmMtCj``0b!MLRGTZhQP^z_>Hp9?|}EmUaqCS0vpF zo|ltJm1kxGbV;0`-R#{qgiQ}4Uc4A#C$sSiSeJz{0#af6gD(+-u5MMFXjm`z3=nt22M^S1Ei@Y{JC`ekBHTaNSwR(HJ#eBhB=PG+Qsjc zHxfKX17iF6=xU@ zj^&PBLH*}@n(dr?%lA#@|8%~P9`?wTlf55?6aQv?zij?ruP`GS3lkk z^e6U9yy{(BLurDiX6|QI)S{GOmU(JaW^=fyrOVRZ+?jh?*LQ!zrUl}M4OIXKZyP&G z*%N-Ac7=>uzL_I9P_ghv>F+ii_)2~-7R!o&OikvLzO=pBk*0UgUn)<*DuCUI@q8@}Xd z8}#M+m@DtHf&POBtvMtGNL+#{7JAzh=k+nF{IDFhmRWzg;>jG+!Tel9%FQ**&x`yRBINMsvc6omd&USw=(;Hj@ zjI^kiwdf3-9ss z*qMdYgxV8!wyj9p75^65fvV{*||n&n_j-wO>Xe`>dFJOwdm6%}cxa z9Ooti6q;lSJ+N4v$N3Il6Si&@KL1{Q5SyZuV?5<6@XwhqZG$7?^LD5rpSS;i;JO@p zX092cjx>RYS^-U_y$SJs&;MB0ok8mLoQLWQzmT}D@|#65GI8K>O}Wa>eb_Fmn)a&v z$9#Wl3LlC-MCy-}p;Xh)Q0MD6+|EnL{plPId#bk?7DfCJ*1KGfl-d2+nRG>*YSk%p zx;~C{>-X1+%SIuqEaq6-7?CeT&Aa&N+Uego9>}#%&X>hs`~TG^mU}c8CH=8RYS|RO z&885f7FNIFk{0`a+a8JYZr&}cNHUhk9V{Q-QV+d1WX=h6k04EoMiZ?XKpOkTYuqM} zF_rgK+XvN+|FwU^uir}=1Q}qH@NVG}|FhT}TxJ~G?12>d(S)&RbEN!rcYIb(;8ioq zjzs5dtU<=EH-`tD7;tZw3)}sOG}iqdB>8%+!ufwa7g~NV{QKB|c^~whcZxUP&V$2m zpBwYhS@s{9P1sw-jr8#U{G6K0uN3qIA1ur;*Bju_WTGIMUv!H#M!@1dlHij=k3Z=lrv@OI?#b zJ2k?>56E&|%Ub%Vi*e^a|DUVQkW0B*hm5Pc4f*<6>|)5y7!*v-(p z=Wz;;@i=zw+A_YTneZFWPKGM1+}U0M`8P2Ho@V7hW|H^lyu3H>xXkg3;C-LclDW=s zRY2#shkFHte4$_LVW}`T(?1Hehw%4?pMKDEJJ!##g~@*iQ@=%gy_}ZdC->1oLuBGE zV;PtG^wvepxFZalBWE&BkTG*?ncSSA5!Ae|)YUt?47=8OOQdRi!!~77*bsRe_O!Tv zZ_^gQSh7F+7jb4C7}icy6h7u(h_jK|QWB=N(Bg4=kDU|~_aCa(oVH!BVT$2Nw^|ic z-N~F%Pvxt)>@anir-8u{XB_x=Sv!Y1jPWPVy;yTL7IS|`T@^DV@DD?T%cDdcS@?)? z#&-MNGc6xr?BoZXyd_*@_A%MWRfiYjr2O8a3Im!@*$}C=S5%42rf#ToO~4I?k~|ho z#W7{`M`Mp?1*6ANRa3j72WO z$R9rOSDd?`v>+wu?uI4MQ6T-ad5aZ9Jh&W&0eUc6x9NlQrwn+k;L@jmC=thEj{6jU zI1Ttr#Xf#J;it%m6KLESvxm&9Ihp-{%nf6P{(A-uBal`rSGYEm3uvpKp_8W;nPb1V zu5OVx1V@T#eHPKg?#r}L*tr|JTw``G$|ZPv8-Fb-Ck;45e{gxchRAK`^2<_mdo_l= zJAx@Pyd-Fr{<2T7hzEz_ZfwoWP=Ho?B z96b^8+yF;-Uw_*2oYhZFD>M@IHF$=lS7*PNtAs;^@5dL6r>`K5-_gc<7r~!UK6IFk z=b19n`P2^ZnzGb!!o-1nzv8#(V`U`RIF#?fMw;2F5R-e@mEv_L;XIR{0%)J_s0w>C z^9Zq5hBw$aDFqvPl3l+HKZB;cJz1Pk2Gro%u>P9_?%L$Hxi#MxzQInVMSXrfFUTBM zQeJj-6SxEc%lcY9r95!5Me@sq70YpCf8?^mz~a}&ex8bg;)_oRIdSQp(vdIzM1LM6 z*iB#Aj+38)O&4{(hvxI!*WPYOA+ssIYd!J27#W>88!vb=^WsPx(kDvq98@5&oz?sq z+j0`pFHP_Cx9BG3i@d54Ur>nM3Yph8ey+v% z^mBW)3;7{0B=k}1s?{)9`SD2nVp}NNZGDLIxDA=}u*Xn#%2O!#_Lldx&Rh6#{^(dj za~tHyJ1n_!b2}F8d?Kpj{R-+ASs&ap&dhJZs&{`nzXiMFxTpr#ma!L58hqmTZC?|h zd~UKx!k|A^GfA<33v`(SLZutwWwzzoVa) zC;ugIJJPoe_^(-Qg(Li#LuFO>p*_aJc=heY^Y)GIuL=VqpD$o{;L3GNZb0p<>_a&X zd2BjOci8*kD|Gs<6F1&MBeVJB{7RZ8>RoA<_&(ZR-vgsD+Q9-Rx^ZB?Vz%yTrvCw| zoC+@GxaUFFNQd^wZ)Us|zZ`kxvvXn#^gS^g)4X{Q!`{1;Zwz3m>%^gl3tiUbjl+}y zTWa#cL99^xl6Hyz0ha8yymb8jbsSo7Q>!|^2kPW2M=7s+ux2VK@ty_26HNP}=C{=@ z6Hse4QE#t{xV78gszkS4E23Ro5eA#1bSUmYi5r?tMWOmGp=-_8|T1BE=m=q~BE9%t%}K=*%Vv3qwK znOnMLf9Y1D4!c~`Z{L-yWawCURAMmM6*9;6TsZ0b5(iQZ@5tQJg3^q%x|iAKaN^az zA1$1Of8j6XQ`c6hRsXa9#JEjDo~@^bK=1D_vj>!J6Z_3=ovEYO56Jv-lGo(u$bZ#S z{?2$&-zFqucdz_21#bn+8or{`9nC7g>JloKO}$BG`Mq{k!X^0!5;(l$vq#Lf&(MDR zl!{Q44pbldLBCKZk7;|ZmT%Z}`oH&MVO>_N4Z#QGQb;jK`zZwjhKpmi?j+{9=}MPt zgU&zeIdGhO%)#DYlFZGset^e|;0dCh_jV4lbH&6554zm9{{qIBwH$WBmC)Zh;nN;B z$kNW}OS`MJd!FO360WT90>VyfZxxLVu8@Zx`7cTj*Zzmy%%@YJP)^{sP{S-Ad?4_f zvAoW*Gd6u0wrd8FZub!NQmFz9f`%`dL9x-J!)GHtLa;|{M1}7^^+pI<6DfJH^C^C$ zNuA$Qol0nB*IWW7wg zK<0JdwZxMZyyLgPKQ1qS&_0lx@BIO)_Uo!n@nC=dk^err&mvpOAK)XKkn8 z2Ap2rls4`>10CHyTMwy_V0bv(MYbUdXD91-)EX5cjT(^0W}Cw@4|TgN6c=(c@k((* z|8nn*x)y9(9U}Vv`(@1XI8ra)9*<=&i&W0Yl34C1FMQ{4HtY&?Ytx0YM$G2*iM~p) zZbu-~ik~}q5_)aAqivq-_gt3yn}m&B8b|h7 zz=;01+iqI|zi&?QWFdto6aHvACmSABhHCz_blJNEPW;e?!v+!BOkPdsUdq$`W4$Zn zq@UmZjDx_5p+%NH9ZLV%@i{xpZ2YC9Vzh$68agBdmFpRGUSwlLhmzUsCn9 z+Yt-mdRuy`jYyymM_x6t`gSAepvNN{(+VX_pDjQ_gnmBp*WSm z-yaHo?HF%Gth+H|^%&pJjwS!o73tE*d=sd#@YtfOwVceSpxE!Wl0o9$0O4Kvv(WT=*X*e;O=Na{ugi|Mbx?7c!(%RW z8%z7;{7Y8_i50`od+V=`Ml$o(F!dLuNlH5&B0c^Z*lBNtCabwK{Nm@ZpU0TDQbw&z70mbd8HRu()b0_ruH$%LrV5T7$w$^5jzpy|ilVIX7Y7l(inomy2nFu58*8?QPaD^K(&&+0<9; zeDXo^rZWR4#`HV-b~i(Mo4L%E(gtFk^3rLazBAVqnS0{V>e@*Hm#yC}TcnroIg0E;aY|)62*_g%{WL-~t%4 zJRc+BG7U|gPd@*(ae#K~b>fNP%UIU)$XvdV$=U~G-kWj)Yr7W1OAXapsckCcMf#*T zC($shn~LC@SkCl+kr$|}osAH>3%z&jj0=U%W9?-A`mk9N7C@-pi#-J2*0`4Am3trZ zfnoXCyXg`gbCephqSuWewO?nq;FgV0Ga1jWeDpWWYDcVne2K94+L|8KCFHYVtH9R4 zZEw4vP5y7#Ps0{y2_el5dVPS-%4Uznx00ak+*K|6rpGWcT(e0l|2|AO7aM6Wn1Tt@ zch}q(O5-O*y5FHZrXB?o4*U1krt@LTk^%l7U3pkvdFgYs1F#=X$xsd z>Nsq^+VU#L4vd+zoATgw!%2E^^R?Ko=;P?7Gh-G3BOywfpGT9hth|m}zg^6}Y4O#tTYz_uL#^CWf;|cDSbohr;aYH1^WH zh$W$&d)1VQ_b?*dvqbftFg859;5obIBGzP=pAeVSMaC)0ch{4jF()B^Vf95#Xx_(V z+*Oc>S(lWOm!Bc_N2U9&mvZcx#Cp-l(T%$kLShfz|6$*bf9A217swniF62~WS^t#7 zo1&cziMY(nN6r^-XNNJ~`bh&dnBduJ*ugNB{YCJ$C&-gx*;Ru{>>4MkZS`5|G05!N zTO%!Qu+$k5^1)QkuDpZLuS#jBtd}mX`sein>K9jOgtF+BWKNGG(_F8A!l1gTz5NuS zM|O->9{%3YjQvrxZ2RW~FH(7Z)R!OkyfFFF_WI$Snm~DT(dEoJ38pBX~rX#scrGjDkh4B4-}I7RaZb~dGHSno1{0Y9az zgU*CJt5qeBOh#5fb>#T|OGksD;**uF%CWo5|J%VNg3l`u86?VOyHy*Z+DXwue0+q= zo>cVY&p2Tx*Vbsz9`9haS9_jLYX4nAoHLXo#;^W3ZYFaj8(!Nu>V}1pA2)3HKqqsT zh*+o_KEj_(2iR{#reOI&?e)@WeAr!aty1zqC&cjGT0hgs)I-S}o70o8w57weiC*P& zJ1egnZQqJio%5`EQ``APTh@6n@mC=A+_rc5ibY7jALU;UEIgXf=`Zed<}1rNL_FuF z=}N9_7Cn`D|L|n>TRxi^7_{aAHn%h^)1<4^Cev_sfw*zfyMx4dw2TjWs9KOYIJONP z67WQdlI^lhmXbJ}ICyWMLj+HCDz=>&jfR#Sza6HRiTwMT}Z=$fn0a?XT|iB+eNaUB6VT^87IpyNF)IE_Na>I|&XAt<6k?4w=|0sm1It zeQS@8{Ot;s`I^<^2&Fw_iSyv#5z8|xhR$NOWp!fz!zldn{60@qK@B!^519UrXu{Uo zBVv9&SvcUDez9op7V-iC^>Xv|0YKYi)W0o8pUf4nQRAS?jQeAJ$Iix&%|v})wYgi8 zIQi z9B?`2Xuy20UcSfA-|&2iS*nM=Th>irO`ws*O&h|`U8}co>nUFstP{QTLNI;@{t9Ls zE3Wjwf~#FFHRp(YU;2r=yE!+Uzen$$r9oKQ-no`1)SR9V6o-zLV%nYbeerHvdE+(_xiB-D!Jp%@# zR=!;j(+Ansd{Qfa6ZxK9UE7Y_;y#75f5LB{5n|#(;_s-4uB5GmT{d9N)&4!89LEg% z2gJ`CV9(!0YoURLrSYL_rc;h$rHbq4v=3%bdLv=S)m8#0X}aGx>G59^OnkM>+JH>d zI}REv3rigBL)v{kW%1pa#C#8DKX|P?j2-cd{mx~3V4t+{(tr~L??blWmrueiW@Mg9 z%}wMcCN2?JSY7!%8#dJL&bxq1W58*uExrt+t637p>!4^dCf5F>0` zTV=aL1owluCg?cD3ZH70mn=G6(;PfNk#v=H=_op{~0R$qQer zhd)2Yn(y;>dZ5*ERuCHE9c2}EIzjU3dov4DiS^hyO8Z0Rm4HrVm(Me6GcoO~so4Ga z4l>(HeG#@yOK7ABg|C$v{h#eJ``7+XyNW$j8i%dQ?1OB-LpevmnS#eoFu1k%kx@!5 zRJOZvebiy*;lk`1gSERg2_Bq|FC?kys7T0lcYe1nQUyzz+yhD4EO7vsbQLv{xe$&t ziy85|-Kki5D*X3zCteu#^zM|p&xW1DX`_7=!B`Y`EczCy7z+-~y48L6#O5VYA62>f zq3*I(UBeQ>ULH|Xlj#cH4x{&@jCy~6!S29K+55K|LzAr0)R_mD$Q)T_yF#|OU{v)@ z(%&BkVLazv;m7MlUJCiL``V3te$4hJ%gwNUgjw$aO-sYKU)Ko2@tD86zczY6Q_gXR z%_0QecvrFcVfMy9FpQ=frW!6F<(<4>Cm(_TKWF*t_Emc#zBH8d=GKKTCvenU>%7#W zXE^j=d3ra3gGl2 z%kfwX0&jlk*z9UDmk1WyT{u`D!@#kw)3;6jRAcvCsMAavv(6v8j|6Nuw|FIJzaRy- zt(Sl%)3KV7mO)58WBNNLY{ftIverjiNLTvJq1ylBot3rjEbAvDyn0M@dl|7FgE`{J z@o_BeP&3*($vOldcFUjiE4OtHlG$I{-?=DP4MnH546J}@_W=Eu_=aT-BGCCtj$Ln= z61E2S3x2>7mgjdSIW`nN7h!oG{rV2B=Yvc>T^P19S6yhrv`29C{p6c$bz)!lj3t}@ z5NN;}{k!!2UM#$1P^la9WRVQPYY@k|vToD%r%=1^_TC0XDV){H^}i(Z5QaLpd-$lY zX6fgzl4Fw<$J1Ea?GC%1SSM>t=C~RpapQ^5h%oK`t?T^)|~t=*cOlpESP30fb+@WkY_( zb@mu&5AOHB{ip}0B`cgTT_4-8%93|HpT*9qq^gH=CD3R2m^zls2R-j=9ezAw&MOR9 z_72`SnFKB7BafmwS^eYr;-#L8KjuNtbE~V}#cW6~mI_#M? zZY(*8b7rv`B0C46)V9-EWzkP8{Gwuhw1&X1Zf@^-@WGD@llUDCY28&&Zt>Fe;j11j z_r0MTZ+RT5g#!*>XzBr~Q-N~T_2(q6$6tbSBo2Vt-OUmC2a};zZeQJ^q-hwq#P7)& zPT<&1U%dKr2$}XU&UWqE)#f(_eX8Hte%P);idx%KlP5jExVB04-HtOrZC%U=-%03G zr62o_iq!YPgu2V#;-!Jm|F!kq7QrK!s@;3(M8p{=zGfl5Q=jnjPLGZ|$)pGW&v_uR zukweX8Hx8ul)FSTajU8Y^o49lxK-MGUvnZWyXnT{)9a|#bZ_!@8E@W zJy9DWUNJJCzUs(^2xi?G&QzcFzwW_>W%!u$+us$KWX*j_y!Y1s{2#0)=iE7?d9moq zgR*lb%<~a4>#v52uVmJdL)o*G`u74X@jKS@&a6Uf#Z#u=gv6$1WjeTHk5Ik+;#wU;6@}Cxe^*>8v)k z4e8`;a{CPog-r+hNK(jnZ)15qiW%RU*FVa?rJp;Cg)h0OaTmsMB&Y`T3LIeWTA(WZ z&q1tOB@nrzaw`<=YPC9fIT9*K@hS2_gdSTm!N2lu&>f&XcoK3jnjeSn3MMUC*aPL| zKiT)*IEyr!qnWEzn03o!j`y3#dcU#ckHXJa)%2Mq^L9qX=18H}oH(F(C3vYZ)Bb`% zkIu20$0V2%9hurPeiUa{_%;8=wXEyXagibCNf%_8<+w_U7sK=#%lnf%G>Lt`@yXYF zBL>Jgawkjuau}f>MCcACd?W1gQStqBd6!wF+!*q>*}e{G-pzTOVNAOl!mGY*i`)KB z{CR8*d+Va~2e3f@mty!)HcSd}AKz!Q1`>;(D@w&4f$m4af^8T60}n`dS>#cVG0Zlb z8ru-f+%NcPB;-E%$x`O`-k5u@#a2>aW?}8Q$1j=sD49=MKE~#A2unYv+CSw*SO5;c z(OpUZU_?Cce0ol@kQ9t;dK>}5$F;sg%a!&D|6E?EIIr74st6;oPh?T= zvhBi>N53`fjP0RVDoVSQszt1GF`Lm=&M9bda9L;`#!g~;E}?BmV}rz>@izJPCrF&~ z%dTA-Sb{@^L3D}9?=Y=-uD~bB8Y;B&KPc5+h1veYZVy%wcz)fx0tb&QVCqZI@R3Vs z5rNNz!(9&XQF25awQ>FT!dKUrJR86;Uo=@XUPk87KU2SMf|L3CWs7y1%8%TjRrIW5 z@n+U@v%Rk1@L19eRul`!6^$7oeWWTlk(qJ?|uqe!}Pm?$Lp0_ zh2*?&>{4n&^Rq@`{{PgLr}Yv3#DUfj`xWIop{jGqeT$btWIo>Gb3OaXFse7Z{?XS= ztV!=)Y_Xk)TQ)b5Dpd7ifO@C)Jyon`v zLdMfh{|O=YVjQV^?{arv7+zKK>b%WHn2rAS_vG+99K%&J@*5Tt^X_F}zq4md1)Ih*BHvazv#jgxj}jMjjx558T&^D1`yVmQtK^uEtyxScMYz-a0A7*#wU)MvOm}PCHTjGDLW9_#?0%YhN zzmfMhim0C*(u)^5)6-7kkSu;{+em@JO|84i_f7)k?x2;`B|?uKEQs{Jr$XrEZAVjX zZqZ-(&;BMZz8*=HBY1MhmtChNNpDm!qb}84pTux|yVM3ex&B9O4e1AOK1>$T{3H;fPfVoEbrLQn!<=buM6;Y7N zq4;>nk7+lMIIM3tY+iW{It9{CSDsJ?#wC$+wQc+uYeUdg*guI0)jUEl&(`p2A_VKi-PR&9JA&NK|6b3kC)U{TF%N zA+!0VJbr7_Pv%)Ox^VxURxutY0{JZT zEn2y6ULW{m(Vvk7BbC}~9cPKW_L_xbN*sYDK+hej-SLRfHwJu_cU>50AhX>Q%UN&A zhE>i-v`#&;#Q|ZfgM$VHe^N&Ezygb2=6GTFX?v!8qZ^i9%vNsZde8FyXZpWA-ao{G zCrF%A>vUe7x_Jvjey*1u#~v(Affc8tnR&_3DP|+jb##FF{IN@ci^g>co~jP3-Xk%O z;$Z4qx_;UxQ!?lDvhyE5GyMiwq2iP9i&2NORcDPBc$h-_I>Ab>ozI}_Qt8^s;%aCU z_|v*_Atx}bELT~v1p?jCcDi2vAq z;TOu$cl+k$+$bYFin$MwYWL|vh@2HvM5I11`zsEqkasCx*q+2y>0D+&VLf;9WL569 z|5nAWuSb{mTp{qNhlAhRue#36kHq?$l|Q;_?gHb-X%VRrqHcEdKxOKKH;ZB9edx8H zf0+6J{(`K*KIoPoWOU3I-jkA8x@(dA{!& z0Yu(PfdcP=ns>U$II+z1d2|pCyB>M>TT>gyV`?kL7tfz}?9O8G*>NyZyyzoX-o;;9 zwQ5EAgq_zE^TPbd8ZK)yHe|JSDg3`)qN+w7HF{M1CFm361ltK7JO;-5}|Cc z!&&=~({71SFBt2%NYd74J8u9IS(iel9K_J!XC_ zM7(-FbX)5dlsmrD`?jbV23`(+yf9&f4GwGM50|tp}zN({MyQ+7$}#g7$HXR1dKHIl;wYGMM{(ZapbN? zNH*zEF zvX7V7&%{LOklB&_Ltx-i=!+3ARj+085vJO>Ox_lp9Tj`9M@$_$V?|zyZY1oZ z(Gk<<%Kc(M8?TqX<$f8f6w72aFVrDzp#E9fr$j8gqqx>cqZ3-@a?84|zroNmr%&m; zvOqF>ulul{Aem3@VSwWHHmK7qp|C?RuG+yG$IohPSlKHMeLjk6%a`T?Rm_reRr3ev zh>PY~sJH_f9SoPm#Zxg^U%To|-3q9h=3H{&6H{M<+MHU>#^eDg99Lf)x33K{ABXZb za_B*4jYih8pM)QvsrlXGg~Cs$rXWjr?;Spr75wc=d1uGKVX~WJ?+` zl-GRkm%pRXHg!lp_)Q7W1m8MtS+W6o50>+mTCw7d_8dgh%WsZBxpdmxGQ0s3;Ns?d zYZ>%EI}^Bn3Dcg(Ub!)~^c(weu83a9b^J23AJywZ%#*gnLw(U$onJLSyg2@{_4A_l zFlC=@{pu~@2WzpI-DA8{4Tm|d>Zso5CG(~5^MRD>;K?3rr0qM^$t^gSzVl*;b3Su}W3STUqDo;EZ6d>wcs2WdlO22eKRd`&K@{JM*UqO>r6C&cZIk=`XYjg}M z1D!?`Pn!Tm>$%GA{8QL^J!8Z5t+v?QJY>`9yb8SS#Dl}p4T?IulMA5z(fHM<8de<8;m_sWg7q;tkRUHb86$X1>*cvF`JOw7116E& z%3J@b=fg~{Qdy`XGcEzCKe#-DrlYVfo9FL7HX{C#`e5U!cuqo(%YBxlXJ_Pu!Q#9Z zsmFP7xTux0BcTB5xVYPl)vU0ljx)1+4L@|&^(;x^8T)_gPOw(!f73Y_g2|<-);hbkR!TT_^saR#wLW9H=wz}4hG4#)QU~CP%*_RTAGv=2n zW0rZ~mtB8XIK-;ph=NNt|LcR;{M<0F_i_icsqWmcb;m6z6L@1G_LqncG|q*n*3(^~ zsOH^5kBz-BqY?i!DEtzXzdF{Zm>~qE(=nzD9l)wp*&gX`&k(QcxN9Cl7`k{r5Q@FY z?7xCcZ!w~wev??TLLXB*~e!8R53$4N9%Qmm!g5;0`i~MTZp_E7H z?~gysjEPL!n*8cMshj~kob*Zptr6qJe{oThO%ft(V{;;`C z8N1>j(rDx74HRaa7AO%NZ}_TqAcaGfnt#p^89Zk`9i4iGHKEGiJ6{nz>79SdN0waY zWm(4!d2c;B9~1GCnsm+5bH6?k_9*$~p6@$$NJH;1
    8T3ORUcE3^Vl$GRKCR{PaDKvHDN?hv9TZ9DkX;v3ft# zzQyWIJ@+-`{9u&xV9D%VOGs5c*Kzj4T1<j&scOgLuw=sw1t=eau9oq)Xy5{7?m69$S)M4;Q66aQuj4AbSmwF2!Y|o=CKg*6hlX$RJK*5nroN55CRkp( zG-_=DbKl@#pGM~WOhc^vyk5wRkEjC}nqhNo{<|Jq_gd(49*Y6SOOLg);~Qb(&BcAe zWDcBlY_PB-6M3`46EDh)?`|No?>|y263?vX$EjlFFS)MFIvT9f(L`5E!v9r3D-k+W z+KE*%LSE8C(da8;`NX$(3Clj4*l7`%xIvm&*IQg#+%IY3pyurJONupQwvB_TaVJD^ zOrrYtW0wT1dVVT2|27f-D%X3}95_4*rKiGZLDN+j;TRyCkQj{34Y&GJ!-zWBvN!8L z-}&Bw3<1u*o~Oiowykdw4GTF6)yoAQj&WZ@#u>M;YchUFd3(idxOgw~x-7-1p^Cd+ zWX`;VFP;ZGVOTOvz)~X^f2~~GFGCfD(sT)q63gR|`Lih`6G_;5B=7GR=SVEAxV5Lw z#}CSPjeK{#5(GsHGpX4tjzI4E(5Ot({Qlmx=~!$-D|W@MqV&2G{7qBMLoZ}}Y9QR} z{dN7CTTuIRPiU?2MyPkc=#=iV0t$nE{;hiSA9`toFKK6U2Jt?gyx(-CuLqiyL;UY+ zvfx%=$1!WepCYFq&APU(OD5}o{?BZ;MN0g`1}kRX9~2a1UY30G7+VAHENzk{=C`xN zcgvY8JF&8>vF=WTEmRk;9$O>Ehe={JcKbTdV$puB-*Wkj9P82LkZ>=k02ws~$VU)^jDzLa+S zIrSZyo0aMci)|r3$8s<^MGa{}M~6}Z=k2Yrv=y($nfe2nYw?-&d>j^#s-HE!#`-c8 zR*2~L9#Dclz86PtD1O08Gs7!AZu9ToyXUgo-$008qo$UYZUxjo7j{DC{Ci)uN?L_y z0c31!s8UeObm=EhV9NY4gVt}(wwwkM*(qJe-{zAK@??0atezIlk z@M|KT6DsoS@yh04Y$%OS=JDv%7o&f2jv7O~9cc8*6`h7@HF%Hq*o9!+j;zI76!dyZ}wIP0)cFTSL<2dl* z={Bjne2mwd+`O~Fo@KnFGr}L}JDss5?CL>w+PuAfYy$3%%;B5F>;;ZCK2XoUyh#4| zdJH!=Hea5?#8ZG0W5#C3*mR_Fct5-tejWzheL@zUmceGN-l2;_omg(N>UXn~CsMB; zaX7NC7n)@M-qunf{0NMx<#%nQa$(RT7(Va2h(qExOjmdlasSXoJqe~pR#4iLu`l{E z;h(75ydtcMHxhdj$8%-pHp5iIk}7NWtvI1n%$u$j1=HJiEnefBLtgkTZl|j#Ck!8$ zn2_AJ5Jsw0ZA!JC%gYaf;tzY`xN-$A!om`V5KZYZwr-63FN2@Fzuv1Wq} zbi~T1j)upC#c`7wc$}SVBemugH};q)k{MKW2Hg zMi@tqZke*aK8dX!f18sCo^~917;aOtMi?3g=@r$#T!9j)#O9|-Y;P~DZ){G5g%*?2YsN-hI zXMth4y11t^Z*WYO;l70HGGU6G636y^ zk9!>U;>^b!-zQ1Op?sfZYC%f};tDeVc8 zW2;bgvrKO+G{`)$)otK~?&g)P>tl&HSZjHJ0o`;1W}czH*ldxD31!Pk?d#XUmr0@L zhJOg0wP6Dj^}ko`aQe85`%Ho)Ok3HvY<4H~=8km-_5e{puD=N1EOhOg5uk8B=EiMtru6x0DUi z*`JtSc2BRW9=kTr=3P#1V%$w_<~ed zJck3V#a`!gx}f>V@m*K7#ULTp=2DCbGoKrBq)m}sp$1CSCFB;1vEVl2;6;Z=Tu)ig zqbRk79*6!mL)q#)-|MafF2v`jvRP&x`p}~E%}Qg*B$U0AwTKKP{C@)*?&Mx|Z^6E8 zZ3f%DzeD$)RY~`R|}%^8+_7uBtF8>6>q<1y~m+fl72GjEYpr9^O_&My1kEBl22X(K_ zi=>!wC#YVOR`jR(pM335StlBiQE`)Ho-4Qa#YEgk=)Ex$y|nrdmVPT}pu}$>ac$Mf ze6y3tYabFl8(25r;ZZSL>Gy{}K~ta12>)pUKQUS)-dT>MfvsFy@5-08L*L>;xi$OB za3&u#s{&)OyisGN(Sfx%z5m|3=8;|)e5m52cuE4go?MH}S7fmxn9rSfwc-1(0vQv13v5nY|=kpX~=g_Ed@~zWuo_OuD+yyHY z96x$)dIWv3wC6g8iZC$R^1x>Ce(01}OQ;(DjGYld0@it!m~r=Bh{}osFmXOrxk8)Z zM{cWZK9u|@8KW}qP?v5vh(q!_dT)!K#mQ?RtiaHQs5U7{CuwOc-aP#&Rh_^yDK5MJ zrebObYg<*e(V`=-C_7U`wq~kBJ*JG_#T;?rqX1yGCtk(Q}MIwz@CtjZ) zF3)$vX@zH*r&XD_6j-GCG9ia<2$}u0H||cb%E^(xTpjCUxc<3*>qJs>FWIr+bCEfd zy(Pj=Z--%99_hUG%zSh*_n~UvLxnuhTI`cLc`gI0uN4Q)q`E-w_v^e?f=-Ymq$=*T zVKEH7J1jM1o=oPP)*4*!Lk=rzkCv?edJFo4c!CcI3uAu-{mb8Jq7L@4dCcB?Wg51= zr3&>1@L^2B*Zj8z3e0|zt0gxd=xD(DcaaBDgp^=L)I(=<SXz@i$Jq|<7@G89o7%o_f{G#hc=Og z06$m8uMD~N z8%F206u;k_3)6}n#xDj8FjXS*wHDnRn<}*o#Z`{tgv00&tL6Wdw`|_-Io`AB$R%i8 zaau4!umW?qKb_&1VdAtylgs^FB}<|Xtf7`8b@7NPl+O&fhSzl?jVU6Knc7Qn2~7i!X7D`=Bl& zp9{>+N-i;E`M+EGm{y?z3$7dV3xDVNe35luw0B=~>)biOa{WNF(ebD^GT3?Y@Z+>R zLcd_#-*`j8G8^hvd0y$ZDIu{R$&BU6CUEFyXCu=62VTSQV$JlS{Y?Kc(3dM|ZhkNf zrPgP*pJYqKq3ba^0pTx*^EWVEZ_(f|PN*;0CTwpCr90BbhhS}bwPq(|oLd9W)t)2ZrFd}@i*t0Aa$GR!f?k6-M`<}vOpHdp6 ze_d5ZHbUqK&gQ<7Ma0u)yy~59^zVf3+%g+M!7*sL6F2=4FJUupiH}w12dIt;Gjg-3 z#xmt)tL)x=1#(l3k;vRXacrUKd+c^68^Vvlo}HVSvB(B1mRhIma+!wyuBB{JOSGUx zzbD;oE8#z^C`{e9^2Sl@ihl${RRm5=BA4?}?3uODKU<rue7+W_TlW`VIsE~%M7Ga0I7Z@_cBua+-d3bN-CDdPaw~)zxln=(_merq zmTk~kC50{L?TjiuFnPGJ_fDyux=0&jcR4t>?rq1;+xH}z=%=yrxzN4So(WJvHap&} zz*1k2%_cuDTM27H%H$EV@lA9bzV(n^^vN8`0>zJ+$`gEdbL&?A==cvjiC>v>2R{8K z;&KgN3>R`gvm$ZuH2pAE7K6Sd4O_Q{cVzZk@#YtK))M`U_w6w(Z^y35`c#_cHJDKQ zy+SA8HB2qNHr0338iqrPX3P=~kvVn0?uW0Bp_Yv=Tblb5nZ3r{RBMb`ACJ=_noc#< z*P&PDGQ8po0|(vY^?9VBV?$oEy^9R=xHcyp z@5zIj(idlc+7te?u-(Br8Y}8?vTpBt^tc2xqw6z(B6M>XN7F(EAD=St|EJUz^XQ zrC2lC1wPgb=pQGspV|{5&$b4JQ{xRUoIL|{Qx!+hTz*9SfDX6d&>%1r=2I4 z)iZex@cO==6IM6kaWMJ-|IBg%zbeaMOAVc;1?Y5FAx|w`C~k+K;wMu$)pk?KLV*N% zCG2)2N+S$jHPHD}nG5v-Y*|`2nfa(tA3Sq``w^255GS10nB)tTLYYVI>_a5%owkh| zHNNXHU@%d6&H30lEbG2*WgJ%mv$c{@kIL!TRJQZ8+VX2SQBv(|8QzK&ZD)^^*I1C* zqiEk3d9d(C66cqgz2-XART%qbS9|Qk_hc@wPX?!i|6*L=S{Vi(1sj$r$GRnocn% zr$kO|f$l}feoCcgQ45gz>-;@(%LQmqFx>fTkr__h?~0wx(}pqoPwJ|IhyxG49s3v{ zgl;=J75gaDQ2QZR;X1;xuBw2z`I8csq1f z9Q$F@a{~uOB;9qzFXB(W{oD4R4uoNg0?m&`M4oBq!7bt+(wOmB?6|&d`!D@TXjT^q zdlan)F%Pw787WNMXP5~IkliIE3WL)zF1P-?fQ-FXaRD18Nt~`D+h5;Lg^!zy>|fd( zgX)Dv5$RDKm^DRdT;m&!!#Zo(&z{|k6C{Q8mOPr+uwzM9-Mb3xI(1_A$cN*Y^-kL6 z(x*C{WG`RVXL*#w(erJ1=f#^iP?-O`{TqSjQuTSq`%))XeQ8K>C(nc zAkL)iv%j#<6H0FN{(T?620Qkhma&-BfPr5|p+_G%z<6AdxoVm<4sY7!sxVj$$x72_ z8N5f(r}fwB^@~|}31OVe^x{+ufu~*TS^S~n)@E$?|8eiobD|ETvP%9N#|)8w)xc{l zDM5|`hEtU!*KuCxuj`uP@;r+*$0Lo%&pdGIm_f4dFJ2h>bE={9@^+~HDPWSenGBsr zZl%g>*JJ62mFA5huH`!LMk3{tWLOz=?sJeMJLCcFEZ5|l@2t2}qvC<;mY(9+@O6Td zIu!%-!f0N*iOZ0!J*THQqk}(cx-|7m`SBiqgzVpMwoqrUZD=;2gJXg}Au$DNpzsmcQ2Hy%O{PJDFt{$`%p$RJK!l z^Dhaq@GBwXjP;^7g+zW_Z#v_T=RM{*3&TTNS;^h=dY;Xi0^#dhaqy0RpF+q4P>!dp zGPPI$!}Ts>t77b+Z6f?ciq>CDYT}8L-?9^$UzBz`ttIN!E@fJjuPM)gIj!LVD^Fb< z`q(LLTq2HL4)?E}x^t4S<8E;mz5U3ElnprPwfzuuSbN$-6+wc^9b0HiDBcs*i;^8o}gk zZT<`IM2Y>}rnu$eekT6}w!fua`ynprdMEfR&d#vh9|-GXI~A zzdUH)c>yV>dl&WVmqTd^Z~Ozn0Vpku&wbyd8fQ%5)wGM;=$PYj*L<%*3_gNx!5%h7(rsUg8mRUC*4UEIBM09q`?PDFoLfRuGf+JfhBdr>yEwo z2@GrTON*A+BQdhUQ^!GRnT3j8xiGmm>XO-4BHyj~#?H7}vK176mp_)JS_}O; zcdCPW*FeETi8_B*a}r+*&3#(q1a$3p)wsv8oaop1@UFh_bm*t$ODQX|Q!P<{PWM~>+230odxqXE57wH(xt`%I=dFmkvGJbH z*LT)2`5j>BnWk%<&MO#w?)Cbx>Hth$<8jn1X4V0cd9L)G6g;;D`U`Yt+_(`2Pd;Ny zTDb-WFW%155Za34+T2}5)W8sONi37c{`<1GkJK?g-D@ksEpQDHYhhbTL_D z$sn_x-_fDdV=Mo|56N{iMbD7ww}-J>x9fW1M$i-TGH~yK57^ef{^dCW|C-rOn}M-U zz;2?xIj4cL`V#zssLR|ul_p|1JxEeA=fFQ?yejbSmiDy@Cr(^ zO%?L=2t0(T^{0(iuNc9x=Ze)Y#yE)cwSSXIb-H|3GENt^)i-R#X*nO^sj#fphE4Luz;dBWYL0_ZyyL88V_j0f~9`raoxXI?H3Vitunhh9DjADMJs^~=~^D9W@2g3xM1;_EuXu9 z+I)Ld_ctNWN*)KgU)W02;|RkxogQmLVxzDA?aNyXNsbxa(P}uk3)A7qpo-SzI z<02luz!WHZ>TdDC5A<|vQ*hzN923r@{crhm;p@C681o^szxO+c zxd#4yecttHy9UEpMv%L}F)#YVe5q=kA%A=~&{ zzACj0V!NOC+1WH>|5V_r-Xl9;aNTEzCB;jCVesl@-rdFUGn66c%y}2Om)~S)8i(QN z*sti4Lp+#rVS|l!Q3s593fx(pPS^)UwDFb{w@2{NS-pDA-cMK~oUmj0hh`u{Txx4k zE|Oam?at0U$7DT+yC;f*AX*|(urp)_e%+Yq=>F^Qyj}RW)OE*J7<{;Jb=lA$CWJ2G z-e;(X-KB{RxqanWGQ7dLF_sw*f@a@~T5rEd0L`lSy?ZnVj>aXO?>PDj@{T{tTlf7n z)E%6YdaYOnUDM6EmP?sh2y`WBZbXmtoCO#li4v$T~`D6x7nl5~W=dHo@ zn!$#wg=#qRq^|aK-$^X<5VEn`@)rjL*F9(*7o=GqG+rY6Xtv*+Y$K^!lQjD4=b8zSZ!g%+5W?Af0*%?N77(lBAB|;F_ynuNd24DSu(ryGea$*`+K<`{D+sCyKM+*Y89&57TE@R`+$_YoWC(zS%Qr1VAdH&)|VSawX)weLZqv!P1 z?}QyRvC}Hf@nI8Y{&7(AJaQS!Ha#&88aqp7zZ~!Q``{=h?J!?)DS+_LHzfLoeEfJ6 z`^UCdalcU^-d|LZc`&?0>LTi$sUYq>~ztfZTw4P(9u%LP4pNdQ-eh2jG3l0knGviV?W~s2b{u`0UI;6N($A_1t z&Y$^zVr$rfs7C~E*O$eklp)?+Y*SNJJ15MnqrmY`;cJA~=t7ZwW_6MAKik6=sRJYT zya_&vS(&w?x>C(JJ*DcDS-Bf)<3c@?>(;?X<3quZjF~(Fkh^0)m)G;h%##>)3nm?bzF~H{U*ge9!&9-mb3q=llEp<#Nt_?)!DW=5sur zkB4kEea+upXYaK`%AObwx%b*yt8jZGbeGlknehF!ep4jybO{i`p3{2`kL0#$>uLIla0ML|Cfh^V52%3t1Rq7F!fO?GV3*`yE zfAdDg8;#XW_4?gFcwm1nc6}XBKNGZ{{>>2Ty}R4KP7A?|9xH8V zcr27YXr?+cht+S)q#dqSB%H<8gT17SkVgQlX4RA~C18HjRK%CLHCsMZQQ`e8tnDqD zC`A$f326SAg^y!bAgZ#^J0?f-6D_d5xk%^C1S#fYt$Wkx(SWi>CNmr6AFcdiueZ4U z5w)a9OVl2Tf>p=Ed_OOfc)ppv3mQUnOCCw{Eqp*em(Dn*%#P4o=+-jr`xE3g?7v1 z*?R}@WRxf3?UUDGiQIcAm+2ZBoRFtvuKtP|D}0*@*G~W4p96axG#^55-pA;@@4lnP zqrZGJ@?24K_VLqO@?tReB&xhcpJd0t;*4uT>g9SEXqbBN&dnOiZP^JcA6roO=*x$$ z3K!ARZ;Q&Q6-KDvr>Z+6Y6Ejojis)lq<9^UQXwEtDP#+V`3b?nyY*oF3Cj@KpMI2` zXMFCKPZLbor%`{VOGd-zE%%VsVED!6SwK&JAm!$i*v{J{`?Q##y-LCg4Zr^qT!m>+l4hGo(mMK-q+El z;Dm^`!ua?0i(>1SYmJwJ`S1Kn?r}--J_0jV2(Ry(ji)hV^ZWL9U9LKxvEyvrgw~C{ zXJY!}v3d9C*)N0fT9|L7+5e8Q1nP~uJA-WH!7aRh_D~NUT6q5P*$ysoH0%B0N9G7- z4-;c=Tu8pZ0&P~hPpEo5AR*b(bjlCYg{by1|JG%?4AV0(ysn&qMsMDHw?ReXmw**s z`uaX2ay0&N|DlG}b1=Klo%K{HEn0HE$iRO%1hqs>zP|4(h$5`;+;gW&P^ZtSj;C{^ zJSQ5`ytK9pX}}-Xwd;B;Se^QMwa>=^OLiEXwyEg$A>qC76nnEiM9|lxnVm!OET{ka z4dv{w_C5It6Xk!76Aei9aZvEyl<|;42+V)i<1~n+hrBydQBnDq(EOt}mv%Glt;5R7GdN9NuB8l>=a_OtfNbx2dMcE1JYge<-ywru`FL@-<#JsI8 zb)pdJ@mw!-vros7zqwq~JiQlvQoA}X>Ve^BhA!8PYo3%bH?(Q~djbJ7X{{r<^pPOpwZTitB($45`l z^zrkx*6z&Enk{C}0{;i;}vWP0QUI;W{6Cv4c zF#6iXexx)2I!t{sE`BE2=`b;IQtC`O#y>JlBXz9*k_*&iua`m>cJ8j`bPD}4cm^dR z`Mz#XtI>QMu4`;$7s4|KJ^Ye}#rygU#PCa29jMRqt-$aHIUocQuAd!!4PAd++r*Db z{`bCvI(YXfRzyUik^4)ZMNdqTo;SOf^P^!p<_9H-9nKH*Yem^AH_WLXT!eOGPeQuI z=YQITf6cIb0Y6OtrsJYv$=$7?|Go~EU2;S2@s0rm37PHHz`+=@tmVHjiuwgj={O!@ z@tN8C(}~PW+v^_+&zN2KlDn-A7sifpg_vuT{4$hqQj*<;oejnpzp)?Wk3-G$$$yUa zYof(tf818~jREn<&F`ijS5WEi2x?au4YU+F<2UJDk7@;JQkN>9qUF~a66Y!GU_IVI zDJhxcZ=%KdUag(IJ75wNRMWncqQR$>FG4@=L&K0ooHxb85>a5cms>y7mo0WA)Ooc42k8`0GvX6+H52fcLr6Ia&djdRLUAmxt+rbd_6VS#{gO(jf*;yge)Gd-&x5 z`(`Pg(ny({pFs^4V+7;Ihu_0e;EvU)hj&r^5hIS63)f(7&*Z6(>3d-$Cfs7C;0x59 zDEi(vp^L`eFgWYdA3~)p_g~{(%_*O#`f$-yqh@ixS zZT&r@{qccb>ONFYbEs&Ys~#=WsN+LNInXSH=Ur3UTWI}iTC%eY>32X-tnDa{!|?oN zS4E|hEe6s21+&)pH_Y(y(eh_&9!co5yj2jjyapr6jOrAPLI0Xpc&gS@u^gN~pn^O3 zB`+ZlHJ!hw?fxqlHHN1UiBwn}DbmM~ zS<5_hoY`YFk}(bwgG%RBrglSod7EwcBN~{zw(!ABRRPU>D#~Df7XX`4j%_nPHOV)4nnpkrs#8xdoga3BcIA;n3mM|u#*NT({rDq1#@Dz~ z-nWNlf(K7I^5cnNTr;d1n(AqBZejhbEGPN!W#+c~%$K=e5jr)1BjY%qovwlLVXR#{ zbF0(-AtF9tj0qjW@;texC4ZOZI-#+MH2u}XYcLh|=+^Rj5R?i|hvxa6M4NilrA5O{ z5baR&eL>w8Yp?XK3_d+9ev?&Y(`69rj@Cv86CX8N!=RG2Vu1ZNd>!Cfid37NtzhsJ z^UsTAt$+6+fA&jHu>Cb-^QnBJ(pen2(t$#@BbV{yG;HdwD*Z5et^0>{JjTz~?^qSt zr0$P8Y>$7-(cF$tTrv|9l}vmG-CyaBE8Z=F*t#&$(LoZ<5C$~nF6~u)1wR@u%n>AA zQ9)X9AKt_th!d4r36HP9tV@#=r-1gh_UB(=sqz^Xux_rU8!{t@*jAF#%DXi=*c9% z5*~i&+vr|@HqC)1?aMe4-jnuOG)0s+V7;#sR_hc7bkGnQXYR;4BFhiNc!jp)S}YHi zx%TSwiD2v;Yr8G*_H5^N-L0@b_L#i4zG(0lqEa;?;mgqKON}%=S1wfG=(VIlhs7D^ zz1?h5em26Cz;Q7P1yLBPHnu)-+76{IZI9ii2Xm!wlb!=(Z3Ln5sXjFHg8cgiSv5>NzPi4TN*wh%*dO)1p9I69F>f&BE1)3{w`YmYr+So%A>JN zs?pCiaYY#UcscA^-$m4tb(53&4r!l2V=hnWk8@!9hqDh_o}@4SLer-*)%l`QQSIs{ zZPQ)`G=Dzy_&V)jRKZev)q&Cqh1_~1P#?&GsuLRzt$A&sajkv$8R|upk#w%u-BSR0 z*d9FMeUujsE*Z*-Sy19=yr|6<>pEdkeX1i^*&H>WQ`6Qwa~ch3%X!IP#`5hWFQZ?J zZSbH;>AMfz=CQoiDC3DAtU6d-ai(>j$<6!gh;YQL`U1HVn%MQEBdTQ`^^Fa+QWTtp znM)p8&nWPyj%7W;K(KGyKGJ5~n{u4uCrW%Vv3lRh2AZ2&j2SIBp)QbPj=h>5%IS8+ z9Gt`Ov)!j1^*H?#V6eQb%l6hH;Ny%7O;V@P$Z@h;G$z~YjJt@sL_%#VY|w2!khA%N z7E(2CM%$6_NZ?bKef?fDJef$v!tIOAi1@ot-yO??VfHcE%+{%UzcG84vPAn6dpsI+E-gLCdK{WNBpqhD zNp+*BarZkNzlRtOAX{217wyORf?9Rt9by!5s5=i=FiX6i955RL0BCflNLBO`FFkK-8^si z3;nU@5<1&02V0bUV}HWjp?GN{(U%d^Rb66xV5h*0;dJJD6Ah{j1A!28-t7MR6dI84 z7k_`&9nG0t*PeH(hThux2Ekk`j@Qid;N4dD5)5(9Ql1mq36n2b;fK~sRAk*M7|KAx z`NHB+?Qgnlj%XpY1$XSmZD>ya&5L#FULoC5_gBRw-WyZ4us$dAuG;B`Stwn+PJZPr7Vn>$vT&fS$bhM*O{+miv3Sb(-oD+Oy6n(O*ofX? z%Yg>$yA8;hOa9k$(7bK;m+PNHQBQ(ePF@-2=k&Cd$F)+nL7CG9-kT@iqxD0aIlnth zP>xe*!C3ac4&^b@ z=cpYo#W~B4*`ukI&KX~iU&7OkCe92me1O`IYV~cIxo8G2lppKx0TI64NUYdP#FIH* zk5x?YLN%@)nesnKejp4EUb{inAp)Ho8Sh<0Nw{|)?6x)B9F10Z=XrHrVS4LB>0;wf>G^0_ zUeWQ@c?_@KnVO>~cz!!Bx4U+DH#yHQ!2i+lIwy$b%lb#Rl*gY%{5@w95{@fke7QB# zmem@y$!M~*L9tU^6LnSW{$OLriH3!b@@o~SqxM%CH`5C+9h;G$LrK3q-auzA;f%gQ zDU7F$@Jft6gxRs#`!6zYqRta?g5p&qdlEV-qP-G_v!FZUTCew!<0$i6G?|79|JpVUCrGobq&3I3!0UAj(o6nhVu6lgOevPT=t?%ortZyD=MY$Ol)Y)hTsU< zug}u3{;bNfUevoK0#)8A<)!D!!13Up^oV)P-WXX$)9!2truX6s1{z5G643i;knf~z zJvNU17Z2>-Z;Dz=c*py~p8|1@zy5)EUbL7vm(sa8kKBfI2CAglpxRhN(k!V5N40ie z;N++b49?Xs+Ua0ChYP&)uNcdN(0ZV=M2fdKnio1i=inrQmIl7DA62bDeW~99FK7MT zFWEnq_=;I!_QJ{X(V)Z!P%h8LzW0JU41LXGAC-ND%C+`09(P$nEAoOg#YZqc+D!A9 zbb|{TXiM?3-8tn8keBnS(b?1mT95AHy;t`TPv<9QN~Oq*I_orQ>BuHvwsOuTI$Rg^ z%CjFam2!aIuj5xd7t|rwQs_XU#6?6P=)3B;E~9EigN2LEkx+O|)?;@IX&&OK26+}E zzXwBUwB9v?h1aNT^i=GDDFNt<{#j*g><@8{HvQeFvAW5jh?OsNLCY|__WjNE%58c) zsPX;n#)528oE7Fosdt$lutoT`(}#|ZoIy*En|=jy3ZblbCpD_iu|sF}=9*1SKiaZ8 z+TT3NgeFa5Zz?4}M}r?8bHB!!0kM>ta^O-ROx~2H9pl!4@thBM+yM;tl6qcWf0LX8 z%2MXo{1mWrZswl9oR$&}M#FnDEu%1e$^!lMRf#(fP(9C;xftEUsOfEf%@u+3c=F!_ zLCtGnuohlc^-M$?T6dMDjDEs+Rg$lN)}|Z^g&8CF@dm^VF*AQo+brCJ?dLoFt4&nNtE|*uE!kVR0U15t%DJX#t%#`NK+fCu;Z$gDA?1kDAuQiE%{BJsposviUiM0`9lE}4 zf0}qxe30z|DQ}K?PhO^DtYJhYS2~j1?Ob4@+t>X`s@1k}UN9>*dQRDlx&oemK5fql zGm6yGdRNt<*JSg)!Vc28gnHI*zWEUIPrU)uryu4RD(nNzvd`ExLO9X4}XRJ zspzI5kB=}g;$MD@+ZQV4n7O$O1W3Qz`1+0ODiaqB-6KE4v?2`scYZhPZ|b4G5?*JP zVqPflW6?3`U<6{$Sp%ayXS8;PCeq@l80yCdzv&M3N5ihM&r?Q~VEWzYfK84s3MWK9 z56ex53W18KSL5XO9W0B^23y zUf<#-$-jnl8~(+?&@`y$ON`gh;Y0(ei$!E&cc4^5pvu`!s~^>elO>n%uB_9DEw&X2lShAYCC-l9=U zCn{N6HZ=3_@e#L@dQ{}{C~D>84nzpCC0Cg^huTx{qBM&b4`y#hMbwa#2x{2r@`6sH z5KQJZsVxi zEYQOAfGlct%`OQ&K+@%bj>0zIU-l}{@ux>Kzvv0LB%ILsIFHrE{+W<&tk&8KD}x&8 zJ8)yr^Cc$Nt_{<{symHW=e^g4r@oA<-dK@_q592>6;405t(&;rCQ+8%FJZ}JnnKO1 z33Ya+>e@f5hSAn{X0ZYbP-3{K?>nIY&EDs2_uRS;4a~P2^=}4)`>OixCuhDu1sm;q zm6ifDM|+-q?+R9j-<@)I!#j%=)iKdg+IqX7rK-gw+Wtb+k)h!ivq6gI;K=>$KFk){ zqUMJ$9O+(hK!4`%kb}3oQKL~x$>pypPM`bqIHMAQnE36P}1Q`US<{7fDLosusWe+oQBwY;m=wx7qC2E z*uER8=v`ziowv|JA=_b5vJMyxiivE%Q{AiRiX5x@%Esp%-FD5xfEi~hA z8Z%0afPpcQTp}(L5!%Wmj@~Xt!zwZh&cg@_pAv?P9scHKO^@Prd< z=jzP^FNIlE@nl_n%ta-vuqK(;fB!%e8nq^S*6L`2C(kPo65XW;i%|;H4RJv*BAAg_ zXl8^KFRLG+`PqTWs-wjv=iE^JhX?1DmL9`~`khJqMl_yGtZK(dMB`!EmW_<8%E2{@pM; z_eN1C<{-=-9iP@?(}tC~DqN5LGPZuQvL-XnlJK9f?7}{Z+I-Q{l4+pHlnH7&>Hn!l zS`Sq%T=5nlC+SF`Jl93Lk-HKwcmAb0>omz;gz-{=&7-+3Fy3@fdo50B+dAp`aHqP; zQ3OU9(za|HlM!)@lcKjX779KXoMHJQf!gkhj{Mo>ha>wPXL54E)HQs{Vg9~I0caW3KZiB`Kj){XedkbvA`cveF$=n7Mv@| zB7?kLFW#BGOGo{h&mGo9Nd7vW`p(cR)>qz;zSDoxQ?dv3%O6dW+`#|~`j1-&q^V)9 zB3rO&8jC}9AM{Pe)nuXdZ0^W=+RQLiI>|}2-v>3i#}3~TjYsomA51bEzeWp|VnHPg zI7B>e!nL*f1W#5;%k=F!)aQ5kGq&Znlo+A85c2vnuxtFac0}%(xmHBhO^EYzP3aQ z9&8ubcZQ1J^I{TO?$Rtl?=uB$yfeC}x^>AI5RH;bzziSXtUxp*!u##7Sp z=>gCFS`rQsmUa)x6C!nCc*i`vp_)P~uT>8Crli3neME)sG;f9a-)KmM;g@t#FEoE)_R?h?kpw3!j@^-Zkg3M@AtIL5kCbw&V%IU{ z3j`8-VMw4zaV$Oz4brq-&gXrFiU$`}_iIg{{`~_@WEL}MP<=C1<^4r8;1jwKpB;_f z(VzG}u~LYpdkXl^+|YtK``-G%N)4E{>%i4!GQ)iRZpL^~Js3GoOn7;u9LKMOkZTe)_OVE3zF&k*7dj zXEuzYJ!Ux)?hnA!uYKMxYaOAG?sget;7c5Nd6(g2a`Mg zLksdzGndh^n1I!pJ8D^hi)h;7ym^R0HX0YNXsxZl?AY8J&xfP(PNVwI-CQ+VVsgkJ5~0ZO#^%!ffO>rRD`A)Jt78!Y5RL!k!;{MjmznbqG3hy>Hx( z%j@U+=+O0rgbRhv6)A?v7IzdMhi>FQA?flVqQKghCgbC1(2nQ!#??GLm7`cTxr7)j zr)(bS--)f48F7Zm`63}8gkAmqBLMUJ`;$Md8rOD0=MyT9iPPtxx~sDC$RHNCojaj) zz(fwKe_isKe5=IRgN7PD{63MO2vZX}@t2$*p_ZapjS3+N=wm%~$6)Xyn#c}ZF*t_R zt4@k*Z5cK5sQ%ui z=tp1Hp{(xePBXc~Xm~`G*tSkr?-J&y;!|KvG(ta^e;N_gW)(+9ag)5cG6;%p z?&CN6br?^cxU%Duh$ZTw|oL4!V{+Q z(&^||x#DB$<3o7z2aVc6iX1S^*sUuaLd4=RVTYwvB{5xT{0Y>L)+S{9{J*!r=!D?Q63UbLU30QheUVcUnxgjJVrZ2$8Bfi8=Hs}_Pc%6y>=8Dfh~{ofa6Nea1ugHh{I%nk7EEX#?-H3O)wQ6> zJGX-?8tqYQ+l`SVVmSIldG79~8GbbD-5Q@r=YhKU(n9H^Narz{GN(|!asLpW(v$T| zsBJKgqLpi~`_+BauYNYXIi4LF=gw6`P0OPpeXGVE4ia7r)m?d!_w5D&t)4Pl(ogZ) zHXd`5@-giZSpU!VAG*=AUIo2BWlmhG*Ft>+?a)o)0(zW&pK$u@1oCW*(8==sh~^JB zxtvxK#M8vE*r!>M^sP|&m6o-mLI_W(P*=Y|_Y!q;B(j~KqLyioeWgk{B5LlaU}MH~eY>toleyeEgZ%AG6if2;U|2@{#bb9fH2b|| zv2LCP;oCQ*IY!mdkXM%HuDh9N;;7KZ_3;`s<`AP}b=v}J3zlaX%0KCFI_Q+@J1^}`YsnrY=1N8 zaB~5O0Zv=<1{dLnweJb7t6^xQUeAGD$O%Vz>yX8ww+w{Um9)k1WBI_gPdjFN7Y4Vr z$LFBO@SidqYTFxcvZ*YKie^*EyfPJG+9<~)dt3pCkBg+;=7kVytnbq9VnbsKGUe$~ zfiV0mcXyMtvzS7BtAQI=P~_>woK%x9;(T_$lg}f*Y6~`>e)& z__a6kU;6->T4wH@3Yr3fF$??YwF5Atli&VJ$P3NWOD#rozr)iWbe{D3E(^Ulb!y`4 zvoIF?>wV6QBTQW1M=f^X8jjji(_H{%M&9 zJ1EYes0U33D%x>4T2m{-d7)c`zN8<^Y3dWK}gV z@Lyw(hj_8u|`)F>2c_Ms59oo0x@Gj9O(6JO= z_ckO3EnV&%QtI!JhdaUih6pminDf_kRJmBvxGlY-b z!{%G+?PqB{B)vKqw<*{BW<;{jV8JISTyJIpy>}#Uv@qmEp3i^oqvx|m#ajF5wU`Ug z+tv`7>?UjYb=b_cMUcCcBOW}@GhN| zTq#!uu(;#ApH;3A4OR!zc;4Enw~?fa1yx_ZuYINSMH5M@SBh^d;i-L$b2`WUfQZgd zXgKeJ2CG*ozn;3mh?Se@FH|Vk_nlwjfBjRp5%s$tl2Jp!g!#?ND>zC$WwpIGj=-YN#l);k zdzc86*sG=P2dkc!I``1MgaO@dYG?AD5U*-Tq)do!|>ovP{M^@)<6t(-=bamFW2&^^|C~C+S%dK1}JR(Sv-nyzA=B z9h+S+-se6&mGK8gW-mxZ=U{w^^(THTpAnY;LUel>-gFvI-l&MHr`?On8>{RNJ?Me~ z9p+ZLGEfnZ2W-TLL47~}83FTAV@FzXV%{!wX z2y87O^^dmU_E7s7{1lZqhBwUk5J+?O{dWGiMQy3?w^F{qj8bw`8bul8bBvC{ge#0a zmfq2jitRh|594bLQa&T%p^@?UctLm<`(2Qwk^#-QvDL`Mh@e5G4cS;htiEE_0H`g_ zrvJ4g^YK^=V>=ioP@h|g+P=ha__=hgggMU=)z&=Nr)}_$Ub5gqx14a~S3M40J-N;QL%o^i_s4N0yA5j3(nme~g6Z5)gm-_xQ?d+; zGzk&KyKcgd9eDvppD^Bw5wWPMKxZ9PBjh&f+2#U`OPW2AYl6`1e?w^V_HQ(&{h^WL z$tyJ1MNZ!CkN`_(C0(u`K87ZFf|`P!?ZJ^rFqf&(Z9w7p@i0TiH5^%plx+=lAsRC< z2`hPo)xB1pq#cd+_>D%B_Sz;;ra@oAg{w-w(P+FOtMtg{VOY89NZpW?4rA9>2_t)V z12IrCm3l=6n{OZVHX3sQPd2_q*BGthS-^XkxTBHt z;fNoqt03F;u+J14j7B-`QO~2_{sN`R`>(?+rS(1LBvSkaRZ0>yJR*r`LAqGei0V9^ zL6S*slTsXNIbJ6-BsQY@juiKnlO%r{#^d(WdpsfWg5eorgN|Nbih{;ZIMF2?48Pl+ zM6u+DH-+(dmeY4{WA#oGf0UlnjNC*M_a6j&Yu$-f*cZ|}&R}@u9xaMcMSJX=oI5iR zb9;CYzPNn*=-PH4Dwv{$PK%Q8*w8CzvoB=W1a;M|X7m64h-XlBeIiQj2wim>ZMz?{ zLSMk%6!h>A3|ltSrA}h!Yk5q=ufYZ{h{<8YVYhhNwD39vqSxY z=6}j~UH-m*@J}~u67MS-;ipoKbi4?QyW6HJxiEiZfwRZ?Ger_iIm$-W$#B3@cdlEV zkjr0ty{I#zehBsbVi4k}y$Y?4(;GkfJz>UdbR#ybYTLS4*c5BtCx-Fk411=Re!PhB zOpQ@Tn;dzL>7sk>TXzj(A>nmUv8>5zc_YSC+rR=>2S0}6DTR7(J~btsn`miQt^LzG zTJZK2pA?xoE3{bEMOa6PK;Ef`qD41;x)dMENas?)K-@73Hfa(b7S*(6$30HM^qgE>ZO(iMFoTF2 z@&=SYNc?B0>}9c|R<|9h`8KrqaM%h)^ddww?GE8-BF}JH3wNT{1wIGyAN!#@!l!Gc z;tsg;vcByhJV04;n%$#se9)5GEjbJSHZ&6dN?KLk1xI!%@AB*7HZ(5M6`|Rn31wq= z-J`AB<6XqrN253FxY64C_Wq?rMQGxDrul6OvH6+u`5J5Leo}n~{JB_XciJr%2;+w`HkU~L1fKl$o4Y-hSiGy&SCdiq zxB*NjD>{=alGZ62nrhND`znuGFTOPKX2j}q$;MBsAYDN;qK!AHip+(c=^r%AqcfihZ{x5zQvQJ4&W7Kx8+M9y;NJyS7QpMtc!lIy{(4nI2DJs z0aAVh9O7=C2ni+OMqqhii(dFjDlEyE^>+BMZ@XXf>Y>viu)VRQv)D{bLd@ zCDh&2dSEh*{jP&MZmPf+)7zc*)jav0hotuh<6W<%l2~L=PBEylO7bf4zVW?47- zdV_6h<_%pXRUeId8Tye6)Jv%1Mk7AXGRohL3=3^=!A4hAV$&WsoNi7Gq^~*@Y zf6ji_u*RQ~C4k3R)Hnrbz=NS&nt{UP2;$~EmUufoNOm)d5xp-fKe=L?t4 z!r=V}_g4x|Kuxvksk_=(9;7hUFC!+3lxIi7w(Oka3MAeLn9jkU4O}vSIkO$l$H%ta zf7|m`i`BHUS7w1QaO^8{eGQuZVd(HEB^Flirnh!K;(+00q0i#S2O<5$gx$npT{L~j zEhcC`rmNnb_*_-_lo2Xqd?RYk_Z-h~H3>hF&HxK5znM4!Bk@eT&7ETzV^L1`nPGMB z0UVv>1<^#lZMyBy`b#S6g!AOK_EiNRP1s*V8fO@mUJhX?y#pf*+&oIpx8vW2r_3Ix zh&7>s`!AbYc5*>ErR~?GGIq4e^6kcBsaf#b?dMbbj2S9emMWi;5zxrKLyunsVfEnU zJ9|T-Q{SUjM*}UTy&ll-Pt28ldIDPB$EuqWPosf>b)g2KAs}q-&M_-_B#2W=&85aGWryNn^o|mC1uT`Pv>kSFQ;0v@NGWKV}+yNRJ zb~R6@R-x)l%}@H*p8son@r}Y&+`e^e+jz~*n)*?hs>8C;WLTM${^l;EzyIFn}mQJ2ry7q;{tAHn`@;EIRiQ$w+W8_BG9S)$T-NeMw zL|!zp@+-N~G<5$Uh6iWb=X#Xf@G%O9%V6 zFP5mh%=g^6SRNQZAW3^b+Y1OD(w#;hV-b-avbTwg9vp()_GYyXNRwcIz#nc$BN z`{;UTE;NMOMt`_#h2{yi4KD45XgSJgL?!tm%)GI_@{!>=sZ=;C(AD3$V2j69aVXtm*$np4uL|{bMw@vG4>~Dz0&=$mg8ZVj~sg5 zax()@jm!CR@%kYkD*Np&WOhR(jB;uOOAH6K;YcUOa;FHs*=T;QHe*L~ws(hpm_*Qo z%7tf7b#kF(-}A`s`QOAn8J&u5(=9_LQB1Uegkhy^o*b$!H#N z(w$?3vfRUF<1=+oduxdW|I-l0swM5mUels!j_dbZC{@u=b6JGM$P<_ze`@%t{23Hc z3%GpzIs?d1eBZg)8?bTI<3yHr0IE-PHmn)pMl(suK1xSV!AynwW;{J;l(IS;?=GjkrsIbYLX=?Ze8j@A-vt3w$1=E$W zz9)^SdTj289Y-pfh~&F*R-O))Oyu9)`00zMRytNN6xo8N?;VI>TBAY@YhRQ!F8IOB zwYO#01L>esCabh}6T`POX`UIOk9iFBJEm@E@{#5-IGt*eRFJ)gM&iE6ISFFAh-pGi zBeuWrs3h?^jd?&0RQB?fNPSd81AMkeUz=e5?Lfw%IHm&$(A4d<-16=KTB}sl?3Ysj zVl*C~nLvQCk)CviEm1V^IHlA4LNN4L@wh0x!St{fD|hXqT{?{?UxL+z*$8OKxAAus z|BrYFKXt0pLB_xPQbpWrKN4RDp6s)+i}x{F6!+1Tcez*}Er|+n(Cqhs30=389fKHe z@M5zPF_>2XYHdyT_SmRHJ<&Xk7@&tjo%+pUPyR*&=g2K0lqX>L!j1-x(nm?j$f`}TkJmce+va~2Zk9DmB8p#W`%$9`kjep)@~Id&4mSuxbdz4$cj zi+WQh_C0-!#aG8(?cCb=QUH6-^7$lt-z+rM#Me4;wg9`IJ*(dl!y7R3nl?==`vr_# zW1YEpsRGp#4zP``V0eYmNrnjak63*`(#*i4o>?z6cRS?S_}y#JndU!8>DmViQty9@ zTv37z3CS47-#jpw{?uc`%m@0NSC4S1lJo^o_KjA$zFR&pRpEB*hyw}#32P==#GR^^ z&@aMRw5$0C>GM9__+Gd77~Zh%%01llOPg)$<)@ZNf)mw#JR?2xWv16^sNKiqdMmpO z>OD=)P~3sVlc#@3*X{aN2GgBG%8LilUst%IIQ|nUKiQ z*-bhy**sq`Zbt``w+@@$eTn&(WA=OA+<%AF+xYSspYMHz+3Bt4VpQ~HdC`<3S=vhZ zZdh*L#ZO*`?D0Y{wD5l(P;(f zbUc~I)rRpO&E{M9XE&4ZZ)ob-(|U%-IP`UWS^g-Y5oY%(SgWcfqZu9jqmAei5Y>WA ztt3Lww^-W(IK}|URn%+pV7ig9 z80v_7H~^9hKuEJm*)N`snxu@AOmb;h zRJ!r?wjMzT26gFE61<9G$(BqZ_|{82ZI<=Gt~;YJ=qRSS^8F%OdBa>#%R7ToJfHFY zQkR3-H=;u#6)LEiZQ<~J)@RT?`XyB(2PjJkwQF;e4fu zMOg!-7w6)Y9QQ*X@#(7aX^gL^o^K;u)(GR-So3XI`m~e_OFC*^)>;_9ey$NM>-Xz% zXrz^K|C3LcK5C0lZf4&VJgg(Mt6F`VE(cV7+N)m`h~do#Iq8KtzW#>D1k>~J;h4VT ztf#na&JBM^9?PUM2-QLrkvDVt4`jf=mZyN)uS)1s;drjemk*0U=}ljs#J~z!-_5SY z66iL6ey(je0ulE!rXEeYh?=)fey&cjfk{&-7b8ApMAXl+xB5SLw#g^r!9zW$d0ozY z(&82xX;jQQ=PC-VK?M1)w7(%oSU0dZ`UjMGZ!LQ&l5h*qbk@ge6ZiN0_LP}w`g@<{ zz4N}D5h8e6iV&);a}v;gfB5s_@&&YD{>F0a7)e(a4PB~{dY6aAo3`9fQ7q_lpjMs= z^Q&#+@X6fA(!W0jRu7p<9sWkr5kR#s7>zV-??I>4iyhIAF`bT~vmNhmzBR@(zDnZO zGPZ$11JNGqJng^psbt1A@$I&L?-l>LXD+V=b-jt(2xtC@Cg#a%(k-!bu96B2Q3q9s%)o7_|MzeYCfb8mN7;Tq72L5d)m>3dkvyEgJUXBAJqn5F(n z!~pefzP0hUafQLvGm{197;rSTIP&vTS5be?pR)oE{*d;gp@?rP9C9k^Ii9My!u&mL zI`-QhXiPL%gUv=4nl65p;HK0=S-6`AOKDC+->!?YFRHcCYG-%xr>NVgrl*Uo=#&Wj z92dOHcjgLOiNkdy+h0Uu@>~1d?l!{Qkz7k7$3|3F8ut3NtUKxuXI++U@<8K5Q*u?V zX)xgQf=6VN9d!i0&ydk3`J*s-$Z+)GvvsJpCLE8Gw}hgzjrToyIS|q9la3!b#xGQ7 zC!#kb76vnhd52wjvmm88=9DvMC=hqhmvvH^oVcs~+=v=g@mwMMXdRAf4OUOgM`HYc{rj10 z&s>v%38pf!Mky;q$Zw#(IX4cu2g}q>N$CIgJVDDZSKKu?$MjtX zs;%R!Z%u!EVt?OgCmkI5Ux9Lmi7RJ@kXa zmKdHoin8JN_}CPlN?&-I^VSw>Z1?_F1uy zkH8E*( z^=qyEaJvYN&p*v>c|+2PfthjNchH{l?-weJ z;=aDKH693gljdHwBQX5%pcj45I*je+|28zPgYZmhUO!v*L01h;Yw!yln7O_6%m3Il zY`@d#GP+)3iB_yDYf4mDzzS(edzXyC5=#7Kv?CS`2s%xZqiMAGsYbmpiij%hWafQr z&!b-RbBiZ`Tcf7yyWF z8@sySk{2^U@zv>_>osyP!u#rB&Ys6;nDQ9wdfx=}F%xSa&r`xksCK1rAn9Cy{;vuq zyjHhR!|Idggkdt|z);%U{ayve5(}@Y2)u(|nqH&NCj%kc`kVS4d?qx7yOh-^d808r zzS{nnH?)nrm2X(qLz`{gMYf|0Xns9FDXmKfmA_0$(laH8rIuG3Zl~ViCzg~TM>2z(jKI;E@btr6CJ&HWB zgIU?k5c;KFv_uoCN&UR^gUAgr9cUp~k@vVT#zT|$;4_an?>u^okL0fyuQsk0ZNGbV4*G?Q9wZENqPi1C zr%n$ZM|GDt_DGJ#;~B3CMDk3^qoBtcg&m)>VA^c!>`OQbspaD)Z#HJ2`+Kh@q?r%pJ~@yAE4@QEPLtv zvHH|Od8K`KpH~3k;&XDnq68dyNSH~E7?$VWTvj|?akUJHX~rq@jU-A2Ni9f-Ky>r^je zq2Z{p%aGHlZTsl*YvHjKgO&gOeA|3%=e$6>kf(#D3tD5Tny~nOb9MMuC|@Rwg`Zim z!T71sNNHA=6l&UbpPUH#Q)IEEd4x9A@+wXy<>Hu5T?oID&Ij4bR*F?jm_0t$5!*OA zi=e8x_ff6tUT82TYOUqup#BqC{w?ht(8q9h{p_3)_~)Hi;A>e#W9(0o`TX|6RCj!G zZi~A)^doNyAlmCEyQ8NC-&3#o)+mxwJ{6(7J~lT7Z!oJ&7bm-Vm@v;#ur$~l{UfRGRMssX_a`kgcFV>)r!+cWeTv^cAJRNZ1Y$=RY zCH7ryzX)TTPyIdox9Ru7NDgIQ)q`Ewy2I_S+&V=L`Mz%De9;(>b*ayr6s00pm<>4g zAy}ylvuAc@+{YU{`schL8>M#x?n9)+=b3gZHG2 z(+oCmzWnxZr(lPX*?s(sE>5VtB6g%c_yi1|Ve}1j62OxkWNPp|J_xhaCCh$@8qHIO z95hywL?ha#CLJtiVbxR5!63s!5Mj#35Ln{JhU&;-S(bcb?R5Xl^n%W zl0R#0#Nvo(NS#scfh@*vFl82SY6w-sl;$rrM@N#*A8g6yT;1tq1f6XR&Ou}(9$B=a zb-4`Bl8C~{b(2x^F=&Zu$xd#?^ku@8$HvWGXM^t%IgYeCJ6IKRePlRG;;Y1wH{&9> z%&wuvPXor8$Qg@!{LZ3waubKiI$^!F;QcVSZ!=T%>u#t`rF_AkR1X~%yZCbE-@p)x zVq6X9LH+Ned6f?TL<2TuXFQ_Lq4|%eTHwK7zr$EcNVv@u`i+xgay{O__@A=9+KDI8 z0)L21tY|v)-r3p}e5)LNBo~+-&A|M>(Z#_SN8wgzjgYR;O;Cmc`L12D3gfWWzv}St z<_~BK4772ulYpA^iGhlSX*9Ph&(`y)IgX5yo&Az$H+-hMYMbDy1zvV>UUGlWnZEr5 zrDQ8jv|&SgqVEI`bZ0&L8O-gHED>^P2UlP1w3m6+s-9(IkFY zSagWKbHF1E7UK03r7qxLA~*Eeh1KI|I3%v>o8%lsv%Ssaxva z_Q%nS2g-pN1IHkO;Zyh~z85VIT#ct)vPD}$v}H$x_P{2O;NIN{tT4wFl$K`r96BW< z;{54G(7gDK{0S!oR4!6S`(bYg=8trLyyyLd1!nt0RLVxMI<4ia?GROm&7TI=%E4bP zbg1lA8iO?D3R>FX7?tIGZ z|GPgy+Vk<8B^gqEJDR>xP2QEsjV5FizFB4+BkddWY7fgUj3q!u?&(RDWmA|OWs8_h z#`wG%e!o9tWOfeq%lS#_6k`0Zbptw*8`ZWbdMnCO)ZPZ>8NbkHu5~u7(hz?A z9K^#^qm=T`Rm{#Amu08C*Zv>&M827`I|C_y3^Qy;0$1L>K(o$QBmSV%Fmg*EE6Ml* zjOW}N)E*|q!NFB4mhJKe#=Een_Db+(8Hx7`n#Br3M4P-&nwRp8qwQFHaVT`vHPny= zrV=bB?tkk*OGJ0=)gsI;8SDr17smH-6zS?M>rCg-cNs3fJi`Q7cj+iiIw^>1U#fFv zTVXhfvG1+0*>n_oOGGYu`!YaV$JKxivwYO4@YSb>Hw2|-IOHCl-V1$ILDQ*wai}=s zit&TiTo_q;ve(4W2@Th5&!N~O_T5fd{vc)9p%$uls6Y)Sk&& zNbDS}-dp%VpvUm!T|2&9ZK7a+-<4D^)DmN%mv!iq+ObyXKPTs)wPuEm_XVZ76=n$- zNzAp|i3>z2zfx41N)n;#nOpb$yrVFw!$ec1wGTTtrEXWh_b!I9-%$~AT9_R>EEzyl zjyHvtB5A6L3KBj8%~@ZYVd1)kx++gN&J{d{1o?O4;vPHD3T+&R7a3NsKFE1Yw^0|1 zGe(_?l>5AV9fnn>*SHuPq4P$Dhte!5z5zw)uYVPBk@VHj?1s$fh~r=Tn09F7B;Ena z%!4|-M|T6Ef_b4#l^@1R2*EREwrG8YwmoSI<9E;dZnBc{7Sll*ad&%|^F06-oZ{XX z=r%#4dR5R|&?+i%usEX}@pnDOc-1c$k#PU0gT?YXmEX(%-UqP$I@GxIl^N@&HH{e? zvt+DJxtS)|`P_;+YTa=@pjA5*7JEMju`Zs5{$c+-0sC3hapsb$)cXx+so46gO1_;( zVDp(e-gXMpi5XMXud4VKfa>~GXn4c7AcgXo-xc@YXsi5+tn}agVjzj>}AyhiO!lVZ>j8=gw3a>hJ%4+;RQ#U;F4L@2O|j z(Ag(}kFiZgC8qw_1ab!$yLfw@<0;11fX|Av-E9|_D!|7kco%F+Z<05VZ@{!kso_@^QemAmHU-H6qXy%@}oTgIx3XT3JM`i>T zVaOmsx5J4kOL>ZT(ry7IXJStu^Y`n!-1( zaYFsZ{zngbF&pQ8I!#G8K2RN-oB?sZ_Lnat>MH94G9;Gsu#~#uJB;b zB`UvF@m*0U!le(^hR?29PUb*wCtq6AVHK>so)+JZPhxpMV&bCf=%yo#Dek`PVblim z`{d;>tHi^A>@DAZGm@PLJv}kHb;X!oz!LvlMX+Ko)MsQ)d@kz!d+z&PpVYu~NGi1@ znAByleYLRLJFfA#JM=skktMwtOd?Y{Bg8LuEXNYIvb;^4hhu8{7Hqhig4P>D@ul zsi`EEai|jYNpCW$9uU}eUC-0H;tQKa5(S<`+D1Y%3Zeh*9Z6K7YZpv{G*OEb8nBz+LRUGjtAA}N0Z zHJ1DEMN|zibk~K->7glF7}EKG^&8E)Q-#gXUky5G}`#eaq@%9QA(g3!SI9f@}qFG20&%UR=YcM&me=hW>- z2555l&`B~Q0{X7W75?>FH=5a4ut+@V`G0s?y*@kbgw^kGI-Jf6P5p^K)q8UbhI77o zy(Yv$z5lJo{b@<4-%ulf|D73Be&5d}67~tEWnPYiZ|50?!4L*ziL`CH%B1i6I;#Wy zsPs`=$RXWi{*%~wAo%AqhXD`jSf?4*Di^~sY|7&^KhVK^EcX?!Fb?RFqZFPp#`1iV zEa?^JgR%Om>96aBKdBVaV9?HERv`w^o8Y(htjibbZWhUnN>RbQ*Yxy=)DWsm|8Sx_ z^b(p=sLb%ZbO5FdoSJqtQFkOm2;N>Z9{t}HV2#;8KjG?6fpJ1=XN|Y}u z;+*24i3XF{JlFkWQGvg7Q-n3i4}kgVOM_%3%g~loeg34QDZJZ?l$qEO0d*%|9;vwB zj>@=;#@75iVdWQ<0^#>XH03E1lxg}LmM-*gUW)NWLsBx`k6cOkXf#egH}^ZG6o_O5 zx_K`we(Uk3I4o~UV zsYh8Bg(e%61*IQi{zF;CeqCCs6_|F-$tZ#0)JM;Zko{4wMac$SEhZe?Xp-mpR^weYsN;0eOqDtXb2JJe zuP1`>;c#v7vQbk|s6(l$9&lXpL?9AU9#b)^F$U#%~!?AxBp+(uyEBW-l?Qc>`)aL|wW0Y3FCMl^;qhclPlHoBI{vvn ziFrZ?moA(@YNp)* zLcs~JEObErY+nr2QYXFVvBLbKiUWB@BMD8gC{7-g!%T|jK;!4`1hzzs*L-d1;Y45R zYZ(6a>*bNF`RM!e=A+FXd!c2@%H5Z*9C~w$&Tx$3@w9Kgy>?{CLd!49I-{jkftYnj zCQOza*6Y2$r0A1)D_~r|_|i%B1~fx zHnsBHC!$s_`Z7h?zw3#$;K_Q`by!HS^S-a|3zhf1h!2Bq!JO+$G38)xn4fj}{+rJk zWx7xaHy5p=X>lv_ccW@hZCZT0%kMvQ4Ji)4G^r!u=}=tCGxf@VCe z#`P0Jud%paN5c8>7>g@-ny>E@6mUkcUVeX(wfH-pf@NaDy=6CAut;Z7O&x$)y5G1) z`cjzY)ZTsVSSS?tsgFO(w}xzTj&m1d@vsz2KT&MJg{neWMtzhOp#40~jaotrstrFs znyt%)HZ>le=j6`Y=70Iyv^$x1>R>WYC2@OI5%k)9UK7WigK zIcP$Dsy;LQlOxbCkZ|7VHVOZNHeYhx87ZRyf`-_QUH%wuue-V>4b{ZK#JkV$osCFyGWl%obq(>(3wgp{HyQeOc-U z9+S18T6wlxUuo0P=(~Hb8xIJgHu_KR3yjbIT?dNmpT{IozNvC|u**vr`?d8>PFDqm zops5-Pw+;4o6Ft_2WxQT-6v{KI_!a&Zp97TPG*>@d}Tc>jPadMtfeRJpriuAU0dq| zvait?yFk#!!(TW`m&d!p@RG19cKSoYB}?c{y2xL8P#9$eXV^%4knCAJmBFp2p(S`2 zWuJX?=YcWQ%1~c*w#b77q9skF?AJ7{*drO^_ZE?Odc+QT@UhCP?DBIicGI4af= zt`2_pe@dcTUKR8lMZI+m%?;Ps(9%XMa|m-GDtem`P^nT66Z)*r``_^)bktx@qva@S zKIi5CE4>(ql2x}-UyPy&T4y%g2m!V7GY_4fpoWF03K<*oM`$>yWap+r0vZlCVP-WS zg+75Ca|JgnKiSoeqy94!j5417e)>ux8QX7b&bqFrr2h9j$5HsKox5NW`*%E3=YF4d zgm!MBQ?%@y5R)BMIiJf7#3H8?{ITIiS_LKW*&M2T^IE>FsPPT4?p% zKXf^G5rz~aPgH6>hC=(mx33NNC%i zQ1vsw)19>%>JGh)-_7oaW=aqFZte=j)2%U2r90taoN;pJk+>yTkQUe~iR?q=p--KZ z)fZs(blWc9URIdWKliz74>$N;e`b4d@+8bQx+UM}S%u;KA7{TGi$gOHKb`uX<%*Wd z^b6d!)VJ+ZJ@<^*;}R@z)U&s))z(~x&B-4N)-F`Ac+8rk^%^N22n{of4N)tP5Ha$S zWb{pW)EjntHcs#ZsAV@i(rxW-`-PItdagq6v<8h%{}B{l z$&nUu`y&kSQ=MUE3L~|nbc#IhXeOr9GP8GV*ZUw+JtG>pL!IEVg00_?4EnZIt~nsy zYd8CO+!=j*PDyUBj@hI0o;-m}s*j-PV5Z7f>JZdPUST(3zzf61O(LzkvAEvgdao>p zI1{XMdwcO?^-HAtjFpXP4i)W!{`N_x*ySBCS+Ba(Zmt7^)xuvR>@a_wz=U?3a43e^ zI^#F)HOY92Kg(T&6Que>@C(>+v`_Rh6gbOrcx7DxqHF0v>ghnRBj4Y5$>;#o4b>#Y zSp9~nFXxMM4!T2K)t*Cpan;gA@h_8Ik{T-no&uG<>NI7MB{aRxZ5JNd?pHuB;i^*nHc1 zn{!6zi76VgQ(s}d?}>)a-m>^yy8{&xmc8;;eK(A?0n54CTV`sJs9EAc+=T6|zkb>4TS~9=(NL~?%%kO7Xl$SD!IwmgN3hh8f2pTR z3Z<&i)SY(Qi6$u*o+{t9g4Ju9#1DH&=Q|WB>$n(N?}piXakmC`8o+`tKC-o`1)BVJ zNkkX-!0K4Z-R6|5&=km3d$hg~WtR8eQX9Dp6EnKJ-xH@%+v>5-__V*{(EMI1DbE@Jy}sIah<`Wd59V`!EghJZ%fw@`s`6*`lYH!6O*%*yXA6J_9XU z=RTs#@rPl(i)&0*)M5M>xA+EEDg1e2qmh)o2kK&8pX0o+jQVAF&vBd6LX{}r3dJ*7 zM7;12xn3dZL7|nANQJ_|64WO=vY|Gm4zsj%)BRtt{K7($$4X3qIm`y1dYUqb@pjEO z(cR;b?1a^b+xN?YgkVH%OU*hLtF!n!mnQt!`Ou;u=K+1mConuE)-6VV1}2M4)^>fI zL4}+Hr;aH1z>2xYC0h-QH;?G%;_1@v50g3FW^Me=P+4$(d9(Qn^p_hMZzy8*2-C&a z%H&4HAST*-Ry`s5xm#$aiQYgzsJ}|IiHw+r`70K@WyU(#N zNVAs+qY2hULNRkbj*L0oOQezy*6%*j5vKKkDYmFRUtCOKRolVr(F+P_^_$ZA^RgaC zHhgY{SIq*(3I*Qc4+f&-$QNZ~clcpGO(yd(B8Tz7v6(H)X4t%ESM`N$60QD7$7M0- zLpwu!N@hS4YSb|>`14j9B}QxHpoRh%4R<@Oy%%eL!A?zm*yFj`Ombd+a{E&pEOtE8Sg;}OM>sN%heEEQ_h78_5smcNFKD-+oU*$$0Bwyo1!?U9 zpifJMT`+%0de*;P=XmlU`&Wyd<*+0Wj8X{?VOhI} zE5|?p!tkPgFLN(LWT(%>UGcx?vHDoZTp=^eb99UuOJF=>jZP0T&H+}ZlaXyM@MGjR zB=qZv#r-7l%)@*))o~kVEI&S?@|ugS>J*^VdrGHLw4u{!YR0AF1}qgR3N}+{;OQ9$ z<=(f~!uT!fub=)dqo))^&8Q<`ePUU`&`1N; zfNLmNP7vl7115~`V*C`-F=m6l7g}NN=u{t7n-t8goqib>2$Fi-R^oEwXD}sinB*x zsqyJT#^+-=+Gi;|P6Mql#;Ib^Y32hXSH%uh_E;c-Gh54+*iRf)Zp3KqDU27AO!)1$ zGndc744wYI;?a|+Y1CAANb@>sdLPSp`!E#@A73rAy}1OV$=*s=HI`8~@z|e*GOWIJ z=B3c=A43x~mf|R`%Xb`2q}NS;ZJS23Zyp@fq+&$`V&(XeAsy5o`pZp$-5$+FzYOd4 zC7_NsOMLO=FHn)`Nd7UyvoPtW9=XU>j-z;?NA6{gt*eznqb-wP7hbsI z$gcUH6ZbO3(Vg0B5xyU*gIaA4Fglhn1~cZj=DUuV{xj}2SMV_I6Q?~a3^UkYYXp?b z5WmrUrW;l#mYJ?Sy$*y3ez{uzYk&RG6N}j)B53Yz-H>fy0JNTc-OTPW4r>jBcS@NV zFhV``fJJB;Rp7}FM~eBuT*p|ccn2x21>>GG5hadT-Ta(CBhKB(9j&YMG=$3a;3)q1 zT%XMQiGG>xthiu^@pO(1iALVkZ^lz@iI~RDK7~9)HrehRhT6^^t$ccE7Um|b9d<1; z;3+9?Fsyy&142SCp8EYun6Mb|bfee*yN+nJV%2ic*d;N&j^Ej+j9t)^_-hg7@5uaE zU_J`#W6P}z=sX&h{E&G{;3ULk7+whCV}t@tHk^p+XISmo%Jq4h3G??qQD60lf#iU% ziZ>=PKHB^+hT@Y=?@9m9SY5{rNMpE&1(r;?xHF`951JiUd=T{_1qSBnct3h!_{OQG zuu-o6z!Pu05Mq7HmA`GgmK@et?53%q_4B6(Rliu!JlVd~Um;b{r=WlHZX*rMBzx~T zb=V$e)856Z_+j|yy1C|Jce^08X)oe3+;jL}^8rmL3(;)FVfEK@6S$I^6Bw?j{K0;w zLns-Dn>BxMk0()+ta*>*0e5U)GB4%iX2^#Lo_saF>zGcxZ_=f`^Nf%fmiHMU&$|L8n@ox^!4cC)@jFu>-#C>D?U~NtMVR64B zHqXTdU!@gdaZ)0M{_QY_GN?;V#SImZ>fQgY-$~Dn(!YMiX5)LZauFa@)Xk9DbD?gU z#8>azFgsx=^6pj7o1}inkyoaMJ#s8UZ-u3whrK@qO;l+1!(CYz?@)d8aS8K-+C)Ew zwLisp+(zOqWqMlfgt3nF!Z+lYzUJ0`f?&ZG#v8e`_js*Y%wAZC`mxhq5v$J`OR7ano#)SNi_?hj$=02h&OYh@MUeIGw{&MQNi- zH7p+VXEtPi;#;h~bm`TCVB+DkP!+DR%zu~$^$VT1yY)K&R?MF46iv%VBehcRA8Bsm zf5pyu=edX9d#}Ts^0BvbPMAJo?_A>fw=`HCVeaCA#W9sdG*qSYr9Q_ST0$SxWnj9w zcnbY7)PL(Lc+G8|H~0^}t|2q0$L&(EZN4_|@n|KytbrEyKUy~D*kE<9KpSXnVDo{D z=Q~g5s2Q=vyo0P!T3=4*AN$(EE zV1CJTi@Ou`6`1j45KjNZfFpA%aaH91pso`l zN19GfK{$2E(_=Lq5H=VbOOV3qyH;D2M>bboP}7m29U)D7VDfz9VczP?XejfN#Bu5k zSm3ddBRYm1hL%Iy zp0xpux~8zie>jZ77cYO(GfalwFW?ydEDT5K6*WS$EDz(EjGwC5Z^1$(E-GV92eOz; ziW%_9uoU(!PJJ4yryurwqw6pvUtW=JleWkCCmQLtTbD#c=qYlol zdf{Y&hVQAL&$hpgh6qDv0zXH>O1Wo)?K_O8u2+3{^@9+m3qJTsgigqVRA&KQuct4) zlOfg9quM;$fo2JSl_fVDZ-J}W`u6zWm2|Anyd_w}Uk?(A#mQr=);{hr^Ral&+;;=!V`^rD>uzhCAx& z(%!Elxd`1BPLh`XaxkQMpspzNGU~5jl2o|-WZV4TF!$~A8s@nG*gcd39MUU)l{MV?A4QhiPV!9{;x#Bkq#EwI=`{~0kok)BF zFsbmgjCb%ejyCg5zH)a6%!+4xYv?28_i;2rKkBqXSJC3SP?OCGMIgxa?$fHjxvhV; zyq0PIY&_W34m{pX6L*9i`XB75o_KQ=O-oU|Z@+=XgX-q?>?~BahCUy?3tLPbP;h^W z3f26=kf9dq4Fau@~})Yu_o){)G7h z({JLCD)jc35&3^<9Hb6=Ml&cWp!-++1B2Q;+Lyr zmjn>OKP1goAs-4H?Qhp8en8bPr;Q|H2T>#ORlL$JBFr-_?a3>xg)V(DzpHC1P@>!! zd_Z3u^6nhzJRY769kj&VnE@1Iy*VBR%G%Ib_0XzR$K;#QlT z(CQKO$c|`;GJEgwGZU_&jR10v+#|!_lE>sq_SHk}}tXqpNye@%_XH&fy{|1hI& z)`RXpdp^S4KKk`%JV~T}7*5yPv3nA$BWV|%+c_UW+SgG@&T0CHCpXYCFex-m&O(!1 zflLcGmZzK>&oKXb0gLagNZrZi$hv~Y(tVbV!hLZRBaN{SE|BUTapa+&IMtpIfxtSn zsn=74x-=rY=i1F-S;N@jA>u;wE^+hTMzd&bDOLGJ{PKUl*Z;47aZC~EQ!aN-;;8)2 zM6)!K);-KHbmUTqZkr#Z_a~~6osMq%CyteLpIn;<@{&iE&;&!V`Nm^;6xB(6Z9pE& zD>QLCEUd&z{rB(i6q<52&jq%vv;Xo3 zMay|x0sX~TUBYaI;`vn%UDVUFoLr*4>;G#zx3!~iU#+PfSq3UD)XQytx-D+7?RsKO z&Bv+WhW~$``#+zvF=oA#v%l{D^Y4f$0=gqq`(bcNtKX(83MM0o-(L5G{`Ct(Y&^{W zpdzJ9u@lY3FhQZGrNCW|nnyk*ez2=XO%viGhHEwd^Ztzs_ocejx8YRr6cgvXH{455 zU$4qoxkEBcvx_}fj41=Wx=$k6HzC-JJo@K*_%mo&{4v(?r~&3c^{SCxBn*7q*>Yj; zU08D+Ec~kR5)FozzN?TY#kX+eHK7C1cQD-Qfd0e4A5XCSahg)_u3H~Sc@8w>&)8(} zku*LyGEET*_js%xp-lMj2Ngj3OY>|Hu|C6D5vPwfjq0jRtD({eJw) zh{d++7xKGSKX+EZ>`}&k3p|Ef93=i==nt;gcKsHuOsCazPS{*}g?Bts2%GBCTQ(23 zo%66>;#@%Kfa!A0(a~D8e8lXTso$$*)Kb+jC{j&fcg`RDeVWBnn2sP~%|4F7J|9T4 zx*T9(xC&#h_l3voz@}EACKkSiz9o&W0RX;3`@W3 zZVj1kboVGk9_&`Lql2Xa zt_S&K_n`i(VCbW$L@1Gz306w2w8(qx`MpegYD(>eQ!hD8q z()8dI)YBFHIrFIz7LR}KE-t|Q&6yIXBx}2RJh@>5T@HgM)X@eUmVD6;L`Tam&5I7u zu^Gr6)PvbW&8e0r*Ur9$!Qd*voakmC?ta02(|IQh^LX;l6k&EPVK#cE?tuduR$b!I zey9tR`DY*TvB{!^pm7f~<0Z7AQOiR|z6>EzfiY)y7DI2J!#A$9QW%6okVG-6X|8X3#h;HO z(`9(URzo`faO5Lfy?X6fe3W=6Cc8{H9gU$f$@pb4(&t@G$wCs7&ZzTcRdqH=sO>O#w_MAjK#2qzVW*X-8C^`Pu{Jd8ZF+deUe--ulJx( z`r`_Wjoh?q>-~-TLT2ASRQiu~sOG83n`4bKc1_h0$Dg6zRpnJa{7Ft+W~}1&piCUJ3!?nKt{l zE&KoOcg*fv)AQ!MK8KC7)*Q>Vhorh19EA)u@)s@j zcKwgM*PpI3dUuQ;di3VAafPpA*!R2M7^fOE@>Ar{6FOTqNO)D%~V(oiajF&%V4^tXb z?nPHHy_mn}QQzsAwg1(Qf4>iLVJd@v{eNCJXvO8?lc%xm^NGXFoEaCl*K`%^CvfurpM6ZuAEZ(c!&r#9)uXdmzy>E^5ilW;-uYZ+& z#`L&85STxtnO0%=w04!PsRM2i;Qz_3R%*>2Is(}V*M+e7&}_)q<$HXxXy9P>l9Y}W zvdg-!)J$>zfAt5dRDWEXu!-^9w4$&dx6eHNzwLKCmF8gQ+aDMYBDrY%u88UV(5^ut zc+L0|G&#rwg*+tr!7x6hooA{yP5Qrm;0xgv)$q3Q7~zcJWs?5NRDp@0+6&dYjbNrKSVdX*AfB9x<+QxY8z7v_HQ1j$ z4OLRzeOE_Fe72}8f96_sdMM0I%daVW>%hQKDSS80C+M#n9ZpZr!;#BPNILSiQsfoIRd$fesdVG%jilAAys2Sx zBbT1AcpFb?;bpZJ@epc7eOEZ2)}qx@$MpDnHt=Mb%0GUd#PWV)oC)7A$VY5@-i+9s z{x3HS4-qN4|6OSBC$wjf5?O1C$jygR&LnczNdJ?>F8(&&LAPl5`d0yzYti`g$HX zpU*P)-$8GZbB#_PmdEckh+zEK5eZAP6t~#DSYYE0y-6>Dd)xaAFf1FG1wMuHl9*PT z()w-R&pxO;@q`-7A8aDqXWUKKQInW}@|YQE|AYK*b>6W*_;6Gi6x9QY$Do zEh_NvP0$R*>JAGyh1gv)lA*6L4`mBbK;po~VmP|{!ZO)MbwqJz$kcMi1a!AF1|=nnLHRY)(oX9OJ30bu zoQ41Ac{W6)y~2H*#KQw4YBKwLM0EPsP)u}ZK0>!w)t9UEIj4j^8)2fJ|^BJZsSGYc72ch z_eEyf(|EF{8Cz5WNzmyqt4dvc;s4+O$hPiApb^cUu5;%)VK_q~Zm&W*`f(}gcb&#h zL||0P^m$~p?S2bxEjAU;a6mk$cy}OjTRq=)JMHqLO}J_TTOns{A`4_-oj7vMz7o@O z>vCW4@rZwjBbOLSvUlu&`F96w?^i7T^L{HhnW)yA5-`WkoMw`V*`>=Tm5zR_+OD&> z8s1ihpNszI`1PnqQ6#yuLPO7_KI6b&|Du?&bJOTPj@)pSNBL_XEDAAA^Bp(1M2c%e=OusPWC*R#JB2wtgfCvfN33 zyR9z&e~$lDx45hQvoA2aC}I_^oPjb7Z`2Gh@xu(9Ke}vle*3<$<(p&rE)<9ksREPX zSUnq&wsv~S!~37(n3h;~k4bj`hVeVJ*xkNi_al$3Rux@Et@vfH>mM7SzcKgRqhoxi z_V;Sn;<=NsxnQCOb2H!?iCXPvZPpy>;f`GQ^- z@#MSaR*O0DdeYf2Hq|9a;YWu4=ocrN&lJP>{Se>8@)1adD^Ap! zZEM#A3)|J31JY3TT_eyTb6Y;?f9|8LJ)x{J@&u?pD0Gb}${TgjHI6ucmV~|$G6TMo zLNH~i;?4bYH^L8vGT3omf#K&}Rs3>TUZ(q)jPleO3|G)2b5<-r?>c4&-Jsd&co*|K zHblyHWmW7)6;$zY=GVqxA?FJd!|zFG)+oQYwu#|xdugvN?G8K+HEfBpbxEx-9h+q< zMZoH8`p#C_KbV<8E7z`YDCu#*M$XIch9w;^SlgE~?9qk_M-yh)&7@H1X0y(2T^&5N z@Y-HE{&AR6alIclbP&~k&muoz*93j8HP06Hl+c)D=2z}XtiPK)y4&6AuD~qIzN%35 z&oHp>!S|szSbgI#k8R2llNi+JoHcr?{}oLI#y!{Rt%hlwXV{AG9jG>G^WPI)hQ`P< zgO>#LU`(de-Gkf;M=QzNv$+|I5(!}(VuvvNQ`>yd-pB*9r2iXYPLAo^8b`x!>&E71 z3SoBlo{Qm6@8QYBhinzouz2TkmydYTffOJt6b|>Rk>Y4*iT}=QV3+Ja_lLR;+hX7Q zSpD>3lzv|3IZYtm;i#WIp9IUV8lAe&Z~gQ4eSx<&2Sw>&DScTWNVF6UkPDYw;nzT8 z71!9eD6Mc5FU@$wSnFYE%{Npgfg3eOfa*Hk0TjeBIHiu_mjdBo4A4P+MB5d7jW~k4FqBx$-7@9fAc?a)c`e3sk=A%V&17;)Bkng8p3t4`;}klq0aG;h;?B5KeBHkHo% zojJ3pbV&mGG}=?X>+gW_p35&@$UbfHhR-73L~6pw|rH@0?% z(t+ zjcZ(e*gyRX(tW=tkAEl+An{6}+EO0b;yhOult7F>GiN?9gRBL_!B;I~Edfz(^Ph<8ZfIWr|jn?-*{1Wbh_501wCmU|0 zzyK3oOMdVOjCbywaGT$O2EHFWTR3zWTIkMQRGRtncb;4jyl=bRU!M_o3~ImH1p{t! zP3_ZP@nri#KYvs$K~-8|D{XXnQ0aI|?BeU=P%){p?0qs5r6g}0_3^>#etR=VxPJ_V zLt1X!n9L?Go~+30d5@O^HorHB!ZddxV7Bw^hO9pqbgh2+(>x!9`pi}Ne15_ku=q;nZRZ8e9>(8W~0M7g2|ls)%_UhO6zbui2H!xNScoeP7iTc-d1r zA7&?6PDzm^{j*(REid(d{QQkJuF%y$7*^ku@1DJzK#lRHE+$tB>#i@s_!Dw2n&%{( zC(M5~)F9`3hdRklP$~xt!?4K4`$^lQFmJO@*Er=kW*7Z=^Q2a)7tJn>m^Oz0`d52! z6#LZ4UY3${8WG{(WR-ii8Po|xR(vrY!Bbi3eZC~90J9gkcv&UtU}hw!sj~y)E6P5a z6sr2j8BZpA{9)}WQhfs&V&KYsXVi(Kc(!hk?U;VFf#h$a*TNPDqizocgc`=^n|65|1 z5-q%(ZI7OOgcetXRfK&>aWLo+GVyFM!~Cb=lMk48-NE*wj*pm*ua`#& zknD5F|9?pP?|7`@|9>2}cS}M_MpiOPB$CHflB`HXsm#)#l#oiPWJD@0Ng*qgy~Fde z_uhN&bq#xd`~2~{&g=HRKHXmL&+Ye@n{%D>Jm);mb3DfValc>JD7S6l_!>oV$uqo+O#~xUBas6gjto_$6JQAsKRRQJ-<3D}Ap``oh!SUm!wmpOce=K>6n?*%HEk1AB zXvorYi50cEf=V=siQ1;$(T|~EagwqQ8sSY95$Umqv06{#;HZ9>#Otpz#h%5+`GWz6 zU=xZS53DRT?w^{8gwZWMG!aqCsLA~7ld*(TRPQks^Sq_va6Pm#-Fly+h2cwA-s0bK zbv#22eTgi0l&8^}^Y>-e$hWW(tNn={1hD>Fcc|i6nE2n1`@;8?ZCCmJ^*fLyZNHKo zieb&_dC4PMJd8BctemKi1G0O~^2g*^>Ty@x7dvH4fGHbBw?Vmsf7dxa)`8vbsCT@< z_Z$s|qaW&7^d_LMs4vQ~x@zVYELZZ`x?lW)#%1OtXl*e*nwggqA8lV@bql3Ca&|m_ z_!inXyj)u@-ptnPT`CLDv~mK!$~bx?;|v{#iC#(NYL?v^_) zT>@RDYpvRXSiJE)8PDaY-4AolqchpFlsW?(ZS!Z-YSC+${gCDJkX-y6W?BTj2~z*n zF2T_b)U|2x`Xh3wxSYGS8jfl5`gP5?dbBzp^-OI22~0J|)>`2np*4M*1DQ|K(X2i{ z@T7agq@Fr{pBWiux1E=kPp5@EAIas%KC?h#mNR{SkIJ9I(|_o5`LQhowMU4x)OeSp z;q76HJ7!eybocMqs=Gb{)z}y0sZphx@ z-hB_vxH#so=_f*c<1^X7a4#UwG53+Sx599OyGMl5FrLQ8Iz1|bf^&pkNrxrv&1tB6 z$@4UQw z??-zy`6PE{yvh*9f5y-l^RvO=cFw7~q*|y7Hs5naa0Jb*Uzk|l4M1d0kYCtVh)6<= z&G(|05xKXnk%r9>2v6gume>c-+(}UbgV`-;GSYw7IQ?N*%~X!O%)yNaB^Msb@6(2% zGwu1s+O3h$CFje}Lmo=JQ`t9-K$~i=Myfag# z=Rf?4AMSC&V_%`K#eqXKZw)O3q#jXo!|qQa^Cc5rVR}-_rkldIH@O3;*@e+>Mh^OU zj|4MpIfS}cdOj~`i=+CH2aQI9g-|t~bVKGf=ATX`_k55Q;y`5I8MSC-)xZAoH-(TS z%%APmStp-N!Qwz|p|3^yoYbLWQCqny@D{9EPHRMmQR1|)RI^>d%B%~`k?6DST&Q^e zf$VqiyhId*KNU@+TWU63qS99elJ$^wNi@X{hUU5U8FQn*{td6V-T;j)8a^z>*&6)? z5pMKfyXO5ECbEZJ*mE#`FXAr$u^SpxxIHw$9wBtbk7CC|hvr4+!tG}fi6lpBbP~hw z^qa)(wIj>`sh@p6&$9!lpj>Em@iR59jbIa9vm+JLxm!8L&86+R6#od#Ne3M8N&4G=pYsHn|IDEU zpXW=7+Lf@hoAjMk0n;byKE;2?lMBN;^}Uju>+{?THCx}sv~9T#OEYm>e)C{;c{O^v z2D*oSLch=FYnK(Vbre7P=cfM6gQ$)!haG1$j2d)lyJ_!YbuAT#w})wv{9#)EeYfs~ z0JJ=QWJXfuGUVcMdw-gcP*YJ#^3*kq2XvJ6cB@)O1nNKM9&8jr$)Dj^_itHw@TMCk z>qA^T<|zDmXlTm2Sp4fJR1Y_2x`yJ=+!>=w1+-hy@8i3(ZPZ<$@nzVfy(@p`n`6I- zuN_r)^=jZIR*$|`Dse0#;wLoWU2crruJXkVe+~9&$7={ z_Zzf7taIOys@@is;+Kwl`C#~k?!=rAURsz=&DyKJ?LKdtp;><0t6L6KaSN2g#biul zr$N>3==--H?R)!Cd4vkicJ>X*v@3lbZovlYgHx>KyNjvLuQOF2H)O!#7}J~es%kz? zqvrc391ayV<7nQq==fyRXWMnh{YDDTY>B<{64lKa2)H%tlFax=)gqfAymgWV+$ z@N9YxYTmmO(Ii(9*Y+2mVOC2~mg{Q`DwsRJJBgMaN4wu|`9^2YKieDPXpi?=r^+{> z&LhJi{`Q(w$EUX5yq1wAibe#^`*BB*pyC?OwSwV&(0F8^Xj?PIFTl}d%)j1yo(H9l zu5^nUMN{9mzAgB~b1L06%J1{Rwo4}zBv3h1{DAbHf8u>msr-r@;e8~YrldIe%RzNC zBL4fE@xLs7?NBrXS@bcO=An#~um6ef! zI;ZnzJI>g{dc;Ax)lC_Y$MoAiFq99vV|sr&4`8|v#H#(ui950U(fV-Lx2xQg{3f)A zac9~w{qoY511CstxOJ-FHP7C!uk5ZRL^vG;#{e$5&aF%mYR;f;H<-t;Uu zniaz?k^NYnbM8Yr?J<@!K>V)!HqvRWt!?j@wei!}VEbse3ABf%`qvv(~XpENTB&UO~Q0LE!4LRAm zzmKENC8rW&3pcbfGOxp7C4mOdex)%Cpu885IM44%EsH>ZW*b!ENLc=DA?xFBR`+!n zEe!ZBm=K5-f>xy0E}ekcR|^g3Tnl=?vIo3VX3zqg0gHuTGMYL&8*}NHIF3m+G(4M8 z0LO68nSDWe9d&FtR94!>q3*OVn#Ij+f5&al{?G0*(D(N7o%vZkCU4_}$G8xTuW!N~FH8ypP%D==h-{uK9A5Y~rXRQ{qoG@aeOB=2w!$ZkiIBN;H< z_PoL8@W_N*m`-dNGubMG2J6V@V;VUi^&z1oTFf8v4Q2gRt6syH%&e{_Uo+&hOByZC>$;x)lHQLrL;rupm%xe z#7{ia{yE;GPcZ%UmZ}FCT0K;}?v!?0T2HrUxMKCfL+lbm%p#Sr@@cEjpPR>^e}R6u zX9od#2lLFC$1t7l@iM`*A1T!Pde85-6E#qXy`SQADW{YNqzJU1E)Wlau8l*NUWa1* z=JUrC>PpNL(1@UQ)UA>=JZtl<1X9a$%I~#Wh7R=oZ&69qz-bj1R_JB_Nv0offqI9n zEFz~b!XnP!7$?$=o+l*H%o!fVQ)<^V(7Ddoehu*I9o? zv)O5*n~zZZP?+;$HI!@ggZk?UZtttBVP=bXvc`e^kju`qsX~WZCzhZh+30mZ7itgR z$$3%!A2`#r6i(GeEU&iM>2s9V3$w3@pC7I5{oRa``?Jck2j~9zeP*2oUfN#!Cr+61 z9Al?Wmvr8ss+T~^>Nh9Tt+D-eK>tk)-oOm?*>7Avae^21g-pf|%VPZZNjri)7nM%{ zu`@KlTAzZGgQAoByFoVyPtdiPU)pV$=rA%z30$6Zg{qqT+23g)ym6I=HP zz~ZK(gdMwX!|c*&j!98R9P{LM=SYDqaCPy}XL$}hYMp;=Apa#E=5=&GR9q!d-GA;R z*Gc|c81ALFzL$Mv57&Pi4=|9sU3J@9dFz1Wx(e!Ql ze#Tr|G=ih)^qts6b=}&LtuIuix==$}nEvLjXVB{RB{~y_@jnhP?2Db9$Nc55dwFgo z1{?vx4TY77r3cU%s~x@BPzue{RoaH^sQ;hsH+qXj=aMMUn3c zBE|)cXvp?Lx!Hh>wC;b@Mbg}4)?^3_Q$eojpfg81G@_upHk$q{ik(Y7q`?$E4~Lf3A(ozQB|@l?C>pZHq%&USxGQTYk?DG zKZN1mehfq(0o0s{vaS9@r)YLtP({NVdNiKzI4~nl!Sh3YG0lvU$_%P=<&@93bQcCx zrM5LEwPE9_rL3sa1k1~l7Q|k*W>~{gnf7+hV@FY&YMNqB@;`YAfaK+V@9YU8G(A~S zaO@*Eejd2e5-m{Ij#_`d7gYq@9`hH;wKv7-s*Tjf2Pph@sn?$ zQhrPGf=3!^WmQ+z+(YSq=wsS)(RtMvM>lJ!dU;bEO1kqcpdT1e;@ftym+E3rJZx8T z+@&jR)NkuT4-W8d-yXIP6x=fH`cD3-a%tK4!g@6Q}!_!V4E)I4pc-# z@5Tc0P5zYQGs7LN*T=b`Ej!PB>+BKKe4(WsWUzYVQS*x;87F0M^g6N52XFj_(Q{{) zvP?Ws?fD~)9EV?{<-=dS&Y9Z)nK@dnJz5A&(d^PpW#q*(9h%-L*mwd+YaZ+u{U|sw zv~Y+|9t|Z{Myo81<{aDt>zt%rSiS?7X=XEf(`dYp_|Po=xfFv0p2|$MyI8 zT*D}ep3Z77)~4ifa4c^=96z@C6pltiNuwV%qpJFqE1MUEVDQZ+(O*vH5K_~C7rx5~ zog769NpAmP|4OaD2}!{0@nI>ZvYkbZXn=6Pkht3y^=_};+`~xWC&SUw8E6_$XTeQN z73MG;7n-*o(`|X22NS_JO!D+G9vXt<1P@;YwysPwMyWkcB>%U02jis;eG48JQRbEt z$!Dw2L(6@=Nw3@YQ67KMa%}ov|F5yp{iW3wSax}3w9f7Yk)HQn+6A9Rqj`y;2F4#@ zaw+H8nr;&e*#(AHr{`kt$-Zw}$Z!uDetMC)4CN{qZ4JW2P`NHN1yksIfFiQ5hF#cW z)Mu$V#a4sGnFx>1vHj#s19Ik4U+#7+E;7BnscBvc%l9sPV8Q3^nuLu{pV(@Bj>1ew z=HAAVFW{-g^yq-jE7W1-7NtH;;on6g*R5pqH&ex}fe@XJGt0y9`KH9Jr=r9IVBYxI zJJHKi@bgoWs7gG>FW=rT<^7mUsei(;jpi%+U#Fw~{@Vt_6+){ZLb(GM!cYWjn}gLR zjVS98p5~4;y)rF3T0P<>f8o$})En^gF^}gqsAZG6zK7Nc4MfNo+3olPE8;k{hcU-t zEK+bsfhT4!^gb%F`D}xYi>_yd3*5oqq2$}^KSSid_QeqH``dDVs^=Q@x^Uyn@;~uD zROb&Y(Vn@WAOzzbpEJEeNoe%!&&{FR{@42t%pYQp;fqy--mdZkKh$$z;6r9mxspF> zj{In6xk%+tVfKo@;koEjlz1_YwyG>D+mQ;-^LHJ3ctG(#beDR!NgR5)NWtx+j@AR3 z4tho?#D?+ZCUqQaTYrV7P zJIeQ_eYhMc8g>9jdo$%%sgNKVxcF45SM)im6{{k#yi}!nuQ?`R9i@lHuyE_eV0m>9 zkc@5IEDJng?HbAI6SDxSZryb7qGsCvybecuJ>K%>g-s~*LB{vm&;Nlld%2dEWuJ&z zo}t*MIOY~bvwbTXFjoJGW%Twm>Ql)mA1H{1{>LwBQ{MguzT#AsjqvVguyMKA zF?(qrsyJjAXT;iqdVcmu=ZIv%Qq*;&Il)?JmA=PKp4CFb+Y@8IDgIa9Ef^B9xz!zR z2t!$Nbak)!@T^BS*_0@=pxQZhQn2_=91Agab6=tfj;{N~<#Y@Oj{5$nNcj)HMWlhj zOPxB_RPB=aoLa>KF~6zEO40VQ3dXOR8>+LQC2R$QXU^aGdD8|e{NFikW$MS#dv-kve+n98-rL*zO%OFq->NCPN73WJ(O1SiRr520 zfjl9{jR;rhJ>n{9bSE0sNMA0?v9dz5)z>3SFJXBkLf3rfv)&>Yatielkext{R~&lW z&IaOWo>X25h`9&#JFQQLmO23WywO$#Gb0#p`u4c&B0Z}8KKS7GeX4Z?)&i4taWW66 zj+4(NshJENhv{p=6RqF>j;AMO(Up-bux37Xg|?;xb@N&5Jv@r>qYq3Mhd3P%K(o)0 z%^ilDFmgK0S|>aTX2-H6LdLQ9|FBf98)BM(l>=`g_H!&Li|{-e_*8=C&SCk zbLI`AF!^O$sqoWqC{nSz^-?fA2Cq!HfY8^M=bXa7oKH$X3|uR%D=_aJWCdKJW27NfS|>2mmViV z`M+AAg{>xnD*pz{Ernbj3K;&gD>Pfddx8VzoZjT;^-%DEP#ND)p+X9Rw$mBL`?wwKHp(F!^`Ubu1|!}1LaHGQ7hDUwvrvFPVzc6+NB z)p0^lYv1_(PblHU6&8(*?J&57Ug{-#In0jKPmH$Q$1|TN@>f?NZ2tc;9SfjQM}wmH z{VrVGIEIV}c?LHO*Vg1pzxY{o7^m9Le#lB{465j;$MVVYRVr;@Q>`F3o7M%d%%&t+w zPb|N=7Tgzdh}9H|BZku+E3`t&71NIvbyU1^RO55%5=~Ss8N=!f?DH(dc%@)U4KW1`Pumv(YH<>?i{(_?kk8`+aSPpB% z&CM!(*#1)It#G7f2z$>tc|qM<#6f7N{19j#A_sH!zf{_Wyinoe=kbmQHX+ZQ&sVdI zFES>R~4SwB8`lTwEYlLy0nmJ7L|Gt6O3VGl11NWQXcp!faj zzrIf}Dx~5`f+AiEHG#z#L>3^tQ=Zj^$)_og!lN;L_LdXv>UXu7P-|pz-b@Ci-Vsf_ zJUS$;+l1L=*T^?(vW;PuRq08oq8(c1+B8(4o`$0@Yr5~9i1Cm7e13ViND`(CSlT1} zt~as?ja!Z7vfti}XGv%|^F@^kSA%1)CT`ywg4O+yjh?29*qA||%GR*)4pktW(;(dc z;)kb`=nUd>DT4{Y9mVZgcBoNlZK(bF^nd&AK)zpQ92ND~?my=lpID9Q^7UT1^I^t* z5XRbj!d`ouq3B2Nc4Q@lK(9=qZBzsejGMo>llC8aB1r?+wgv6Qc*F;H+4yOHrQ*AS zsYd~gL1~GIObAt+k2no$xzCIYifv&1&;_x{;z=0ZdnOWR^%j=+5{5~hSiMZ`y#?V* zX|JI9-1=tNjYAJ5C*Io~67u zT6t@GjP|7@j=_~%neXsF^JP^1+`)6iVMj3~j)2N7-!@)%kcCMr3Fh1L=U}niZgfx3 zE2`_3*1d0B6`DZHQU)xm9JL(bQ%_SYQCg=S=`wkh~u5TO9CEfl-WQ*n!N@x>Ga9{ z^qZjZ^8RXk4)%Sz@dwY)u#N-CA=)f*BNdkZEbj|cDu#AJA}NeX7iM_z>lWPO~A{K9*EXQRX&p#N){x@0rPA3VUH5%J{*h2I883ui60TAx7g_Jo1b zaHs$JoqYKl6R+HAMiGvBWx@LlP~QOo>rcTH9Xp5sT|lD0%h+hV!1ECLOYq!>qP$VL ze%&`|{B9U7u(O?RtwtmMvu35jBdBlBliFRzZqV0N$Zu$73;hf|2e|?;J(`grp~pM< zI1wT|TD+Qxjn~m%Ogg3w?5MOWJ?PAlc(nAW{%!2k4OE;&YoH^RfqDYZIJ(PiM(Zsj z60$KdF#2FL>h_%vFgEmNeRp{UEJ};{4@kJfqFNR^``8~mtG=Vntj9j6w7tegT>T2Y z;>Ia2M9P2+s!~FtlywtLi0=|p2|S2rvdt9W$iY#6Eo8)(%CFStsSp@lqWK2P{|&_(i+uP=#d8M3gzOcL z`8)rw&j+)=c8Cp}DuUrN!mn<)Q`HrtdYvOQlN$Xn_QT!hUI7JHhl*qm#QivM7$#IL zLzs)JsIJS7u+`?Bc0+YlDi0Q4Z{Kevoa#m8S5VGxl&~|7 z4t@{)2ZN-x*7~FI&e5-tXC9&+hi2`SU>1n)vgF>QpoAJ<3exO283T2eFWYcO+)*>% zHCeO6N6}DY(e)9fDoB50cROIG0@Ti|`fs#h@yB(J?jzUG3WS=Os2%Iuj-%!FiY%bP zfwyRowSB5T^hlVUIm5yPr92}F;UU6MT`gg!5K@TB)^Tlj(Su=KiC%_XQI zIBC)!$U*I?NBA38?n1`Ha+Q?YLDWz-qqgr76B_?+_rdLL2O7{Z_4q`uh6aMSg_Y@F zfbq;VW1b<5uYHm!QS;mC8lFi^+oknOA{t{kJ5KkW6HR%@lv*$xg=KTjr$QT<(9(Cn zxlWc7jbG*LqPbd$YIRa@X6Gb;1WslaM{c00rNRrltd79~Xjbg{m4VtD!ky;t835VL z@0h#eOZYmpn>&vXf_hqe-i?3ug|?-um*Wl!!W7L!3Yos4GvOA0xKU}-G`RV~Qm!c-vYK5IYYWGpGZnGeo-IDC zLwkoq*w5ZWlN~FQstcH2SpK*vjb<{I@2EKQeNO+~9T;}0j^jM&3zHsdC$^sb+fLtJ z9*YzUM81}+=~=gcLP~U>rwg7yq#rewszX+&UJuV9C!>a^y{2A4^T8Homk-P;qgOD` zMnB_V`5pQiok|rOu)5fszY|2gXbw8}MTnA&mZ8U$m!JJ4R>y!ZyXg|)e+#u|Ixo$x zIHKOb1EVh|!coTIPOS#~9UvW1wEfa2g{O`65AJv^1rvu2g*iDfTu{UJ{3j2$4&i7* zyU{B>5~_I^zN69072!8hn8i;}<6`F;(~M6=)1x6X*|1^B4Al6!g!fd?HdJ>bbG|yAPgy!X=sS_JG0aY4ZJETPU);dn#4l5XA?V z2tKbnjIx$$%*~7YP;aD{j>b9;bzY)nVSGA)Ci>EkuI=uE>aN2Ao)Qv}D*mGKRL(Xu z<3xaW65KG7P#r&c9Mg@QUVpnq?DIt+21v9#YZZdNAIvHQQBPFCe-Yn&k{P-#jeL(k zl#VJDAJ&$C$M~3LAN^SO5m-QJude1fw*E7&1_bLZvon3Q&~dBpbAA|>pPp@MYP}d1 z3X2wMUt?i2o*4eYn}H0^ZHG4T}`ffh2f`K z4#+eZnSTNz%pVrX=Z6+m-$8B*G1M8rP*?kz4*F(k7>Z6-p}Nycu}Lqmx{RR{n#|?- zfJmAaLQ69wNOzfiu&kRMqf|SBB*WwJACnc-rzp z9<%%}P+(NqCs`qM_yw<+WhXr zB93A|pe7fMH}yn&7%{9lqNVf;&F%Y9d#sm&-+{iqC$CG1^+1Ybp>N~L|NkAwF=!u1 zeOg|CYL`+5FDjb?k%OMS;R6rMb+2AC+wX>>um1ifiud1o9_0bu?I+c#^P2=z`I7S{ zg$Ei9T)N>u;z+SW(5Qg^GdeozI^0>+l?$iVFkX|=@GU89m&fspe!ey*g|PTh$Ca?V z2Q;bbSN{5abrHI%YgEtE=P`KTUWgAgy?7%0>)?6R*JtK)evbs6250?ZGcSe*>Ty*K z7Ap3Fe0-(hmXG?V-0rf`sc93`$1u7XZmpo&Fy3clu92uuX7*y=!U!yW%Vf-GvVoG= z2Vp#(#i+c@>`mODI7&EoDu>vx1(kH&-2Utt7wU_+sCZ2#9mYj}ADVd_0y&EeJNu+r zQP0TiSoFFIyuBN}q~-PndIZ@t(njl{Zmg)c%GwuA7w>IbTz`*6CYPj4f_7j;$6!maRT=Z>oC$)|W4`Afq_dqhyc`7X@^ww{Po!A#a)a)9=< zYc?%Wz2IH<+@azWC9Z;+MLC>mB!^-BlCg`bNG)nspM1Ulg$|zfcyhe-v!9SE&7sYv z&4$L)AO8+LV+JXK_{dgXDMUUT_EU~d7teI<73&-PAgqN=;~p`eMI$+S_X}dNxbC>k zl+~qD9u#cQaoyrw5Y_jpN4B3GJ28Hx;vFJ?XzKN$sD-K9)WZ-BwjWkr@6tt~4@z{u zTi-_OGkrk~GCXK@AUDIUYX{0bYBBkzR2P<$@@1WmDnjGRL;r7<+fgHT>30RDC#b&n z=n5fC2Ie21XK7Eqj}~Z;3>|;Rh<*rsvSb?LL=*N0d3KfBK)r#Z?$ZO#Kp0Uf-Fj{w z`3ucI@jtpBk=OrNI*I9{9-BN%gOi1*d*yhwz>Fy(?iOD)fA9ikZx372UGhQ$vhg)> z@6Q79fhf<+H5TZ2r6zv(MlB+jR4FpDJ%-WNNICW-dpxVy?`d0W63qR%bgtz~Jj}3P zu9Hdo17n}J*WYl|gBHf*!wyn-=sorII-!>q)i*Ll7km^0k|^J%v9L|3ctKpc?%6DA zpAc5(-d7E^k$g;snub6SSG)P<1`S$#erl&%A`PtM6du_U9EBQ+ndJ@-N~3cB$%6jK zVd&6&rL;r{0ix#&ZldZQq$gXLR7uaF0pF*?*{g*>j%BP3j^RYZ--mzuM&CiivsJea zoK!?(M|KtYDh#0UA8w^@0)?S(T(^ShRuR;voPVAB7>m>7?@zDKsD6YRH5#+!$hMH- zCZIlSmxhoE_t`?{MLdmiYx)+SxX>S60>`1+rT6ro}FcQPwUt*CE8RrwhWmHq^pDXnv7<10b~ zEL=wB`X^B1tiT~B+0&>cqB?IW>oA(_G7;Bz!}i7O&h$r72W6#c5$Fal5T9C^X>(EQEEjeC&OXw~2K{SkqY{rW z9ca>m@6Eey$Dw}c>QV1CJgfz{i1@i?!_aXnMb(j$Fjsf7xclh|R9M?zH!t`L7KJ6$ zPj)c40jI#d>`ZyR`Ypi(O-43^4DPQFk-2YEh6Ds|rWWAoj0 z;!I1WDUdV}?Y&Xz^&q52=W$-{+~XeBO$DpqN&8$cS zR)?7Mxmiuy8q34Y4I9*~Extgb?T2Ym(!K+7e)OhNtKC*<4;B#0hFn^VnX${$SY|GO*A!tIysaA1T z2=)BR`jF4)0sSsgS3P`Aq2``Q8-*NU)cP?!T=FSZJulRRvR)IB6M%urtFMn+{6eYb z4yKwhvxqz*&!A+v_wW3A^Zn~#CNzY8-m{SX+y2BGX~EH3c=|i~dsmmO(W1-#<6dUR z@T`_?vz7c5I|D8LP_8o#_C~!!&1tz+kMT5~>N_NPeeo=|ct@w|5774P`t_KgaOf8o zJ1Hvo3TCe8pUiNRz|-x!BJ3|l`uDh9-k-D4bOg<42%e}~#&pfypgP@H9M{LsNKn7-9`NCH>GgMEZ-ti>YrVGl@ zUlVyxwJ!!WA4;p@Eq4HXus+|BUH_MQ@|HmE<6;<3Wlz2M>2!|*M5<-xH)O7Y%*Q+h z)0$X6Ev`7%xit4-c5Iyc&!k^eI2RO&zwh@d2+RKs-Sp#>Q{slcQ!>-+A(;O;v+r<{ z^q2`OSqX<+dT9$|qDHgiFBCjG3P0o{vn=V2qfY=+5X zLmb1VJ-4bAFuhpP+h>Pg@f*W}r-n0!`uC^^jS1^6$_NPuqmKBKbvu|Wu(l*Rb-}m1;$`*{`F$FVz8|QrLbruN*d{<38IZ*uunHkrz#09zl!P*4IZxE-Go2-^=No^ z-;pHA3phG@ns;TbA<&Wg=(Iy1hVNUtxKF*xmy%aR(L5@%pr?RPs&aop;RY zhunQ=g+fAJqPIg05iwe2b>ulY>LFTzL|ueExr zG0I(uHFbPM$#cWf=Ka$ovljHh-u}{Mb9FS7Zm2X!u0*p3w3s)=yoVY7!Kk%+^w7vN zFTwi=LF4YLI)_!T{BPcd*+(94Zs^x(KKe*D9O}Ba^Q-STgBrVcMjtd#!ur*fLn~v0 z!f%IXXl=^gQ~r0~`!*k&LDECb;s@tr=%&zk=*{zYCKaIdSi0-kXrdm{!8zRzjWPI5jfQB5!t_LW{LyF5z??u`az}NG7-rimb!%?c! zW5!2;V83O7>%#?@xtAWcOMVX;IIMB5`tdZL&f({EO9v%1UcB9Tm)|^^jp=c(zi=Ab zw&J{2G-gnRmEDelFi{xs8|&yghUvEVb3NL`#Ay$6ZuivnZw|umpH`g#M+NZAc3Ml} zN1ajXOM-uS5mns_8g9&>+2Q{U4Mc6*=#6rKxQ@fu?s8J%w9s*9bHp#Z?PyXpT#fnV z9&8++zb6#v9f>+Jc6vDe&VkC$w#R}z_TiaSIjlK6_o2~kR+nC>q@w&R);gg*)OnbR zP4Bgta2UVZVAD>U4S5G3wsEQ_26w{(zJ{pgTZ4$p!rUKMo%g; zV?OLl#n8>Osg|?a6D?03dwgI9K}yrdGcBFB(ac#En$`#1l;^Eo8mP1^*bR0ug%@*v zg`jfDgbTL$GO+F=70x*}0=dT(iwY|`fLuQ6TU~!AYjkcOorj(bFIQ}O6Pn_YaQ>(fe99QLzH_-G>#kU-RQ zQXmlc#~&zY=inJ1DVwCNUxlK#dq=q9T2Ox3dMWE5KP=-6+QeOMK+U-=l}durXz;Ar z`m_DpQQy(}(W$~3sGMI zj4eIK)3h$d>fCfeM9QR6cEqFhQI@WwtpAI`15jOd8<%Z}-Eq z=EZ)W+&YM75em{SxKM(|TD1g~>Mo-xT_Owf)=E6HcE*tROdv#>nA@R~$uO|q`u%M; zrGL?4z5?;)>t}!0xpSw@eO6KP!IL`KpYFl%CpG)b;USdJ8dE@{PsI~}+11vp`xNb| z>O22_hw#Z;UjO93r}R(r%kig&a#jAfg@JY00Lio zA=`NrJ1;Oq{*o(I-3qFjRb1obSWWAvD5JV4?47%U_R17KLZ4 zn(ff?j*QDvotN-z#=B{CGpO`+F+0k|G4JeQj31^=@ne9{3oJj|CnFze9**fGbnglK zva+=StZlD_>qK~^x3o*cm^8DrAM;=#G9bLdZkxtw`dXJzK2=K>3=>KBL8(U zCO*AGjj?jqKQWv_Nu7jf4Fl5Wn2HY<*cnm zE6$55jO-L0Hbme_`}XIV0U}uDM93|?hIy`}E)UTmm^;n=a_%D~&WMKI>D&EQqS8-> zWx*Hxd^d{z&Ns~)a`F42S7xpMYylrEtMc1$R#MgNK(Xq!ja@3f(2`N1a9?l(#@2eD z86Cv%VQt6zlcsO)K;tm(Ykw zZ-S-XCo|>c6n;g>&)sLjs5uTZUoAQp)ZG922|3KA$Ns5Xj@lE=RE`~$M+@{j?%dPf z0~119`ksoL07*9BdX@AuSR&>ATAywH*Yz-3@}a{U_s{(U+O6fclt#9qI*D#A;TE=k zokw{d;>f2j19f{KaRx7TY^55F?%ng}mJk&m0yd5wD)qj-?+jz6xzWd+AGv!h{2 z1W(6#iV0uY0{w~}-`+;_!v@s(oEn%#J)`{*-1jq}knxv}{5uvzwvp$^!q=gew4g(a z)4QSKiE!Ep+C!*ti-)F{mlx{iE4Rqol?bGpob*4UUm~Jw*nalPb+o`a7BwjI9W63$ ze=n+RjB5Qhxma4a|JXrg=H_0F6|JR@VAOG;gPX%`@}vm3q3I3-?yR$J~f z`|XNG!)X(Z)<@o>@ar5Ksj)6-@I!1(vso%EGH!Yo+!O@_wfj@`Z!x={AT-5nGc^lM z+~GqB4`%T+jdsdCeoa*0YiNH_)zaJ*$nsqByVUs*Chr-gtMmSW29pANRfaJ{n7E$4 z>+Kxsl~>;RLH!jB?;D|2E5dki3MDm0Rm6|NWZwt`rFNtGiuKgmkvM1>t0#XOGoYN; z`|K!fL^@@h!-Netl6*#~vq4GBy9!GfGSF;YbVosoIIJn8OE~S@482PI6~Z^8U|?%5 zUG>p;=$1&`bg;4%ku(at{kU9EDEE&6=WX|>#!0<-p~48CDN1?$<^$hD8APytbw$u9 z43*u?k*LvegW}wfa956ODD+kM`X>A-p3y*4s3O<~eyZ4>{T$MX+CbYZZ`T{>4Ksh+ zaGMrSGrZ|u7#@4?5r6xsZ!>q%QekmVwc`-#yH5C|RLy~>8N7UZ@Z&TZ`Bbqxi20-TL^3ScWz>;RY&K zx4`JJ;?lBRWzhd5P?GRR0mjC6b6h>BgZlFp(ogSn!_zA;C+)ps2E?YzKTHJm&|q}* z@zbT{sO}~!?;|%yv`i}(!K!5j4c*1Bf8W`Raz4ts-ug6zMs%9PuPRqV?S=E-)z5zd z!iQHn2^LiRc9iebZ??s2e=G&Zi{|c*9X z0vhZtJh_kjQ;CKj-KESS%#rJ3GwK7s{TX$ci5xUNf;hKBj&Xo+K=f} zHfGH~zq5d-u1ma^G%<5)9(8c`-Z<}x=`@ZpGMOZM)I!pL>F5`(R5XusY6;jV0n*oy zJD>0FM8v@3fy^5e94WL8L_Zt8b^!&Zw9;1d-GPFPt-n9rCqgdCH}#X?JxKW4^uEw% z1U0LC@hP)A38RswkCryeLMQK_k5z*XDr?NxQ@tq~)j1u-EhG=16xH&yrk%8i*zb1Q zuca3Hw%RX>-a88)>Xyz|D9ga)N7>BqXc=h!&_NnI)COZVyL`(yjsfW!F)A|lCbV48 z)qHnc7JA+Io->$KprJW6JLf?MM405S{BbZGwVN+{7yh~jB#kdQ*WQYt)$@vGCq&Co za{tzuvv-70V{lSf&5{VJ!qKyZTos4eYMT90xipaYwC6(KsR_s%AsC7H8KP2o!MlAQ zqVaU}zU7B7UJR))mx6oRo|u0(EFgmp)=r_UQaplP?V zrsv;hS)Z&M=e=f(`kNXLOy7A7quVZW@7aF@=8aCh|E9Qv1}sW@Ot>Y0aQM+>T+mw} zs4h8rwdX*$ZTj|KLMqU~(yITWEDI*YRhgg1RKSv{VUE^g4EH+yQmkCdg%Oqe9-6E5 zq3qj0_*f=YQPP2<(M=8OFUIPMBrLT}eGp6hC$A}5yBIXP1LMnrAo4agRCG z_nj5jr@d)Kspmvp7dn~^Q>k@UM@jsLCJ}{?2hYT*kkBnl05Y4mg}jLrBJ=7MJ?K<~ z^-uexGPh#5=b08I&8keSzM9+;zxT8pKM>gD1-UBXV5yb$^4>k#cqYDpEk5@dVD=fK zBcEM7Dm|Gz{VDb~Y8yMK`eEO8G#3*@z&IX2~jhTJ|4{v)+#nL}<$j~{alUWhfil>=NRLW`2rD~Uk_Hu#A zKn&`%CGZ$4c^5MKUY6kX{ROOndbR)L~#kI z75CB>_ZyGO>u7EVk8gvR%?*AGkz!C6Vs|CL#~u|tia37tfdtC;%%Q(x@fMABs(()~ zBBDeYzql~nH!ybNeZzzIB`2l{0##Gl-o<-4!;4){jEb;`Rr|8c&+h z1U(miF-!{@zK4(t!v?mbkWXt$A`=F^W>h;RK7x1j7w|x9vrw0W4`ExII z096c*?fpah*AB`T#T9;|;IwcIcAUq$(_{h_l@Nbz{z52)vOqc1+l zhX&Md?pi;43|g8lv8VU^f?>v=mFFrjKAu78pdNXgG+H)2VSCZH8K#*EbnUp8QRDOu z#YYC1j^BttcD^uK6BTY}4r($yi6(+A*1FGgqJgM$A5KhILY=o)n~+N*zlf zquUDlmp+>uydaFKHl9Vj-5HJ+=<=)&%4XpiqV2__&K08Z1Eo>;sWn*7{i%4ZDdK~gY&flyoOu=I*L-FCb8h0E3L#j_=`**%Tr0b(QrA5}E{r)HJLxeAAMXv0TmJOBeFjYIPsXw*H z7OC(ZsQ3Et#P0@P6cV46EZckrHRlU;7Ku>fQF7cDU)H_6il$|?X+#}+fiS+C?dtb4oIpCQw8bd(Fs!p(^W>arK%~8Uf7l9RJasdV-E9=T zG2HC5$7}SB^DH6_IfjMECBbO&`Gaca(lFQ?fP3aXLpiU<&ywLO^Ai|y+INL^F$CfF z=p7LeKMo_z=93eTcEMmmN{nwYri(b_R?hp}AP@*S4H{G8eW<$nTQX6eQlF1fk4`c9 z6;tdASig_IHa}b|D$Gr$@9%oPIVJ;Pq*z5;r^=X&Q1^_z!)~0pY&iyno+g zVyn;_t4TI&9TCe_>e(Iw^`eU*9!?a#el#|{_QEwM34KdES%OznK!ja!-JkiW^DliV zp~Sju3uygLkiV$-AMYb(<-obi7>drSY}$L7+7DWI^SCYb;D61xz}mRhjgqK#G@F-m z*ieH9R_L@CC%GP@@$9{!;p>|q{r1|=Bbuw&dv;4R&L1#?`BKpy71Ml9G zVbxh$>t-s(d$yRLG5o`<2AZ9oJlOeM3np)f&3R@%g{I`$!McZ3`#1VSG}d#w84kp4 z$JzP`Cx9GMXJF600uzp!$v*CVFek(;dHfh94heHx&iNILmMKqEE7ZRBVaA6C!2E>P!gk3BaKS5C+44VS{enq`@8g93~W=HxZAVtSG*uM(z0cI|~8el?>Z z!ry(OQSSbeTpj4V#$zkF{QxZ1Zge|^s=*ZcoN^)TA(UHKt(?gj!_j2sN6Y4iVB=<2 zv~FbWI;=ZfE#uoqoqz3S`*y|c%w?3RwkDd{CI~+zb=pe^DnQ77r~7q3rsv#(k9}@( z?;0vD;<|80v;UvtsAacj!_So)X#SoMes_Br8WC;pep*SXhlQHwVcuD<-UC^Q$Jp68 z4b8v4si`4Q9`Oz%M({P;>!-T)?Jham)zF4-BssFDk%dXQ9 z(AR%E>42KnlDFFxRRUQ#agUc<5tNm;X61aQ(nW_-pI=Q=#y??DW$E<|6RP?J7%G`! zObh*mdf3`8Nj#+Jc)%*nr2KpFOE7ofMik!hG;|$oG~SGoD9?+swW3)4srYpN>wI2z z8~3-r`wg|TpZ{7SM#;0F&Z?|?t&<@@QZ7$(@<@Zh*H2Ec#TmiqW)b% zpCz$R$~71N5DlmB#DDr~3Qd|5?3z+npvoXq>e;srG+IwmSi6522K??jnclp7tIM~vIly9F9)%czn9h3e3q_=)3e#_8FL6 zPH}9#j_C;YceuKm(4<4_YWyc7dTc#kKf_F`_}~(d`Z{IPIx*ZEdG@qR&D4FUGq;rU zd}fU*rRPh`7Bzs(TVt14EezdE-r~oEl!0Wjw#*?C3?%b6RpJ%QP@A~KwiclROHRw4 z+UXd-`0#LVoXg%DKwe`I9h#=_ub>`LrP;d~w{VQK=hgSB&?15m?Zx5gmq75WlZfBn zh+~*~^4*sc!-tZt{5CH%KZhDLf_OGF1;c!~0p8g@A60aJ@zad!{=1&~s~oBf!^Tgh zsEPCUc_2?}#{a2qfCk1Bzn?c#bT(mrK~g>@V99Vp&_$XWq!%r<4C^PbxlMsGHojYC9-|>cVonBLwrHKD6VE zi?tyPs%lFgTvdiex~QbEWM7zNj%J%VZ3>xUUS@tzWl*cMxyFQ!C(OR2Yg{=<#jgk> zUq5ahbKymIe{tlc7h~}bLc*L#Ug|t-tO?s~v7)*Twf;;I=(5M+6AKD9q~v8h#67EJ zWOt>~4}i7>9!XWwWtiFLm;YIqYCciUYvnl^pXOMI2138%MP6e1*2AB!7-;D90~w#` z_IcJDrmpx!@+(orDS(_TY5soruiqrAOc2lrgaVR}W#?b}rB}98+Hi=9{~XP+`ZL&f zA3y>Le#gBBM5y@aAk|t^p6@@U<&d?=CBh>&8vh_#CHd&q@3Kxb#uV`h=P~*0X%osd=jhr8QT)LMZsY zzw70b8=t?tM7>e>Q@bu41Ja2tUPs#LDD#r$i2DOH9gMo{dY-GLenIUtHdm^v9zo|o zdU4AwjOV<*=6+Ae*H&mY&Of2_#S%ztvChnCSbnTl<{|zHx&TDJW3ft#Q8*^y$W439 zDZETjY8L=oJgC>9Q0}#l4^NvywN!+Rj2pQ8edWIv={`yZ?+x9ak zw&q&QpOY9r>+Erx7mxMk(fCtst8<0{SU=-_-qPJohh{mBl|A}|#g(cvn#_i*<6!=G zuUk*WCs;TnzsuK2oa(vO2Haky#Z&Av=u3KQX_H+HlN}GQXE!LrY%|l!j}gju#nGAn zVPX0|^IP>dUGdo7M z)jg!zS;SgNo&UQIEhK(64Td?5*b%uNQJ9}|#8nGQQH`h0mRQD*84Cv zl0L*y9tunEx$EoS2chZ4R>%9D7{AB*(ygy4+~+9Q4;Kaz7HNuMW~oXMNfQzN4#UxJ zf9>f>O-_Y^JNF^K|A%%wy(y3*<7!f!5fIpZ&ix7Cg1!dL@10VO=)0)d9?5;zP)n=A zV}qH^K(2Dwm2LA8#&$nc`Ei8<)n&ZPB9@XU_n#8%+@jcmJ@H!kmj|=lo31C7)ld)62+<(weWwd%KZxW7%%p`sE zksd6?sfXjvQ}e1UjZU>*!RnyV>Wvo#Vp$Vtn7l;zHuwpM_n#f<3^e_(^B`|iG<|#^ zmUoa*biM#Dz z_ruZUrG(zyDvb5lHhPsC($sv5Bz1>_c1#$aAe8f<*4)KFSXn$hWsP%31w3Lqk6wQc zg92b!vimn`4Xh+4i1xs0)-GG~_orYq`^TY$92;n65tmbVr4B=nj2bQ&TciHhJ4IW{ z>VVj`d9k+s9wIhoieKEJi$+&2-RTj+?7a3D?5wTguTs5ULUFac*siqN#CE;g2-ifrhYw&-G!%LHzLTqUWB=8 zI}4>!7GdMfMc+LKv3y^v)AnCd{#vkDA|IJ@)dxm)9y;m!Aqd)oUGj?yhG9jpYh&7T z6vh?bT{PW&7^Yp}D)E^M%;p~8%IxWeK5_aUn~5IiJ896pZywXD8Y`A>Ul?kGndql3 z>M2;=OXqQ?vpi*IfV|t^@zRsO`(0ZM`PDs5)IisKHnTYk78BTBMR8E->0o`1zA&y# z4oAa&`skf+4lu%cqU`F5A!@Li529y|L^(E1b(+nQsE1Y7R69f(&A(pOujwy})E-zTF7y}EbXoKIkUuk*iUV?LfJ z1&@dCE_g_Z{PX)XTFZ1b4#rUaU(WA1FL>u73^T2As}8lm%y(yZSLQ}oXMa`~H}@3f zi2XS|{N@B|V%HwJjp>2_xlF*O>pLH;6uev2_Y4h0_dCUSeQJmMPjn?{Ek>blvf6Z?L;izHhID7$P+Y;hNU(@Cs;XZ)7d>5pqhp#bK7Z{9<-FP(c3ra; zUvLb%XMXe>MzsqJ4KNi9lpFe`GXz9Qsb6 znd;Y+I0quiX@#9Jc>~1Yx$#Zkdr;Sn^Toz)|6I?|BE6Bb6aG5Yanb<)suxKbHIAUy zC07|xF=J`!`4ws%9nyZ6hyo)v)FKi3vYAB;I*OiG&Iofs-2RVO_n*P=daGB<>x~pC z_6A|2mjU8Ij-~6T|L!yXAFplMZh>aIk58-R8vt>z zpggF^8^+m>IDTI-g89vU?akLTp^v_2<+#8mAYN2b7xw;Z=bgJ_t8#@4&$zQ=zs|-z z=q*0=ZJ7HDn)7PGx9_HmZyb|Edm*nYRu576y7XPQ;07$!$&qP#u)4as5;>xHMFbGM zO%AsEZwKQzpSY&Fk_(ns`^E*4x zn&K2liKqPCFYeEYzr^s=?{Dj|*BN2_upu1W2l7oG_vXkS1SMlzb;|`mH5_*4J6z z-R3Yv{cu_}yw(bO7w2O)DH)QoXi{=!j14hj7 z>o6P;gn=t7?#%=XG>@X0QJTOxHc~Zn*jO`C^pZH#i=h zMd4|JUUlV8at4++|2y*E$%L5zNkS;z`G7Pm8H@5rj=G{w{1*S@&-N$>`1xit+R%dM zLo+#}G3e`mm`>;W5xOQ8FSP+S{;Q_(az#34&`<(zN`%V=H1m36Rf^<;r`;SIG=Aa* z%v+o+j~`nC5+f(cKb(s95RIqUQO*Q)mX zv)xr94cmb@EPqm1-ko_|B@T#>%5^I-of1^B<@lN9I|QohN!94BBaHLLz1ef0!UOf! zZ`#)XBr6`}`cCp_E8Rj}{F~H`zBmEF|INFd_ubKu^r$p@A!Qyxv$)G2=huQzvuP^& zqOTw1C%L)wTVv}NQBq{`ZU9B60tRpQEM9$4j^>`Sm|0a~`y;8qT}g!bGxXEEez{S# z8wQW}y~btsK$D6P$-(Fw>JCu$w?B6ZnpkGb8LBH`b?K0>{pcKOxzsjVy0reUen%b4 z%JUpi7_N{wwD6dfp6Q?8X>ihj{b2y6H?ZV<@`k>E503WO_xrDg|AQyoxpH*Gk*cl+ zR^Kv){^I_>*u|GN?W_n@0wT9WflvMvEV=wxq1n9~A~i1m$`d~WLvCNo_?d-pwEMqk6`S`Kx94qg!c^G@VEyoYDQ(~^@D<12|__l{OSm#`~W#d>F!skyP z`0siDIco!t+b^nd%!-Hf_uN<73zdgs-^ZgW{SJbijXKvq({&dVQGkUFy^%;^qU)tu-#WgG3O;aL+7Gld}qz) z-XF!#-~Vec?RF>3;csqtJ0=aT~CCPW#e$_7{;N`w?^1ppgSQO_i@KQtgEFm7x5{7SM5_c|PHxjvf1d{_Ld;_nGhwPSNbdrz@9sP!PySQa8k&PPBZz)VEh;Sl*2dMkM@>`K?L- z!G^_|(Mud=`eQ03JE{DC==`wttCYYIs^j&#`9|9uF@Ecm1jWTCa}=FT=wdEVpfjM@ z19%#4J-<*Xs`VNXD%d#n;w7lsUy`slczVYZ2pO@wV*@=fw9R+ju7|RIqZwWESMB5| z)XaB=cHPtx8np!D#&q+5v_}TdwuKr;IxM!>l)>Bu;{&5}$vX9@@tl~C=k;nt?AbUR ze=!=ywB)vD+o!<%l{^tPzTap`XZ}l9*hiE>_v_Seb&6k(#s`k$i5Ki)#5%N_tU&7*Bt$+R2*N3Xw*An$zF^{od{aTIZZa5V_+f65BbAmLA;Y zt&G+{0}b2mN}GwJwOz7So2Thf=i-$!=JsN!<@rkMJyInan>kLtb>%o(R9zPt9dko@ zAA3Y>OR%`ddjGEMGsDVwnyG%X1`S(0>q929m#<~evV2SPn6xCGLDG~dvmN6X>im4{ zgvZAfJbT}Z_UkUHFmd;%`zk{nHvTS3lJ3tuL_~wL1`mlQ&?m`zhga|%^cIEPCVehK z<(h-!crT2+&-c8`P~Ury&-Ah*p6-PBn0y);&$4}|&+$JP9(@c~wSjv?jYH@Vix@U~ z_!QNV1N%%jkHDYMROiFV|Eve08ei3SmiTrPT=6u~Hhy+hL_{idSlKOx#i#myTdupV zv!lh~a1PElSl(>$-4Rp6Z4}%L3g1rLd~H(_)${e5>^8BtNfsheZRu)O)e4zEOX^$nUz z5oo*c!5c_{y(}E|So~;8V34jnkivJ5dZP+B-~aMQi?%Nx8xKCCdhbj|b{S%Z9d!Th z9ac@XL*yXGM>ioC{h8hq&Fs?)C45cEZwD#35j6TJIoM+1vT z&jya;@C?_d9shh0M1;x8nH!-r(73HQIA+imM~i#^_@xA9kF{GTDXxl^K)=W@?rsxH z=X~ujS=HTw)7*mbv|p272tLt9B%`ZQuR|>H zjA-ZK(c8)B-GKyab`4pyWZ-O8q|pzBJY2~S*tSs3ACki4#M3jji1^Y%NP}(|&+4^J zl+9Zak)IYD_mndt;-7hI{iZ>z-%re|9=&UiR+sEZ8=!^e&vG~_oQy(UJ4SgXzs;e+ zmg}1Y$T4X4VB0NIU5f4+no?hLe zpI_N=Z0}Jh=Up)zsT@Kh{QLSo@p(X<$K|FA$L1lDg>ApWx;FHtAG}CDTMHv{5$-i7 zF&s|m+2jf<+WNoq+6SIp%Y^3B0XXvBW*8~$C&$-)KxJ%uiH40BUEO0byl00h zUN9qR72$hH8B*Ec)(s|Ux6s=y#eTclF5aBBEyMr$ew62!r2k~GNyrW@ z>K-64tU-bX;`~GR&UBQ0}i&rf+ zNc~R!`~$6a{F=+#MbQ@kJRxM8SePg3=IfQzDl$a`k%E^Qv*Kt`=-!w!KM8ew*NQv; zN(E2fxgO%Qx)o6Eho}ty0Mss2YLopGhkol2vz&{yAYOcsE~k4BT79jTm6cDir&07b z{wcqxay0QZ|FiABgV3ZvEAU1y9E}=jnm4C31OEJN1hfiho?iPDr|=kBO#YPP zi!CuOL9r)cSwmtb<)aC_u6tAHk$McK(oKHF-B3m&(yH#EA69@gqkbeuuN`K&MevPD zV^G5K_4JP&nEjo^bhqJ;*hjtjdsq(T1+hP_Z`1g(*h*B?2GK;3}n z1R=%)#!oh%+@~i8g!Qdjx5s4AoLH`U~thZ6I zTO_C2fgUKif0d2&oE>%UzrN#^v^MsBwRdvtQ)EzYWPeBGj1KDesk&JC{nkInO)S@@ zG&_S-=nR{u3CW_u|3hQfdnXA27ObB-DiegRGGc=bK*E+5E3ku%+u-sft$q%E> zA4m0mEtBvx-8F~kcXv@efAc2U1)s(k)L@{zUs6mH#f<(mJJ7`itvZB#Ym8Vv|EtUO zZtuiK=y?B)WfS8T)Vr^Me7KCV&Y@K8AD)B94x)jhH4S6F>Ug@ZhxX|E??Q8Xn>Xc+ zm?H9(gS)o7jiLo9mv-xINAdLe8YWpUtMIfc8C@YJ`KV2}E&uu95=cFxxp{F<9h$yR z&U^K07A9;D{YVy!L8EJ;COfjI`Fh&J9DOb(QXQu}*K~HxIZFjrXnCeQC+|Isr<)F! zm{k3QXD}Byw0GCv{XJ&5;_K-=G*?`_mK^Gf=G}Sn`*U>>(fop9GWRH&m2~Xi);)ph z^1PCxZ*ZeUhd%GrUM)0cd+&wsmfvWqV2mMAlcEoURt^WhW8&9C3kUufUOLT%mi&kV zi!l_u0wNKb-wbflLi?P$VdoP`G-WtjIFxJyeHU7E9Aa)l3-a%*JW~Mmoi9C-S?H+V zCztk1>eP@wl!QwsO7%QNwF=inP0lN!){d(i#tDOnd|ft4QLGS6^?jU<-ns+kBz`@1 z`Q!kzeDds-mG*cVS*OcoKZ_9|u;7+J-$|4}NI#X_is=UpbA7a4PMA;L*$w*RTkWS;nh?n?x-VRA5LFlZzF+jZfu>#EZ;St6 zMr8eA8}U1`C^_R$cfD8#o+g}E=N|hA>MED=V7>Db&3{?ZI6Jfjl{|l!y!e))4*{j^ zCuh3EU7^hI;r{saLdZUQ`_d1h8qBTCDMfAn1bO>#xUI%^s8PbUR^C7lb@<&1zM-`S zwQ_MR=d~Nrz;MF!t;&D$h@dFVv-Mq1D11q%<7eRA7e$|-@b$uLoge~QX&1h9>IBBy zS(xMYNc$32XFPs}z5K)Zf7%gHZ*51gK zp&~QxZxdOTl=Dd+>MIFSKcV+kpTM2()cekc9DXCX4dY9i+y1-bwj&W*E(a^gU3?6L zqwEc5J-u->QxAlKJkH?gIxl=j&%*!pzIggHuiw6Jhtcw`Q;j;8XYjP0K6kx85unex ziDlFktG8PH{6OsbttB*bV2b^R-c^_x9=*%7O38nqd4rGp*t0QS%4+F$o}13zut+3Z zX(ultf`qfb&MJnB=^gwL$$X(1&p7zpS>wFU|9mbSL#-oC^qv3Eohbi0t#XHw7el>U z_p~gP1jEA5+QuChtI(AG^Cxdq6rgoi&WZ(rgzeMv-QhnYCSZzPEKnff7UU*8*DSDA zh6+8ElUF%8(XwI>i=0^wdhz^|SHsD}FzJ&ZbjHpB<#}{$6nFGMr*G@Sf~p{>Iju6} z{9PJNtnSEP+%JeK1jzP0^!m`*s@UvEFV?SJQsEsEpCfgy)4WEMa5q|_NyI-tWMkgnTQ^d(pJiU*q(;P z?Z2LS#+#u520#6w3lgYfVtPw^qc2oF3e-ucybew5i{=4}82*IVe_kkHn>MPIyi$?# zB@LDE(+cv$zDGlQNWHF?^r7*ME@@kx6RP4J@XP+J3=^tKhMV`&-)-dz zKE#0a*P}|L}3b#fm@BZCjAc`B?;xgtJE&fv zVmBiAnC^6V@dw7Z9>3B)_7a9^`)cxjFQP$vFUDFu6{yJ_GrYi#@r#y~-x9rY7K?8# zc;&7&X=xzx_wJFbc`SZUyqL%pWoHiLupMN&Ez?j=>Y!~{k%MwQrnbu*2T-pGqq3nT z6B^oLC!^H*3lZPlu*o7C0r`kp3eTyxXfZ_kh?mwA)VWWBhO?>=)!rBWDE%iL25)aX zj{VY&@>DME(QBE4k%%(ZpT<&XPP={DsGkhP&7RE$?%{ur_t4~S=|+vBtQOMGsN(Rb z&nl)@8rO%$j>kVTUa3ao%`P}(aClmV-sm0C7f`t$ck$VtQq=ut<;+!K zaWuS_Us!m)5zX(c{F=1c0WGz8k3F5?LyN8N!y#H1Eod}+QptSyci+}G?Khvq)82gI zjC?ku#s}Y4JDqLNxK4`h?45hi$N;CG$1dS%4H-@AtFNFIp14ydH?yOW!e2w5TXdmg zIjQ1zj0n_+l{IuppFxBJ>23QO1W;3AL0qHr88pvUreI=-#hr+njvm@a8e!}_qu}9j zj6ab)Oee)2V}izK3Cm}3#$)65#A~VNyZ`Q!^nps7)k=VvaXfGJ z+7g;1uF;%KJp6ZG>+{Vv@pWv={dES+E^40MedT<9CiQuDV;dLAYyO32zW1%}(DdJZ?#TkDL(~GA=~S2LtkiF)AIL%@ zi!Ei}T`6&2G;aOk%Epm&s8GAv<)Tj&x5BfG77uPrQRz8B?d7WjonO;{oSN0hM6kiL z0BhCy4{E$z$pg~}VT{)h+y91symLz5f`;wiMReh_P`Y4BYf&+^?oDI<{xg31zfqrS zI8WEHOgz0>4b!nyIy{p&bA^#Nrq>#&y}s9}ydHITRs@|*zl>&f+}kqrbrjF|KD)((vzWn8j{9$9X zC_8bmvRe_)tX`t8L=c1yZ+c_VLI!A&+`FjSW`Jh)ruyAhy@&?){WSRWqV2DJ_TyWY z8ii*DPdDVH@6?6m6G^FO@v}o@G$wyo;pY7$JnK%2UkT@^_?*z{(Uuy$EG#e3GXG90 zu!DjNL<91D#~v2`^(PLpoO}LYJL3{PUxQl|sJ5 zpA}hyM;{Fk>AP@r*_tuc{p$2-5?F7HBSNyW`4t=$?+IFMb`~~1h3RPdzy27Ln}pSQ z%yYVmi*!Cg{r3bt3F&$$|9<)VmMxfn(w@X{Kv07c*Mp$KckR**CWyRp!X+^!3$CIqQ1MOc4O8Jgm@rN=(H@FNv(J|F{6D^hxhr`b zT&B}7a(;NXVF?Al4dW(u9=i(-p-RETqD>cD9|^UY{+kdUT0Y;3s@tCp=Q$70LhwG5k6%lgQO)VD`e+kNJYAcXT-3$yc-m?8 z?>QFQP$H6j`mEGpAWY?(exS*K4*z(&u-bMs!Vgig9j#Dx;MAGNvCcs9_Bj_&vlq?c zii1fhFVVck*c)wk3cedHR%)0;UA2R0rmPc3;3;wzi^t z;t@*SS$Wc_^CRkey2?;SQ-_ig%fBC$bivaeage%H%mK42CJUnXD1IZ#t53Uqwf6v& zirlxxJ72)l{m=>8ao+|F3KXl#<4e(ej}%{GDV2U26z$AUI@V{1R&A{sTJcX|MI!BM zTxmLLv2jkgU`!8vw;zcRr7-_)Y(UC<+olIF6Ifz}Xo}YYSFALTq3$ze>G1o-%ES^n1-P)w;DV|lx)J|1Y1J8P*_WXb1+18E-u8+k2M?K8So$t7& zI+!so7=O#w0E88;+_Lg~Jo_13ysh77JX74$#EEixJj>&bC#8>R@w9Fy=hhTw@a!co zw}f-_V}6FRqup~Y>3`1SfwYz6ui{P6Ypk@gLyr$0sO#`^`eEx_$C+miDILjZaOK7y zFJU*zaldZy;g!^3w6f-)z|1R)?Y~FA-k89P!g3RzN6D?Xf5*j@{trJW`7p>e#>))k zN}}3#&S^R+RQfKIcGB!vPnYec!ttQm;PsV5G&I=z<(<2H@gbH!ufyL$iR^SRzt9k@ zx~m)!4|CllttsJYR;ORC5A1?@W_lfOoax_wb+}dM5P{mAhn$|OYyr~q&9=wx?ESmn zjT@e;p+TfN#R4{wG;BO1hrL;3D8ezts^TZFrov!qU?9U;3Z4{4vp;&O+w$Ky;fYZu z`{nsWG!)OFu%A@(uk)zhcToFvrM4Z01MLmJwD?(#76?r5Pj~83zAKvN{yFfTl@a}s zNQ?`J$M{-0QdFw^mi=*b^zZq(&GV`EA4^shz1!Hnw{&aYCMBU(SRPRf<`~2B>1Es{ zsuO)fsC|0#9aEn~Z2v!LyF+Q(4QAN{?p|>F3DW_`=212lr<}|3>rZ>L35L$&*Dsmk zVe-M*=B<0JVB^Jwawa!^^ty@VgWRtpFu`bC=wxdO4RW;0pT8CZF=#--v7Qr8*CTjL zGe!}JF8-Ugzn+3Ag@wV#aoLE}lJZMtFNS{}KO&k<>bVWwrzY4ex}{M~u&S(l#4z-u zrXahGA|R-^+1`nw=5I(bu4uW&QiuBTTOyUJ??U}|L*yQh@gP+54~4n@rq-cZ@S3SB z@WASeD{W*ePOfSK!Fa*EIUmD2v}~(UNEESxg^T{y2D?7t>E0$3(9Q5enupo?3IDCI zkoMa0&jSa@7yo(d?v4vEaq)Sdh>a<(w`CW(!_%0Qx254Wx_FVND}oe-Cx;YW`NT3qIZnSU!&heeUXi- z7%q1voQHQw%m@)lb_Ytg6Jep`1YNy_&EIuWEU#ve8Kuyuhq6I78rx~K&Z|d-Z$#@q zZjo~LOrZXsmH4^GiBNiD&ELgj8WFV)prYVFua(_R&)0eYI3=%lB1N1nmljb517Z+l|Mf>ay^S}VR^3i@#EU_j!^?j4Tkr|L;X=O^}vbV|mJD+<%u_};q3(f))v z)V+OhUvZLz{;Y<)+(Q_Kk51)|@XEjh_dTz$ z!cCAwi!0m;|Ew!VCZZ|lEvaCn(il54$a8kpYPV_8zRZKAlF|HxO9Z#w<3Jg1nc0^0?NHcyNhH!{@3myUcLHjGyEdPl^9g+q?%tt z&K$?jD$T%zPxZr>-WOo$k+@p-i6Q82esg2j3Z?EJ<$rE=;3bS9e8QUy4nBF*^v3b_ zaON>6Yumn}$B5}b6IjiC{j`*zmQ`%sh#obTCLlK9wpBkU5|t*YkN2rCzO~ zJ9*p&2GYt*vJ)lI3|Gh-``5oV0e=Io?p@5!UN8x7?Shc-iVYsr_O2rU@Bt}QS3V-Y6oRL$al ze}c-BDPdR6@}Tm8_-}L^J*b{}sx!RT3|ijZIlR)w0F!TS1oMhTK({Mvd2XpT)M#vt z^=0LUA-#tdJL9*aeonQ=w699hMAzG00;EcmByrWHzsIxnJcEeSjJIv z-0%W}4W_>}almVNKTR1*R+p@rTEp~JHm=82aPF}{U8zgDd$wY_hSgHPoNeCmz?8p^ z*R6y(%JEJ%18EcAMHK&5Rz~J)0%|#%RMr`=gzEkJ>h&WI!*J5*A1PkVlQ7dbB$H z+~P|pG=HGxOmB_a_N~0Trrn4V5AVD@)|3cYo7?YY{G5Z@4>yU-{EuKt?pe%r4K0`{ zi<_xvErcK6%ihOhWcs-ZYmwc>MH8F@4Mxle~*d;&1?-{DnVrz%>(O)F%arUTg9td z1MhA#DxGv}!!y_0(u7>PUQ0Y0wzf=9C{Dbb`6q&t7&Uc&b@cgRwAhS$xF?gszX$IR zO9gfKV*HMbwreMxTGXIPb@lepUUiuBl$K&=y9kqo?MLUIhr!gG;Nh=5!>Fi+KGJH} z5me=z<(e7z8VI~euP)iLqY??4%aZ*p=-vC!Uyn~WmbK_KXiJe@c%LQ-{Dk;{~rKsuNIOdl~gE_kkRl)+h{0}hEfVi zNr{FcRLW`y4OBK685wyWdt~puH-~eMy`P>xp7Z%$-}7|+e!u5>p6h!4ayjQdpU*ws z_kF+T>-By+3>7ku`kGh*<)f9JV zbHA}Y@dX19I?~+r`vgbs;^4!@_1FCTpf~*N{+-*oG4Zv)OKo>=9Cso~J~fBQ_`GmHTQ{j4~P0LG*g)r#9FzT1c{VwY!WK2FEo6|c8Y2@Mj1hEdR_YcTPui}E1 z^fR&9-a!~MTGA{R7=sjv?}mC0OtH;jo0Z-lbsX``;x#Gz55LX5kH?Rs$3b0zUb<4Q z3>Mt<mXRsyU6slz(>N z0^4gWzjV8NR$?bkF35>4+qn^QDnc^e#L58K`25!s=X`M9m(72w^8=I%6bG(rO~aAE z3nf9?B0#m_^p4nsKyghr-fa8_r;YKn_#a~bw7dJa7h5;spE>3=XB6oEQ^+afV(#c7 z@>|L^HWkG_+V|h{#jqaHUxEwPa}a!;d7ZCg6qb8K$Ku^=H);C)hYgMuMpG3xv84at zn{R%7K=QvM3@K&vK3FEMTXHhQIb>Mm~(thoKc0 z?u?K)rQLfEYBBJ={*U7v${P~(BN2YDc0pb&T1bQ1(wEWQArGK8Ex* z_M69@S=CriHzygd5P;?-)j>h$2)wR|!|tuZOB^7c+qnLsH6PU9@)Ftq=K=KXN!|V` zhHgK>aJaH)fWauS&jf@p(4S$EDA{@IX77^j6qe=-XDI{8<_9m z5KC2(!} zavKhKN4#WMzfn0~k()upJ@aG0#L6f&782M)j)dorrCOy`%x(w%3oS8+Z-x1VZvlJab!*|e`Sr@)!e3nz5d zHg4wtT9@q<%eyPkSbH#>`qc-g2EB8ZJ?X%#?JUNj76C}sD(c&mdIkIS`R)&0&<5J% z$NkTrt3pP=vjo`!qMpBlt!qP!Uo*D+e)T{)xdGfheelX{RD*s+D^`8&G7_7H9Cz#8 za;)N3*&}x0?my*vuJFdUeM*Wr{7~q$nwmXQ#r7wy79{F;ry8T=cBIkw6Z~;VVB2tg znvj#R-*H_Yf9?IrPL90uW;nE~v5dWoKL6NIeX01`(Fz>NVm{i<8;k?YTy|wrjtuKv zGko-YpkW=1#-y#e>{bPxA8vJ>TxzVw{N+o;)K>x@W&$F58?FvjEmbFCb1;k>~grX2aZ!-T&^v<=F1~Ik;;=2TJxE^@>~1VR@?UQ_Z3=pm@^Cxs9PlxrNDk}vj zyFWtPp2dV2>M{m>B}aiQRD3w(IDy~Q^Lb~BT{beOU!Qgj zYqR&;*vLxb_})B2>$Z%4y^lCAnP!c-eO?3YmBC)Ml||5{sG}%7X96>20!rl0WK0S$ zR8K!b)J>DMo*GII+$OOHyf;d$BXD(wuD^EiFE7CUDN+4;embreiEDaVFs5@Oju~3) zyyHvv*TAS8*`9Ow5RRl~=bhi{%`hLWFD~vAUry(XgCeO-kM`>`IGdx$PDckPJ57Fw0*&emwT9m`?E*cATj^Bmv<-C8Imtw>N*-EgtLajc~QU)m2HV8 z`DAkvORz`8xkK!b(Bh>$&}ao?EQbY+94>6 zg$pkJ$u=6_fyxaV>P_kA8;NOnqP z3Tv=wO~cH_CG}&WnK-xVB$w&C-H>`}7ssFhQ4c$DgzHqRBqvlP=Y|!`-v#o}jv)2C zqfl!g7UE>{0^+wuzO5)t!)dF^d6zpOP+3^Ex=7j&>vg`beP{X$8zev53Dr*Hbg4vT z+M|m&V>x1Os?dpn_E)WL3b*0daP<4HEJWO}%{VetZM1{f|N4I%TArT+vibL+2NvwC00T8~>uyAj>PVXTK^=IvQ2 z+Q*>I&?6$TFcYUD?u9Rp=qB;Vy;RBZ@Pv%V9oL$B<8V$^I*uzR7JHc=KR+600##O_ zSJu`OxKbU<5^UDVH(~t+2k};pRuX%2;qd9BtbfnBGw+uvUH@zM-D&KissQ!z;G5=; zoj4qMZSAF^O*m~3T(Qbmp2YP|{5^}23pRe^=vhM=!vTHE^G6HT0a+?XpUm|JW2<8S z3Mng=_?E4Dq;>#1&;0&!fiDVj?R)2c9{q-7fvayWIiA5e+??98uNpg5%&F$^0}G-? z%MCdn;)1P-?A^dE@Jl^Z^kbMh=3jnksimU;LvdS<*c^L?%@>|DK0f7*(--C}HYu%y zMB$VC3rxpxbhrE#bvHhom`Ecx_K_j`xG+i7c@8}pp5MDDq-faIW)c#rHL|I%C5?4CjM{;)DPr z%TOqJ`ApP)wKdFi3hv%;&K3`!sAdzd=Edo75`zHQ4RW)Re1!6@2EnZF9Aouy?3A%$}(Ts2#Tpk69~X z&HH7hcU~-q7ON7YmfW?_al^y(RxJ^qA6>ZZUt&SKU-*!|vpZa}yf(@O@zS9_35NkTOtkB1N+hItH^(|2`9d5p5N@H#_M0FC@;jw;%ZZ zx(y9CI+ZRdbuh?h%g|?$Ow}92K2S0f>s(O7p48Qomze%phY8W?;>MJ!mp#OmmhE?L*B^kyFzV8CZ;PNw zQU6hptsb^_%uvGxnQ=DZYml&DAq?4EVVc@a#QE!AB~15!n#U1mp&hiLjZm{xd#t*! z3Wwvj2t4(#hW6j}(~f6XV=t$qQn0W)7F6dQOtyr=BR_M`g@v`iY**IG zQ`=b}CNU>4^y>>4cUOF&`;fpfd1)kZ%iMz>+UEb7dc->>prAsmIP{J7MU5dwRg=n3b;@EItj)xIL}EF_kN1suyt zg7!Dtzo(x_z#hrjC);G{ybI8^@nuZQ$W9pUVFf)s0ccoFaagj#7N%aY3hHl&fS#=T z3bWroKzC&;ErKOkjftAW5;S!T2D5#WU$`wolg4i4RY_O=eh&|>oYVu_r?FaV4^ilz z2oG`mA`WxQJO$XE*5ia8$AjFcyBN(GsQYy76#V{iBrw~I$WLzC>vNTxvsaDvZug<~ zF#;#xEA{i4r~Sq-BIatgb*~>x9Iul@cGwB+GlTdXt*x4=F^u} z9J(K3Y1l-C5rsdQ$=<86{`2BR!`xyRb^*sLz>p^ez1vrD>>b{RgM~k{tMLG~EpGU; zftL?!{$%WYdA@nMoBiF*`_80JB&+u<4p_)Qycai^Ys<#2ii`u|iuot_ywRtrB)nFpd%~{!r z&W`|tR$tCGM4iB9DeghJYaTEbA{%n3!IU_!L_DbeM4b_d^V>+FAayXY(HwpZ9V;FK$`2}_bJ7ML;P-n;f1S}JO ze{yGUEs5LCVl2(n0*lx0TNFCshM`h>KY6WwjAa!$Wz+XAVb^V`9aj^%ASeAtmm@bP z_6n30T}#~$0rxBKkUe}L-h7ocHGU4e-_Gl*3zq#)gwO zHbFB}aqHKibtI1I17?}gsqodItMYptz3u_)T*}Vnb(&$_*IQrOUK06dl$=<*km*%8 zQQ#mE@qu0+qQ+YHbCrd1IZR5tXCE-Khx%B%j2-d$B<6I-5lO#Ts4uVm>N3zyVikRu z=X06?ACA~}=L6kWoMy!3rB;9K?M(g;evi(BEM6iEc(n}k?z8LcOoiTE7_`2z=a|G2 zhVNQa|EVC&>W*C^Pi>dy7DAVBvA_MZU=l0m zlYK@5pRg|~J+giY!M9UMIUpTT@Cm1Ug@5T?JC4)UcaqZ@USQGwKQdz?%s7_4x|Zoy z9kwY89AFE&iZrxZ*4to%tz{*u_lM8poSs@}fZH_^i^Zjy#&gm*9QRv<$MiIg-wM6< z!BYWxjcTe}%^Pv{KzM=5%^lb%^t|5OO~1C<1-e(wepcVYha>NmgFTa_|1mwMjU|qN06P?(ce=aeK(8)j_YccSv63qi;zD-i=R}gXF&NkcDybbSQa6^>;^I-L4k*GbbEe!nad z(V07deG{)*Vw9_(UPaTe(6{<0;(BUgdq zob&wva|jI8H>b1t5&6g|FRjjGw9oThpA4w56_APvVI@w&wW& z`49a{kU=teBcqTD_5K>HPSGKdQR*lVNE(B8JCmuqoay+1*kHO)#JfKUn=O^Te_`5< zoqMl!os&O={XPc@j(Myh*7@pS)sKj&D`;k*Se$kXJbBnOSj*#8qKKKCj-4 z?J>zMUe1R|+{b&LDL$)%g#4Sc57yK1gfPT3Hj*uq4^Nu+YFb(vV7pLZo^EF={d=9iya$jKa78Uomu^=(XYXD1{+?tCmWmrI-v+~z)(o5!$Bqz9`v zb|aZLXE^YUI8uUGOP-iN06djc87EZp|N0&HNa^SW>{r0(HEByFc5Oq7gLbjInkw|Y z4O^a8vj}}dJHpmn-wd?tT{bbByrJD@K>NL+0`&DAH4fi&9t)DBif^C01!dM(+SWOo z!b(rmjD>7Vm`WB(QEjH<^I)6R0ec~vm(U=z>XNpO9Sn)er3IWch5m>QX9wcgu=2_)Bv55mgq)3ijJ5mk+nJvFYd;k3?C4Gj z!N#?dOJ6=pgBt33d$SNG9Oo4n8!@Ka0~jeUez*wSF}F};*ORgZ;@q`+GUd^lhj>Xl zYx3bSoD{BkepxXSVtTZM4=q(hGN&u=#+xN1W{$!zMK}h<-*(>Rc@ly1>PjA7Szy{F!T$pAZ3IxYqq=>7LOK%6sHe#_acmLm21lC_T%(_xe=S304_iQ)e18gJqv zKP`giLMie!mJ2X`eI|OyqX_#$MQ^D7_P1~;0x zqERpor@h!BaI_aXBnLfu$+GbK@&lu~N1LE}gRhR#jlbuNuTvA}wphST`My`XuVdd2 zy;19HA2E^Tn(A)3zxK?jljrLzH8HMdVRuNp5Oj)^ewW_h1ud*+^Fn8S5PpFA(P8DJWd_Fb$Jv0db<54alH z3@HZYJGO_3AT4YxfjdM7Ygz8i9GW2X-_-AOmb*8k(AR%x=gAqv&AHewJ$LAWhAhI#EH3`{r>lYlBhgC3^iiR6Dj-?apP#j{Gm(FBkV=XAj4=<$;>m|HnhURDla= zmL~V>&az zrC%_D+ErVMLZ00~>d)y}ul;YJXW#M_*;#JA2A?~$`UA9ns8av-cJea$C z0jVdAuQQ90Aa~ARTbV@U^Yz@YY+-#y=f#1FK5PHygJbmPkT#qyKglXaVvTwqw{ZLw zHg$1d%sD~uZFV)S?EBE*0pokzs^bECgHSV zb5JD*4Li@IUCoX(!I6edSw$aKVdIg;`}Pmi0L`>lSc^O2@A8B5IMs>NhaoSnZ8w3Y zkwX_xSJvTxwZ_Sjsh?Q$tc+GK#zA6L@%>q5b{sQD9FCq$iNXQ?!5}r)FreP85k6B+ z_~F}N;uxoAFt+Cf)=7Ie;&?3AlYuTGf4)_26y=AGV)3iAi8#YzXgtf+P4bF_u@%E- zG%03KsoA>2k6jZ7io8g_*Udvu)>M?*Ei)1mmyGJ-`IRstG3}ZX(}*2h(TW~{^@RMF z-mdfC?gDeG#MX#-4P!^=m*Wx}zu?qmx0J$_JJmQ|4hdLpB<5{6{DCz0;6BLk+o^tF z{m;MW2`y4)%{Cl5ap(PZuQ(`RFJyZ7h91wsS^Zi57v~G0e9OHOMVl2c9PwJMi}xuy z{ZdNnD;30c(iPUFIla*EFrxrbK(4=czD^VxAI;e%FGyqGG1IoK^QG|XFsYQZMFBbo zezo`?I*N@^v|Gt?pGfRI9tPX3dT~+pj4(Hs5Bw@U3l>x!q~@v|j8h?aVe$<4PgrXc zyi~u++K1NVFyI4YT^^5Dnqm+3FQnalZZrWM+n+vPUh0jBy4z}OpAh&fy+Il;HLh>M zcEL^K>)*K{^$7nZc}q5|GjHmT=b_-RQK`A!ISruftQJ@Gc!_;0r&aVTdLe!?<>t-4 zZJ7P0j#a=v2l6Sr97V?~AU1Er>!do}&O++i=~*Mbt@Qr79&ewdb~WLE*@Y!$>HRRo zy`AQMe-Xz{3>T6Wcfsfx(d}7h{9$C@lZZ=i{UEWxwL9Yp9S;pk!nBe?_z7I&np5T7 zC6j4T%B;Oa^RHgpy+r$jT+k_K-e>UA?xQFc3iC-xsGfx84Hdf%{Y-&O?qinEKAeSo z9}-UQjzY?jt!rD3bwO~+{BCBpMyQC(*nZ?_0Th&K}Qe3r9Z`fV?SD%Vey*J^4&xHr&#N-KBPz80W*Iy<6_M;jqt*37(J@kT0|==gkly zKaJJHkEO)vxb={jUNdNY@($$ao;rUlg2?|N-E>km$f<)ulglUc_QgVSD@iF&rFhd5Bn&Vcuc)ka%Z^&g7)*Zh)L6sD>B3+VEIo)6b=k1VTz zMydW+a(%I*m9LKDWuFu0(`AtfN zuPfm9_NH&2j@-enJ3qB74~0T+iK2+jMFO|Elg8)y!N(83*#&!kJ+6$yx0<6P@hE3)s=IA<3_4p^)q73( zK~w7$&ZB`ZAt(N|(L1|-7*+aie>WrphY~3c5n918^3-*ISURDAX1syeoQJkwsLKv?QIE70z^=Ki~R5mc(SMHTg0$ADYHXgKc_bp+Ta> zq^`^eI~!$;oVPOgA#t9x&w%>yI^?9@Gdg^?1S*~h~I0>RCA3_a_6sU5rqc zJxAbkjh_8>F5((Jevbt-zwb>>JwOS)>vq8T4h)_0`M&YUO`sTV*s=5FUWWeG+V=^V zMk_$QCfCo`c6u-_!zP%}^9RT;_YZKD673pSWM1KO{3q`6|MV9Jmt=&jyZo70&l|Ic z9VQ;Yn924v8=l8PTgWYSepyB6O1;S@k@!y>CFpQHuk?Z}g@0cV77Hf8_;8IX;W3GJ54xfnQcI~s2N z{<(Dy^BglHH#|Rz-74N>)Ai}_&2rLns$Yr3+UWYyr@#%Xf12-p8AJ@v64m0A~BsDC)`;==7MWJLKw-E-u5PZ#UMmv(X6#rAUMmP^8Xkr1Y&-yY%36BK%;nI|JD4_nwGYSBnT2X-jnHGeT=9lZF3vutJ}l_F zkL~V8PH<%<6x?%Opxv;5(IwIYlU>qSVoYxQ{>2(2H4bkvJkN!5BSKQyEE|wovDtVN z&p{G%#))s??Vqs1wm@WX*M6){!~7BrZ_Fv)@_4z8Eoq7Vi0(B3Rv5HDm~19k2lbQ~ zk718b*x79*&T{o3baO1@iIxuk=e`}-?eu6D^G(d=XzRT3gGOTRc$O2QE(P5eZ?5ya zPT>!AoXv5)VJc8R?Xg)Vst_nQ$9GqHUB=9|VY$=acK-7{iJZaZmhZ_> zwP#b@bQ?WS3Mo@U+zb5nSW5FD4dr)X=aVEaOPAA7zhwKDZ{260*RxzJ(YzVD`vV>M z#6Ccmq^!anUn!UheDR1Dtw-XR+SV;l$G`)C`I&Jk4`hD*t3S}a_x_8TVK%7CT)QFX zeho%$EpES)a0c7wvs&}+v*V(;cFYb|bNptL*jss-;BjplseZ!BqQ>yPF26fIt|fnP zC{kc}=P`zPfNCxwp$NbCB(|eXw5BJE#Q49>xF(-^2yGNh`^0<#yhy6utw(JC-V0`7 zVLyIhhKO(V^3Zx5)@@qxjdLGT3oK!Wiwj0*@2YOp5XBTHP3tJJd1%~uRB@2S5<9-i z^-t$CLK9z%m$=0g*82JkP%lluPsce9J5l;P!3baVTb4$GcWG>FeNfgz9q2yc@U1P2 zjt`FQnz7H^d`h4``mCo(dmmKZt==3_M1jiay^k6%67`uKIoZ$b3*JG!Zkyx*4>=%B z4ak3PcY%+6cjRsK;-J6ls;-t7V_kp$oR?;2`xp)#ZoCj8cO6Eu@%x68$D!+H`0F(h zS=br=V+l`2E2d4yC~nNnfr3dl9wQYc9IG0ZSgH_&&3B4f+-g@tNmguPt`{@5b-Rc< zC%(eb)@ZxCZ4o#wyWgUeyahtO-g@wYH`ohoASIyZ`q7O}toY-fVPSD~hbo}AdBFIjnPFJV{3)>xI;a>c`pn%;J) z?Zo|@weSAB734_lNPb0nDG-Atd2)BhPZ9Bf@Ec@f(#90XRywlkz|vz-Zu47)f3GC8 zEw#PH6w4gcV$O?)HM_(FJ@h{3z1*j~%>C0d0Gy5yv@6Jr7}%z%3rN2VVl zEmPe<#SaIQ22)A)yC5k*IpR{#7U;G=|G{&gD^Ol_)%{Wi=)HL}@ZeKNEPqqoDf!$T zr?O5=^P>-xbWB@1tK5dn(##8&_T0lB2^s47%in-3ToLQCHWm_7SL)E8&w&8Key~uoq28sl9|iqI__3t z&Bn)L$1jw_n2|lxN2^X~BP(d~@7Ka!{)y>#<#fDo7-F9bbt}7q^;{`lKI?>$_V-=K zHo8Ks_M@QA$Jc>U`|C!|UR8Y4*&q>D8w$fWsE0-O((M?i)V|4j>?ysj6q;L7l-FK3 z0m=D|w4=`hp^~|5STpKB_|;q=sE&L*2pwh>Th^O%{ar7$w^9dg(Z5f=FI5vXGk}S8 z{?$GbGdTOE_}A_eWiT=_Bvw{tid`K+Loff@?R}GS-t3w$VV-B8ag)tH_*47JQMbSl z>jPc9rPuaDYrpmi=F8`Cq^ophee7yArdcJi@nLyvXGwVz(((WomvM`%olL;y#FeFK zs$MY5{ownJw>dcaO;F)A(b{W!^}nD0&--z1 zkTlXUOXmT@vF#E(7w6VOZcuu-)i*t?@6jJ7<(!AXxPYv{Fh%G|Rw7j^%5P=tv)wE0X?8iAQcYH8s?Qt1s*QKPkP*t&l z>B;l&hYYd8ch7Fp)={Kx7-BkiU>g>G)Oxi}fgukK+f{|$T{PZ+z+} zy6YMEtf9-|&48iAX_%PhJay@U4pe^ROztYC+uPWnW^qb&buNGoZ0g{z_irenvef_pD>1 z7U5^8XrFW0%l`~0_QkTaLl3Z3I;5m>lM_rXUvYRtA80uSUJc<-*`(|Prdy*{8ZS>GX>IZEYy#3DW_{c%1x_FwD-Og~Y z5$E5FvoVo|Oqj}Py*D+5iQ#+iH*VL+n>nC+W;Vl_I~h7V_^#h}6JmIOlmD;kZEBgA zSLt*nD}}MnV_bo?<00Q34BgN9)6q;023}uaI;T>CW}j>AYjiI`wZL{G?yO?$d0i;! zeDX4k`%Hty+Z>{Mj%JK@;x zIrEi4MBYo*VOgoIi7$Y*$(81XhR~g-{P4`+rhl&gnDLfoVP=MUJBIgEKj!x1Whe3* zs3Jv!*Jp!pS~rLN6`Aq51J{`=yp=Y?#BH?X&vS&q*>LlaRBJV+Ghz{4sTNpob~AFT z^FQxf`2Hzq&wd*k4&E1koT8Tw83uuk4&Mf#^+NWiS1}B{3-o)4i>HDoW@#7BI#Hf9 zEQkWt(%;BQ?GN_9@9i}AiU(@;8p&iqYv|ZFx@+@FI&V1k^<>$${p5xAt;?mXZjE6d zlUJ-*!D$#}&M-N8TocA`C~eGtHpwtA;{{DZih-lpmZ@jLetH~A(=|7nyA%7r{dTkR zm(s_?`fOqr|E8-4u}f{v9ya`e>29wBsy~0m&WE3#G9PZhkb^eUpKR%IckI%ux_okH z2v+ZL6SFrDz=qcaS2n4|L%h}dL&qw)vCXO_JCT(W-W_aI5xNwBw3GTj>eC0%-KM`- zZvFbd^Y>wXxM4Z;?U>#r>`$-T#E#oc_I7;TP$HEt7gZ*SqfGKHhl0<*tV*&_>Z@$* zuN@}uVKc%mtk^U2=@7%Xt9p;AW*=jyPhnWcs&Ya8cU@-x*K_IrU;aYrtAvSwXAfbh zwnF&^`!}Gp-nAwjj$&xnw~3OgEl0l z@p2zAhjQ$B2|tm4n0@Bq%-Xgl_+`27md$hLsp+!yattZOb5Jg|1&R|{`|Sn&ijse3^>b>thvmE zEkGY;_$K#OHfuo_ui*PflK2hXtd@0O&XN_X{-oHNrtJg&TQw3Ca5nEo47RZM`CxK z`Pe{F1zPkfvDZF@lekWsp zs+Hb{-RaqZWeK4oUjx44z<$`H7fWq-qVFj@5Ny9s&7nX z#cs$a?;Sc?^9;u;w>{(aI0rS)5@Upg=NQI8yV)CwziP0uz>d7(r8J4@b}7q81364! zUUletH$A=q&pu4$jQ2TFmDZx`^jA0;K|^Bz^;U4 z*K}qGyJe_i{qrW)*Vyq;{AT#m+qiH`=kmz}3|7Q~z9#DW0`ETeyO- zBl>L5^oKDLi~eT|;SM6s9r2ROEBe%Wn6=?}<#6mIHg|HTk5L{&w{9`%kb^#Uc?vJN zyoZkW3bkMJs|8Ywp>bsM%aHR0KrSOc*WF3r>-5~07RTo|D#-&n@H-s{t5K= zw3r|Hly#{YyNoz{3qSr7znR3gbaLy5Z|?v7xf*MpgFoN&5~#I`v)5n6*xxd#nu_My-LWku4+M-JQnZTo-U^YzynRl*NK$6HXoz{w9aJ1n|l*e$R- zOiEHf-VU=q?+!j4xf46s1B34f`9Rget@u##ZW8yna$#-d~~do< z>$1~pNo=VX)$OklxJ*s;>jf)K8F;(Zn696Zw{a(MJUjZn@8UL;Ah8Kaw#h}kfa%xC zTYv9yhvw>o{`WHJ`A#qqyR}ug${7suIv~$H5ZW(G<<2dQ%+6pA0-tvaP|HF|E;j|62_GY_M5c-45_#dX1Hx*r()b^qV~3)3c> zua9^Wb{!?DCR5wXNsx9bFpHJV1!kmb0!|(j*z=REoU~ zmYW_PRl@n?H^b#Ih@Qzz(ccwz%*x~O#uD8P7C3*d&`q@w} zsiq-rrw%mTjp7;10?-(|T6dq`Dd>AMmU)P(2;CZkULDfrM@9J+pzeDH5BH9Z z5bvv#wPN=p91>j66Z>F-B?8)pByUE%x#1>D_wy_8|2pqj^HX7d_~ITQzYO^kH*E^j zI$zf0n!d+E+VZU2LIMYkQa<+4VuTg%hqc70zwu#se+$cwH7SOOm7hvX+Mab`PWj@L zSZX)aPel({eWJ%rVWNFl>UU-{_V%l*z7HLMfp_Bhs1gpfNdwhmeA0jIKBb5JpEz(p z!pSVF%oK-xe~pcL5_!LrA0d(>BidMM>?%66As$B-BUZ{L^FoQt=9QKvP57lsF;~W> z7fQNk_%83D<1=GR$MD$xSH+Ngn*cOoG;dBptb zH_eAQLzdK48D>JNj$5jA83_inEf);$r(scBL;4OMRiJRr32Zg6 z_O;N(Z?)%XloDj*Ka+mHw*+cD>N7r0QJ`&O?S_@vx={INS@m`sN6Zf{cw8m=1WGNO zN`s|7!k-*Rt&9EFFl*9y%Of9c@ReP1+ikxj)=@3`KIzCnJ)e!(-~kgHcDL>M!}bwo zUKmiElRwHdgoKVi=NztshHMO)$sUJRN|I z5(c=q->-6b0b@8PC_2rM$Nm4YKGZg}yJ_Jb9iN-TY&HB_pDd-uS#O+bNn^kV#mX=a z@thlNFn9c!brQ7&I`NU|pZm&KrSe+)$qZwjz+}V3>4-y*Fyj5dYz;q=XB>20Z2jT| zG3*Iz^5k2_3JsL6{LUgmL_Z!|WnNq@3I6jU-Ze=?TysG`D}CGO5$Mx@xI6Z$4w&ak6*og{?eq;DeWl;nF%#R_l zoJg%#x@d)y=XbR}h;o3;c(HDC{TC#bc)MVU(-aupa*)ZGg#kyFVcp2jY+IeO*27|Z z8(Zi!Bko1~X1i35-IwX@d#(3Lt<-LWuAm2*{>pC&`B%*Tz-sgV@xA*$-pe(a*{=7t z)5n7iR$n#CS?G8o*dyTng6CT-Iyc{z{+Y6Q9c9{SASytAD zSr{8XeR(?f7u1h>8B;?E{@*V%7D>Al1)$}PcFobVPS_e=%%+`K2%VDVIr~Rk8Rlc= z&JV>8AxW60SmRdO{|6FZvT}2rR71)^PF-q|1&&#c8I?=v{6F#RSG?VwhjB;J#X>qK zkmAU)qs*!m$)Er1nRa{v4Xgs%o8`^{b@L6T;&3@E(yg-J!o~T|br^U}eI~gPaO(Az zgDJIf^v@^Fs&sarC3uGVZ%Q3WkNb%2r_YUtd?oS+YwrlTSXi22*;gS?d8=j|TX8vH z;?fPMp}8;TS(H@cSSwUcTB?k+>lq)rR*6Aj-sq2}^3B*hz2u$ap4AObVFJ zG4PtHu^n}}*Ajl7{y&qG*|1iz0J^ zX_56>S1Pt(@nT$x#xpt}3FP;EYkJ8X3F%o6ba`j&vC-v?WlIZ#AA#PU@~r3naTA98 z>i=<{QIB0w_~NRJ6#pguPd_o(2Mp~4Z?~=9_JV=02WRho_b7NPM{hUw?VNhaPdd&h zkP9L$MpPO1U!IiByb&um!+T~cnkE_!?|}L0#i8K{9$5VKyGON@IDbY*trEYkR0ZPX|)b1y!9$uy=v zr_m{}NeT4ulF^zyMKE@GEq_Hb#1ru8g#_UP@lVq1ynNqh8T zPA~hvuGjz9^n>A{*XIlMufpKcRhhTk^08j5E%!&00S>%+%KhV+;`Cq?}rW?Yawe&xDc47+Kri_$f!sKUrchXnRFBvPN%Ni(x`;JGq=+fnM42kd1}lNdzuC+Ho}6Ly=K*MR!CyEHVSEMgQUrA zwRhI);qamRH{M3m?c4un9b5al(S6X&w(O)ow;mMlDJk7};~V@cUC%$3u@f4rzKS2w ze*^8p)=_M)&yrXQ3^X(Y34Fxzhf2FXETj9?abQ#s!FrrQ9u{{U|GpT-$1whE*)y`W zdp_Y}v#=#I7vbNU+`c%JyxjpJ#P|naIQ~aHTe*sf^;%W%hxSMax^|cJKi8nY z7kX}GZ9cZb942?|_T=O{gp;3J_V}uDW0vKOhunSsmVG4!@VPH1J&$gK0YjKCgGXF|PNA zu=p)`?Ad($s>w0}ce$#pGCOgC0@TEPK38>(|F?4Wf8MVhPh6HXum2AHmt3YyZdQ zi%-Swsf6)Wf>Zcv0cf%(;w4?zLaEVi?*jqBte9-a$mPWT*wFEuE@okAB z;!8@1`7snCE#A|`@p`k)z{m9Z0_?F#c>QH3UC+VHI^7(blu{hEE`C$iM&M%)%NIo4 zBGo~ay|{vom?m`bEa!W+o~Vm1=H=FNw0R5#I|j-+s_FRE`0c{8W9w?uAa&nhhUG50 zzYV*6j<<9?q30za?fA5fwEhG%wD6aOPMTrguWRKC%O3&RGsCwMPD4-k_Wob`6v+7b zx|6*BIyO18ges&EJaR=%Y01i$hA?akPs=7<1DxP8->Ptl;A!f(dycx8EQdWG^DE*% ze**H@*XBj2hJ@oAubXfwVk-+pSK?pQL-hF zi^R0e<1oj;b{G$ppsK#V2JyK!o-p0XBr&HhXl-{_g|aGg=;GRMIPR%@E%BEf!+X=2 zUqk6BS)iS7FPF(w$I+I;F_~I=`%-o_sZR|6Q{m~{JX9gXy!+z~IA?-5gpS>F1 zXL%V*A6rZResd3~Y)g6txGYHAn^iVFyFiyqm`lF0p}g0F#1wivzq?BVdm~;88~^md zPdQ$r4?dM)%f9FEctit6yRJx&xg5hTYc0<1{xu|4>Gz3|xAU=m?`o^p>EEzVFPpSa zDG&RHKBZ|WzQKXhKN>$TWsGZ2m~+cpXw?C2b9Uo~vjjiUSpG+?+++g3BmE)w@okpd zv8UB4C3cga8f*Kbtssp74BAY;Qv)6yb>2cxZ44{$h6%eC8{ zMU81a+1quBj~M^yJC~#iPe8`yQ`>BQD#1+1cw&|gqg^r;tzo?+Ka>7HWvRbM^B93s zTU%*ElBNX1@P}^B!{i3+;!3)g|HmCB!S)jl?J4@cxBm&aK6L*3Epb?YZ*kg9#0BIr z8d`zJQ=ekx>W^pKuCqe%-ik<*w{*T8__gcyc|l7xIn0M>tQFVaa5fr&V<&R?Rw`ez8Lno>&L4$(*M9d2lw(@AjT% zVti?HNgK1*yoN>3qdhI*0@$a{W^IR6II1pA@$Mt;k>LjEVdvWiplnIpHI77Gc(!GJ zL#oF?soF>|khzn{uO6{`@@oP2a(@J5Zhi~BI{s-Ttvd5t< z(U9#x+i9SMe788HKZB#|))`&$5yGM%3&l#$)0p*X%_)|TbUPSFy}CcOJSRiz><-rF zkz6=dz%*yxAdVH2KTI7i{J*)E+a2%m@)hAsn38-WubdjYJ@g#(P95O0s)auB!0%Okgq_v3B3>jW_ae4*eu$8KM#SmbKPwGfI%p2#G>cq` zPxLr3mMz(vcJDkBQa4cVe*MaewFx=53)o($v6r@0u0GucQ@%UGoxhy6=?MjQl zvPCC_%+;RIb9jvBL*m?wkfFG@aUL9B%SvIQ@juRzcKiy zKLdt_Zso4L_(hHL*^hp1cOrkjg;#leRn=yQ3A(rU_mT;u9*Xf=+aHXy^3&d&*BJ4_ znuKZ``(zBDcTL0xbyj*j9^V(mhr15HgYM}H9<0h$m9X?}FZvSaa{>Awg*Kk{A`70T_~IMk(necqCRXN2CaX}39xvho!4{Iqjd zXcj@5$JrdN+v!mM&D2=Rhp6wXf2rc!7);N<#J(SroiEvJq4b5ww4NLf&hl!9xP+}x zW705{RX_U<#xH&15}RpZ=zn)~&8%^YDz=4YeY~-sz^SNjsu30U?Z?Fkk@fZ->#_O7 zm)J$KYdAQNd_ckZ5LR3^vyeB4AhGr<ukGav9EwfmJ!;N%XV z^&MxK)gj(ml5|~d6Sq~ zvj0UK)mzQKPUbsyQY;^y(6+{2kFmYs3k>rK;kRxDu5o#U-*&}aYyKvSl`7ufH=SCB zt$m@QAG-)SDCR96R6M$o;rEj}#5!(%vck>-`@ZFuCtyOngYBOZdR{3sB*d#2p0>c6 zKUKN&a?!;4wLj_zPwbhGp9Sis?zy?^n=`CKRYRa^WDYmfo(d{* z5G3?@@?fI7N+>}2+jlFMo9trv-FVCL0N#xZdhtK^QBU^mZQspu8J^R|+t3y2!NSnq zPw>wU4VO-udjyLxdi7DrS z>7ncNyeMqUF#m2>%b*uSZns&%29t7nJIchVReQ!0VW7^z&8mpd|9clc)jCX00QsHm zz60(IaR#7xmYNl^al(w&OsCteL>#zyw(U?VgMJ1TXX6T|NMEsBzIRJ4jb48bjv)@y z?~90iPYY3#UkIevCBQ_7oYe7?JV?2!$uwe<&9FWd7H72L2hQLh*N$(0BKevSJ)gtmtoTxeN2585fp+ybWn1{?%31>I~0s zUKDwCfWs5JXVio4kO{wE`K!1m%No`~&&?aJDPmUmcvXaA>ktV`6YHn#PS{~@;mdbN z&(Y-)+bf;~TG*MXF$tRN?pa}j6yKPnG@e2nOHHDoi35y%-Rq_Fp&u4h)x#XrYBBp} zXbSd;L$fpwZ-_`2)(TvS30z8#r$NWasDp>qVQ4l9me{m64#wUqkhEhQp?8zULiHrU ze=+njXmu%-2gb>HR)*7#&=t301L``!^8yS~VJ9_C86k0kTw-*4n{f?7K?!w>Y&NI#P9g z3Ye}DdgNmdAzQI$(7?RwmVE)eJ`uWma{awdSwW88Im_4fBS@=wIbf#d3+)#2I_Iz2 zsIdv(t~Ga|LT!d*+51=`e`~&C&q$3FT`y2$da%)iR)jYRRRJ zodRmlooCm8INvK;@_3`xz6MIqTPqoI1ysn#-B_z=0fQ-OS2z_2e2!7Bi*bq~TS?5h z%VlB(Pe21_>#q|2beLs2KWx|1Nn+a5cX77iGEU1poH%Jm+*f1up6!~&OJUqg_@|`y z2!^fr(ZA`q4~}hH6qD)q#omvf-bzM|V%cEeS^m?@NgP!pv(qA5v3o(@=2D;oiB)ZN zn5)eTXl<(8De;rYuTRtH-!N3Ai4@smOTBJC#L;q=Ru|<09JiiX|LfjoXn)qU%H`oH z7@poCy`!)W7TRCMUkY>uvPY~`Ubq0XgYW|u9U~Z34ZVAKOEXeGW%a2j(f#vktlqH8 zM_vHlczMkpI>QSS=7AHh)_;dt4dXR&wbj@^bHF*$x&x*y9C%XdWwGK~VoLhmZfH=c z-y5~^1o-v|tHSZUn3thgE6hFc*It=AsCG0OvzKMtU%%dm-ChQtt_CIHcw}Oien2o# z{foi`Hrv9W?VgN89R(Qp${D0CQ3zFrc3)OG6=3b2n<}Rx)MMyQD;nAJK^4pI9wXgk$n$pRd|yF=e?BQ-*?h^)12~fzT*>-|As+&$O3!o- z@%|f^n0hW-f2lQ*w@>~kbX=>BEO+B|6KBv4~w_a?kIZhV&_(jIi5pIfkJ^3&T7^*+%*az_(d;J8XaQYg?Y9Uap82Z6f3U-iytGt4=9ukQ0Y zhTY!nJbEkWeqNXmxEE04A%QKn>pIhHA3=YSg~oH^wK#f4M)2rCH>mAfSy|jd=>OHm zZ8aTvt4Pe=m-gzt8pWQHs>I{c2_zP|w_fWhhhWHgbXdtb9I7YJY7;j&n7`YCMzV(_q5td9c+0>iY`@d5c4g)r{^4pL zTXKqyqkwI-`ty@_9x%*5`2mL$Ye5D5|Bgy?UFRJ!YFsyh*reXSz@AYKN#AB6H5TFG zS(ZLW?3l4h{yw+^D-RSu-<>TAlT_F1XJ*bo;}^~AAKnsvwy8VAnR(ZBad@X`$>swL zbzeBpJW_FX+Z%?@%NAypI&*dsB z=;x!eegFEN#B!dgrBx$<#G-rFp^uXZhV_m8ZxyA%qHLDe12wvwt8wnkl{>*%t;Uso z@>S#Waa`cFk=b0)pvF0+`)u{{X(*g(hiR{$*lm65+GklJ@1)wotE?A-TM0V;0S=5+}W_vmDe&A4;H4E=wa zpw*h0DgbKXfy~|8YM@JH7fWzc6;89Tu6Zl-oW#D2X>9v6Jw61bl;uO5e11?mV}GT# zJRd6GoOAgdc86iUCM5%E_9?t(_8aII+M1Tp@W%$q~+p1DFlrE}y_&lwVTf{SR>mof~9f1~Af z{58^Ux_|Z6BX|gMAKYCMee17X`a3tnM7jVd`$(#r!iajGv7r+WX4khu!?mFiKZo@& zry{dFTdN31`8)Pq*CX(l8kbh5j4rzf?N9DUF7XV7-a(hKRB3|ewo6AvWL8uciXvSm z`7?ZB;>zb;tFP2z&pHaXR5-z()m2cic8BLFm`!69emph>;}`b+xIgq7zFP0)f6o02 z>)6&vcgPTWal?g@ckwHdVLot2NA_7e%ynu$RjyHueP%v6K6dNz&nxcNypQ7`W?gXR zK^}9QT4S%sNOjN2-V(_mtL?&RRP7l_8Y z>|Zy2hlv-r`9*~h`#j5w1+wj6xa)1ImJuR7-N3um}@=8ubq6AKEQ?)J-*!zx^%lkjg{E}cjoNEDhg+k_W`;c zgmn_`KjF_DbS@iv)*3_L3-?^PCn(e`2W?)yeb=|CVOjSv$*;i#?$YeS_4qH^+hLMR z@%q_#!tXmZ$vijE>xi_$Nb8jw>3JJ!Jgw1<;q6=hdtYGlU6$*LZwVYSiqP~H!SHOT z8f}VhTWWxv7XyuENc4Ii=&p0U^h1Tf|C`*Yr)YAc04C!#1jP&AV3(-O>92@p*aY>y2EX9cfc87J{VuG>b82{m$ z+|DHShz(UuZ$qJPS74(s>fnHQhvO+W1%~mA*YUCT#dNE&p2*q8mPxPYg}Jf8HP!Mr zp)@9jXMn9+jrsE$l7-(Lm{wgilPSNG#KfD<6s}edITzCo@s@7ID12_PRV0nXQ8M`D zrr|yq@;vhE_1ksW7_uUu{pTHs%TWqHx|h(Wr~8^q3-T|Mm@3CyZ=@ZDA@dPWp4^K- zdnRY8L80e+l9;)6R`fhggpmr(gIk#&L1CBlO13C5jB$@1-1k-h2JR;Kl$g+Q)^X4~ zyJb(ZE&MsMxm{uvAJ78K-j{7=#22d7>RpY+#J%0rA&13t8t5y!$GY%L-G8@NV|(`M zje8zlU&6+wVJuD@lX1mh8%o{jY_&b3*vm( z_ti;?E$B1`yzV%baXu8gK8(qzy8Sbc8J`FY%{%2jg{lp{O)J-5gv1BVH(M2|kz!aQ z-iXX-CEK!MJYkLWK;g=aw z<;=rzE?!oJrXJ$CarNJx-0#M=)4Y%GaSW<4OK@EZuzv>=y*a7hyUE1(su!b+) zoI1u8k-dSTULIqav;K-O!NWA2Xf-P>Bc#UBb?;9CM+;<74rZ;`P54=AIFk1Y`?_HL z$q#Z3pE+TS)8^i`?F_h*(7AE2)j5kHFB$Wk-@%Tz^!PY-+%_4F@TBvSK}8+ww9a8^ zqCXOiw)IbM0`<9@*q4utxVD{Ryz0(Y8wvUIce_&1ychc)-6~f${;9^+)6k(WdJ@Yf zA1M7PTFvnLS=Y*YQl}Vs*4syXZ)usZtFb$kUfcO$6ni5KRv1{;KuK4e_N^zm(BrER zb&n=Zz4%$&tH#c4Xh}WWhKc318rfctuykyjT3u5S)Cszt+QA)%RQCfGd@(h| z`g8669yu3>?e;P!x=ht!)ZgoOU(h?~%&$>g#(fB8A3ol=3dz{=u>2GMjbBJD$Wd5b zK=6uCkFfd*vP57yn>%l>^b>}0d|^(rZDgo} z!0{s9JG{!z)VOTzIOe&TvG0=a<*(a}uu;d}(*X(m(@sx^db>h77?_=`^B>Q^0qsBy z$q!M`6O}-loUX&MA30wSNw2|KzXqnW84|eADRGE>_&CnwnCu?rZ2{yLKUb|Bfqfng zr621EoS6LH-~*?!>3TPi@*k90Q7_}zFLTGqZ#K}sJ6l^>R2yj}c6`U#c(87T`U0OU z-420_p{6V&$QgAvvlK@I^J|J}iZJQ0c0l(%!QZv8(D6Qwm$80) zrr1);_p}v`ej2w{-NXhxzjkcRXYGQzVxGNqB{Zy1dun4ITFxN%gY0D02g@@KxAs}m+s`RUKb{dzg|Qc@W^JZ#|GOXb-%+x4g`}tGdA!)LtYt}+ zoF`Up$!j!BU#G?j-FtpKq{|bG8`fS?ZWMxrs>KXysUl2O8IY(1pY#9gdkp9yAG~XL z1bP;4?W`Fha28u)jAfoaAoRWBH_o2sfiwT=AF+SxuIIPp6FjvOO6zh|{QjNCKV*Nm zo$2-`Nc%9N@L055jhR8i^3&^@h&sYgp3jy=YQNQt(Pp=*pN;jwsJ_>-YOP(^nSYG^ zY}YU7JrLc`nKTC9#40;pu=hae$TQE;7W(;wq=FUCIA!FpW8UGc&yVde5&oLgJjH_@ ztsy-;r){8j+3jQOKj=7e;BeACA@EfhgB(_PSR`vb{6=Dq{3{=I14=Iif4xYxZ7OTUJSSgxcGAK+R;WC7?PIl96XtbvsaEpe;%)-zrRms zsT_Q`1HBgn)n}{-9*HU19p}xftLg1%r&jso*dE0CIR(QWnLrp4AKJnoHrdSbhxgL`&uZ))C(1}W zEZ|qgu>DLP;m;f4_`_?VL+JUH_g$vf{kn;xJNt)Cb?AN*HLjTx27}}2jQeyx?2&ba zAS}wsefXS7uP22`--E5nk9Hw#daAwY`akiSNKA+Ktfyd(QyP%=rq;IOjs=lCM#MlM4V`Z#)H1^w=XC9J@P#LTOW5>faoZFveU@xES{hUl@$So6y$qlx{?ptcE&l_Z*Y{227j!X#9f>-ZT zI)4pESHyOxvUJ0z$DWh>+qy7^tTHps;))~h=Nj}b7&G)^QaIDtc}p4o@$s9}E@;5^ z`VVZ|qHbbc5c`#%W1%onzDhp2ks&V{>Oykw-0|kY{`6gL$(wgU4oUZy%h?eUyAIbJ z`)RrzfRt!cliT(G&a0kGO}H9c`&W<2ex0PMX$iIN8w)LV{j=^EX&fIjIiraDM%q3F z_J9W&Fw}puVaW-)UmZJE`BwR{{UNb8c_-92Uc@H5lK1isA8=T2)VoQHsIMDdQB$IN ziNG_b_UNRXm;JXNchZ&PzTRbeJ`*I9mp#f1GsBjtpUux@HvhBV9aBP*r$5BOh*o#{ zW0pVA9a!Lg=o6vWhKm0%H0or;8OpACOxeDRQGaY-ovGv#BKFUI(k4ARt`{@JFR$hYWLJ+p>~ZN;2O^=+ucmwZ%|4`ho%4F65`*O# zr-$t{N?JwFKbJ7ljaWaRw!_({yN9^~ia z&;K<)*cU4zeDzum4k@g+KGNWa6BCZVXoMYwJ@4O_Tjc)-ZoDEFOJf<`KMHNGCDggm zA)Ifxp=PT)tj5j+Rfgf#gxtpMGMmWU1hk>jsIaV0P|V-ELFVZd7<}~n=Ly<1NR?fW zp@IuY+0A<{U=0s6&uYJQJ<^O5OS~g~SPfy5bW?q4fyiIIlI?1{E*lK0Z`<53NsmWD z$J#IHUJiXwB<}CF_v3vM^9D3DNE%XOk27H9|8*DJZoe&aSSE($N?hKoBO_2L6ir#o zF~(8-0gW))7K}T-{kPKVRO}Bi3GLq_kF;=dg6Y^+m@16-ACFIBkf#~)^4snXH=#Gu z@2bGl+mJ`mc4D_u#mctfzKy3C>h1q2m+hD9msT6jLaE=y?;l&opqnN!I8j01HDT(~`x-E>vQyHK>7VuOYOJ+4rgXd&VRri&mY;6Lkh8I@r%T)uXjTNI%hYzICWUjiuuDtDdEtkm4+)tGMl=8gr9d^PS)#97^7@l0#|4 zZAiA*@^yZDDz?UJ9nHUW588A0MRJ}Mfl?c|Z!1PWe1LAw5zp&z ztf><^^9GAI_^x{Xvj*CORr|F()1lLsW6i~>030vP0$aDut^84k!cnk+aHJY0T@@quyTFVK~FmreA~+gR~9 zjCM-y4b*%%u~PG%EwpG#yx(wOGj^=ZJo#165Br1Xfqkn8hF1%{x?wE|6uXfHUJDb< z8Rpn2I8Wqj=Vf0vnl4X-o;8}y*EGHUdtLC=FQqcc8amzvOl*|lTO9R=&iO^-Istn9 z0(7qY@QE*(O^s=lva9M1f2eqFHe0!d@ShFt7ULT_Nca1|q-47FjbEwQK9mrzahcAi zpvJT}NaR9|1(FYKy!u?_7z`vnkd64T9;o^HpGmLVAwib2-=;wiDXU*gro9<}E)V{W zFHC!ps$=}*`A8|WTW$aLlD`xv-rUJ=rjO#M)Y8hJ%wI@retxD>v;*F-{m?2nA_>D+ zC&C(23H`mtH}9Hwj60NC&0Y~LdV?LOhF7=8nZhEEr^LiNITDjVgUs(A!!Y(rF2<|e zAFEuf4|)~NL-*!|SE4KxIJI}*Gxr}E7+UbfSmq0nf0`>Ae?{9t+Eh9uYbd#_`?eu0&1~PR&pSEO$jtEEIGfq zOA;q|f7~v=Laz^2V_R1ea_fsJ)EYH$l&g5Faf!&VSM1mizbA||g*GtcvBR|IrUx4j z{eX`0`Kz|JU!bDm@}(g+q8?-1^G$}eIK3_w(w6_8{%}nITYerpw$z6gr~(U_)C78d zGER9Ezh@a8#+H$-`!mEe=)bF7T9qjK!2&ax;_{P^`$FI6P#uq_g#WhMUA6A(%mWf@ z@}k_>2|AxI7G5}2_S9JnvyOYSEUu^X#Q=40fL!ubB0q0YU*!0O%XE1M8js#3-a1eE zys&IZmfQB_5iaB@Jg#5i0P|h-K@4|ee{sqb%D4}TJ0UwYzD0E@p5MP9b0-(w^u%9OOZ&JwIYTc@GPu@;Aye(IP# zN!0(8ydg`v>qJ09U2QItRVFM(<$lk7u?xs~=A&~FB!`MuB+Y)OQHKt(G%crIG z5a-Blm&x9i<5(l*J#1j}ProB}zFf6ILhB{{c>~6)5)_gea9m=ZDtU@tPXX=BTx;8{ z>G2)tm|1JJ|9Udc9CYb@=CfOkCH(eY=JPKhb<@fl5_kama=`=SZoL% z><;Y^@w32s-GSML!xqr?F1Ahg!w$^7AK6C}bQdn(7-h zf%f}rj)uArdHD^JC2Siv_~A&_j+d_b^nD6_gH9Q-w)`+BSwNMQVS{l#S=OaWuFy{# zwtlck$TQ`Pd`sC7LmV3GC43TA{;;C|Put)xT`*72TYxb3-D|vsxp2(w>Gtt#g8w#0 zl~~Qif`5PHaNIk z??DFdRUB_y_TYw{m>TQg$mM6WaHyXV=pCW50WD6gvvPkOcBxu_^B5vsS; z1+xzeUg}yAc`!BYpX;x6S^~K)z{enk3npKfY!swOVz1Sfo57QGTyB_=uo5&|tBOOe z$4(3%y@^Gkg^$c?4-n7ES$BU+Q{dlw@p(X$%hkVfw?E*a{6pwCc!z2_;R1t`TKc)0 z2)|;?L3>n`sCrVknG#qIqL5I0!4520O?aJzAIcVP`b!OGDACLv!B@Jx23iGMz zUO!kev_)>%)!q|a;Gi=w2Fk_2E_p+!% zSn+jDT8ZLu7$Z%`f0}CfyRWAsBb{$y7jtS)id`bibU6;5Kl2!7oVhb6?a#vO&V5zq=Q`}QA`R%7SUIr_e5ml{*!CZV|ipuCl2l6_6^KsSsse~5~GhIN*FPc;YIal~fRR_Vu?ICA}pPZIZUEK{l% zSJYDi+R`U?m)=Up1<5}3;U`I$a+NzV>0JZP`rMmY+pC8;%^cn4kz{%zm>p2Mq95b9sl}LDzR)r<~Je^wF z)zF^R*;RKp7@9-hACI#p_^-133$6?NJ_+Q{R8uBiLJuoBr@xo4EDQ#v8~r0h#c?`d z=NG$^0+|1FyiMZkFX-IG|J<)_QxbR zC?sWZs|QP=wol3At5h*m`DDGh+DwlNL;n87bWxF)STUtoVjh(Rsq)+|E7A+0vs!4| zi_#3@xeeFTkC=Q13aecB1{M1FuxD+q@{PnO7%i7op+!$X&AaXSXP*zkBEu6F91o6^+7leU-#> z!6o1*w<``0Jqq4hZmq_?EGkA{do3m_`R+Y_LjV$PvzcC6P3-T`c3+EIONl&;s$KbA z_^|~?1xi^Xujw+JhXXmjkp)T`(6Vf)vExpL`cA?gFC3Z+F{b;?NbGv)Q6^pk*pbr5 z`6-8}7i)W^-uzVf3_N{Uap-Or6E?4}l{AwR#h&sdPk!ff!rV2RmoY7#IJWTy%lH~1 zf27V>IQ&jW8jJ-$k=B*_Ctfsz+;qGNa8h&`WcYlfYyT$0n{<3TobJovPmLh(x+q#K zsabCWf#P8`rvBLzXT?}X4oE6t2k-mZo%cf_lf9xX?(h~E{^9&+)Rc~kj%8;KzP!`J zhn=go$-kUufl7A6M6Vy|5N4dQE0X&e3>?nq+;S%p$nP5{C$=nL$!K;CbF?>-BFI(t zMem_f^sSbMvM|iFSof1{8E{7+>*$voq;(AW9)w}X-Go--BeB+nc!Y)(> zzQq>4Y8_B7oGB*9=tJ3hQLS$hN06L&{)f(F2ju5Vi@q^>0i9|(dSa&u`tpWwS3cdkBnAov)zEK0SV$`*o}x-T0qZzSS`_3jUH z6Z9`bY{sR}0y6aYGxXI+gqhdWK;}Lx$-D$FY%64o34ao=#+1GKg`n$^zk1I6C+Ta7 zY8({HFr9=GIKE1@PTXk@3+DPtFMk(+Dc1UT5tZk#`z$;21NJ>QH?(nRynw)iUpUh_ z_x?x*j(*SDd?ShI&(M|AdXrJtq2(JiTQzrK)h-i>n2bb7mp>S-XHo)vzol$f?9>M9 z&$|Oc*Z4uV?Fo_a0yfMqFR$)!lf(*fXa6_qw~%V(a-!T`4XZnO66*!xBN>!AoHsJfJE|6Hzvuybs8T_jCD!Y1alhXS{~@bkoc z_ON0+ROjyPVOijU;U`VKzoUL3rCV)YpoMNv!_=%}+Y;#AzM)!dng zL;8O_ZQ&CcgOoqP%iKolV8}wC;!(^LOutxCcEv3Q@o~`(CVhuXOUUnW;`H3_QtKl?wLVDEQSF6aosyzg9fFZ| zh}mId+jAT~&m1i+F9tolkwfPmPh+CzL9>f%0-$<#8PlUya3!w$-*|gbemkP zfV6|k@7?Syu}th->B*=896RdjX~x@tb;=x+FA)tm@!V$>&)2Ki^}%wYV3skTtHX#X zNhH#O-ShXbt>Vq;K3Wp=s(&jf2D+ad6K9TB=^u*0+~ODPHU`2lw0_QS zML7*e%xcd?UnTrG;{dk2EthcQ(dX?SQUstNbhP3_2j2!{TYy3Fn|X&_6kkx$zASHFpr&@~lUiXASG zTa=f5M%u5WT!~Z(EIufEV%N|%nAlS<^z!-!j6G!$8&7?WO)jdvf)!%e8zA7tJm?D* zOQ+ALKQY3JMgEW9_2l8>MJbk$Gio7_PmLm3!NdRYtuHbX+=v!-iEb z=C@hA=Nyp_-Y}%MKZG*&uYTZQ?3HZ^VGMR2R{q$saq-y~C=uOAF5UeYhF-=*_wkgF znAd-}e63iPc>kSq-L3OF&~h^OYjIFK))^S=iq0YMsylaFxp_jZ1SdIM<~};PZL8pk_kPdlF`=zaBh)6b+0_@iUcQ=j{vc(6cS9r*Jx zX99`EY1y(e<;^g}Bz*VC4!Znd?!(4+&G|w2apjqXYOxnMk9;iFkLdRUjK8>=F?-Vr zyS8l~^km7yk+soxPv@s&_g+h5@*g73KIXaYceVpP9!+9dm2m&o_iT*UmUeIEyvoq8 z&J&t(>|)9|DSPqo^?L`ffiEh2qMU(G9=h3#gP$F9fial|ZO?a4k($jN;LS(WNma5o zmMm@=hDO_nM>`bX!3h7}f}!kCY*QATP;?x}(v7PJ*KagHWWE>DT=EIJ|B7!cn-C8D z@*Vb&rsI~u@E(Q4p8eZk{E5yBr9Y}bUWjHp04z{UC}h(-+gchG#RUY=Vl406^*Elz&B04?{q70kDW zsIds|&9)e&<3!-Z=f{R}?;b$?y;#!FJpzZT=*i@va~aAIe|Uc}UZ7 zZPi$IMi*Y}$OD@Bx`D9k#QD~}+VRpr-e;UzcWCVmchmpwpBl@H0`H=QoBx#GML(NZ zxmt$2XP7yl=|+o4!|n%`Ye%%`_83&XOSNFn8iFq>+pp9LKE(<%Z@J*6S*-89RVYwu zflcz_nws^wP$Q*kle3x!#@PF*iSudDaYbtQ{+>sWmS53G+H(Yw_2%wAJ3`>Tv>hJ4 z!JNzs9d$|dbzW6aSN_O3B8cFZpSgDMj=9qUGz@J1)amdY$(!;O(j|EDOWZ9cJF7aZ zQ@3(iZN&rA=dO0Sr8`5y5>81*kUPVVUgKP0bcv=!(|!BUpc@yK)B z*yyZ~Og2h^g$04SEqm$fi<7(G=2X*SA zm-Fq0D)z5jCz>sxugvaxN&E@uDvq^}QGExCf}$Tz8eBzEsfM?U)G{aeu!5qcjl5ov7=(yZEWY8<}uXUr(+w_8z zJ&mOZdLI8MdaQc@%D(sp3{(?1M$@JIMZTx9Gsn?B;6E4S%<%g$ zhiX<14i+Fif5KbSD~O|8t#& ztuolKCHqtI+dB;T!q9$yhTBpkhT(UEyA9>eic~YSpDke98oVxqq5bGO-;NKKMQUtm zxr0f}1YU8XQCXJTTN!NLv-e2M!xo&_`TgrVHv$i&Ra9bKYULNC_;o(ooJGWEd$ibi zyc}$S_AX{WSZmV+)lkBnX`QCFpXTzcXgn3Km6=`T-ULt=ZJpgNIy?yJV55k~d zP)gIGySOkq862|pCd_gu-nQC8k0U|h7k>k7pVcHD$NV3MEUK{n)#l56H%ws4RLs8B z&=Llf4~hnS=)uBczf2#mWQD<O`Yt zaH2o9?~I}k&brhedVG<<_vl*)q1>}M@c&@#zr(4H|M+p(-btm5>JlvQ5GD667j z6b(w7kWf@oq$omV?^Kc^**uS#z4tthz0a`c-S?0Cyg%2+^|`+H{kwjDbse4a9(<|81MNvAyS80> zYBhS$$|}Z%V{nft9k1jB3a^AhKG5=Ph%BGG&tD(o$F7HQW=~FHxT?;=tSr&`vv|hj zTB+grA($tg&7x|N&|2Xshxm7xKRNWap-f)O7nV6NQ&@dcA>b7nwA^>+jnY4VF|6dfrV|afjLOtB8GO6)(bSug@$@zxpsFcX z3yNd=vBZB_pW(|T2tOm#_5K0YKe^D^=)PjuzuyUPS_lqa-7p?=+*v=**49HG{wPn8 z*exKPAU%+NUc2FY1IzJ(o3purY{_|aQ-cH=F@E>7(bx}FT@-!ww;Q1m?=iNgwHS^o zwP!`W{39OCi;1sA2W$tbg2MZ&f81c4m#T18`#un!+Fqc)oCrg{R6~K5C1?&+e_>Vr z85@7smmRNX{@|JDOWwC!NJVuL-##vJGyv%>{^cL@FEEyKNX3pD)3fe;PY+KyDzI^X zcBP}?qZ$0%EpmJ7=)bt#`*`f5A|q5~)hYU55gM*GL?*i(vFNvX}a)@HxU7WMH zgl9~guKA%{4a7*c(Os|PPPCH4)nO4|wu;{$Kl?!_e#M zX0>-qsPbmGbMAIbKWY4&1{VlQ!d!#em%nGQy7Ag`-sc^2buh#q=JU*shVw%4+4bM2 zs}s@6%SY#Ut1PkcF7wD~jpc$Vx9guD3$_3?cv(H=k2p+csb4zB=L+GX*@N9jaWG@L zKal*}6=q8{865^`x~4FEV0U>i6Ky`A@!4&~oUlrik(l^B_X(yGQ0?tFvYC$wX07tI zrc?i-1(xa0FK&y$%F5};EKq||<2dbe&1-0CeeDh(`&MY@>dw5lwHQcwG75U38}ykl zKRRaNOPJ%U#Ii5>z56K>7e0xyK zcQpi0UsA`;+50c9lqhU;cg1+XeFUAYI%=T^Z)}$%b>abPIm7s4HQ)&J-!K>IIhTf} zne|mC4igYrc;FpZ(HEGvV<~)f>k5wU(mVR5OaJ2U*~wcc26mvj_l|q%t4~8U=aakl zsiNpBTiuIGRli|0tH!*;@&t~KPUNa|I1PsZJ#T8=Sw?8}9jL3>TXBok2qvlpN-c$= zQ0MK)Tl+*v=oin9K&tQ>O6zjvlNzD*r9ng=tE%rev3!3Y|8UXgwLZvsYj?v*>>_IG z&ih#EN5cuAx4peRo?8hp^-0Y4%P|FLmu_c|wseG<4>uYfR%k;tt)ewfAR@EoV<{^w(FSZ?Wr%yoQCM^iUlBpC4&_+2O9M z37WoDZd2mhg@zuDFs;&Eg|@B^HX|pYfA*BiIM;asrkt_^`RCJd4DV&KqhU{N(uT&9m{xU0`IYe=hytDD8XQPo*`obugXh+I?64EOM5kDJrQs zYx6&U_v$$8_$H19FJ4WSeBqB4x!hj~*PH)4@4wl9yUdJMB7HWG-5-IRnpCaM_Z#y2 zFxsNi$9+8?n)!Z_WoAZ!9ID~w+IAAPUY`E(Qh7t4R5U%Iytg@#8^_=xy{l6DE(}_# zo=Dtx4rV?M{0<_zq5kEHZ6*nHh{*lJFY}%b>N|Jf$1xRSH1_%H0Zu_o$Dd@^eDU?o zr>G}GTtc&8gP(@q6sKX$`1}nT;oX{#F8&+;$*J+*HWZ|a$ z4bMBnaqHoW?K`pOkMBJb7DvN@LTAXPSer}6h?q+W>(x1p=7v>%mY2RoE#u#u&Suu4 zj=~6@YN?|zUA)6QiC+y07#JJI3l(6NSn*w{2*dBR9dJ{yw#tRc8r)OIwJcSk`_6=1+&H>GqnS@fH`gJ4bn7UdWSsc>=@1 zOu1=KW2D7hwh+ z4*a=o2O7``tMtw6MK!61nB??s(%!3hr^n=wc^Xin+{y@GnGgE?>0YUbnDycyc?-sq7-tyb_*+4%14GXfom0Wg8~y*Dd;WYR zW)B!{F*ZCRiq#(?SL`1IZ)U^Mv)-%L5bZ<@JIo#lA7aGQpUAA@aQXzKS1Zml`+uPZ zxm|n>o3Osb6I7)`_7RwV;!M||=8v;V8}`}!ZSlhWSqGuPBS|V@8>Y9?Z&yZ9K1tKn zMJw-RuHMN{0WxRvob8Aq%w1)$4JyF$QzV_K_{|ZvkS0~YPoKR8IgKaY{7Kt^`UmD_ zR<`uv>3@%mopQ~?F?cL^Z(^pM>o9omy>)XY9iHw<^O@&+qtSdouttxT4dn0_?HJre zf}R7u-Y*wQVfNvGE~f#e`&*r>Y|`OtfJUFsraBX6(DL~&LB_|i^J4k^BfDaCEZ_3? z(bpv(=RhF3pSF6)?FtnlGJA@2vHZ^HD;$&9@JZ;{F0c0cQztCf#!FS7`saT!AB{a- zx(CPkQ-1fi?FO*oJoM<|=y^01B6o0KrVx&qrTq(o_XfN5B0dG9d{Wm$2~fF>GvX{;&dZ#=wahh{%!l`<4%{?1qRibvhw?Cd(agRdNNzk51u{9rD;~w_K0__Fd9~A+J^{<&?N3`XuzJf( z4F24<)h9TH*pf_ZM~qK4cTU&wqRm$zcy?XWE&c@)A6%*ekI+NK5@YC5=V};L9N>&% z5rpRUTP;(<@@TSkXTwA0gJ|T{+Aq~L3N*dFc}dl93yi*hU-$4RJtE)pu9u!}pk3cz zKlpDJrniFnhz6YR@5kaeU2aJK+;6yHJX@qJ#KPO+V8|=L?rRf@^l02Qtr zQax-<4_}T4w-9SE{rmE*&Ki8%G-3Lran$x3nC?usBG=N#l|VFf^ub?+d|DnEH6_Yu z6fImu<365y`d2z&;(@JCr4`mco5;S0nYG6i^}mX%f4jN{pl9_>%gqrHomnCz75<{d ziGIzs5m!jgJFELvMf1Pk0}b3bIyG?-PL<%#Yf4Dh%u58KS2a zRjy+F0IK=pyu~%JeiRjy8Q?rOhNqXNKThb(LsLb3WX__4Xtc#|!R$1q^E`2+VB{<> zD-@X(&7CRdg(Ug2Te}(0W9wj6XYp(U#m70ZCpDRA9Ms9`*bnx3G`ikKg|63JF4V$vd%b+=~T^_s}`(+}{Y)+0{GHVflnrb)u{ZlM60 znwJo;+Ib!IZzc_pd{`0$MzyF-U@CrSP-+SZu+kqHGpD!Au1J(6oO(fozXzWY=wz*U9phZ7zUWN4o z3?`o6^rB1{Ey-`5zn^#*>RAud-IC%*6G}qk8eAi2h+pQ(?UHDS+7%jY%y=EfU#27m z9lwjM%e`lgMb5;*Y{P?!Q$fl9_It={ZJYn#-|X(&_QDvemsh-V8VmZU4uvsm)617t zVJ>Z?Se{P`Eo-$%B;MVq@4U`vesAk+CQN&s_WLD);WqmNbW0odxWICO#NzD&Oh>FM z;_!FJmKZdAEAXZK^a&XF#s8u`+ZnBwo?!GJQANL#)vkZ)K7oys`bE)*i8rW+Nvo!H z1mo%S2+Hc|HWi>nViBFwfq(XW;*?^~1g-9XC{Co1g)Y?%&s!?ozjNm1|LhkJcw^2? z<9GSkzLz++NjX&;$DaLu_ORN2qO>p?#jdWPjN-6T`315}xwLaZvu}ecF{B za=yL6`o?~jypv&c@g(ZfZcxAOhv8;B893L=31@&ZsmANw?DW5`hni0$r%oEsc;JZK zWqk}iF-MJbqm3%gz{1yZc^1waw*9939h} zjvWTtX#I=ERUe%~Aivb({M^P0?Zu0~O;!;aa#|P6c_WENoo_K*Nyp+7O1ErE*ytu` zr!uDPQ=5e8^}p?{oSHE9oh~nZPZuQH95#ETgRRR>HsLStLb=dX*PR8ZJl);~YuwVlJWx$DsONMqE0dnxqVFqt~_WBpZHUvn$)zo*r$piW>+$8YEW zs@)uPsP)@z=#(&HRMk&Ga}FxNvRV>JEXb-FH?rOVZQnL*}#&4m|=;5Ht9bnc-Oo=HA5vZP-N$6y&GQN`5>9SgY9{Yv>z_Sqqi8S@S z;VuHC0LjDBrP`>^C;80pkQr$A{pD8Tx&x60)qP5F_R!)_5>I ziW%-ty{D%{q4JQ5iMi-)RPK@=^EYoZqDn=7wBXK$5AR-tY$nj`4x%O>xU$mD2V^S4 z^4bjtG!*&bSau*bzxc`adloAPVbsg-P<%rHDm8e!eO)&w9OLFc!Z#??yp|^I^{ucUedAq)1I0Yi^#-_Dtui;r7O+zB% zT3!Y+$~DEZrT^Ip5647}u5g&~;ZN3(=fKfjQ0ALB#tJ_cGEO&#|AIEZ z31U63g?&c5rF@I$E>LdJac}zcY zmY1WNd%6gbAI2*X&b@>x#!G?RqFrc#)423z;zKkOVl{E|YX_o6$tI>^{TmQIE)x~8UdNU%imJH8U_-y zo%>aX5rx$ICw76>9~g~jnSGwPnGJJqlx`0{RY#*b^{+dUcSHUC6LD49NvJco(KX*2 z%LBK6kh;|^qYnMJalXUsgJ}MgdE>dkqd*F>zWr6k3ANiFD)4ZSK&|tmV!;*hXu-%@ zyyomRSoCv0Z@O5DO5aX9aVr?3PmS^d@ASD~^i=xvgPmBsRo+wRY}TIy(?u-DcYF?n z^>fByxaaApIA@Pg@4XTrFHi72f3pM=i2)g_>q~fg1BcIMk{Hf-IpNiapnxnC1XsTB z4f+CW_Udm>-bp}nCu_b~`>Dd0qT481eFn;7PcB`)Nx|&l)AbW#J|1XkH|L}rZlg|2 z-0uU7{2J$>)A3vJX3r$ls=BRY=CnLoG>-YaL+uP^r}X9nUkG9O+2L24y(ab*LQPZT z)ANAo@Jyaq+e8SGM1L4^W*!>R?uTQz?Y8JJc@=8uBa{z+3d7T#IT0Feb^^_AO$6CU z0W{Vf;`E2p4T1GvpHF*e_y<(pm+-t@(+oBGURzfY2mlw>x(PQq0%~}C-fmNE0_yQQ z8}a?VGmh?tPt-X>3}-Nx*mT=BWC}=3vbSRfn{f<9J_WB&WBmPkzuvfuk1xRbqsIoO zd{?1L=Q{qB5zQZfJ_%!u-3(;txwosgXte>=zT=eK{XQDTILnU2cBP@&TE*#f+$5?$ z-}$J4kPYNKR?z~L|H#*9@qD{NzK`Cke0}Kcj>UUR-*24ZEuTaK=kD(v<-_X!&6~m+ zEXz`I=*DU*xV(yM6jePWWvAjfO z4TRnc^Gc{|$7|}oq68>A!0|1A&=!sOx`ZBuAefz`7&iY{M%0ldroP?;l>WZzq5W_% zEWA17FPy3ZGe371USh`Ly+77ImQkY)DAd33eXN}_Ojg%F9eIr5-j@7-Tc~W}z_A2) zGN`H_K%~;keoxZV5M{5Z?cSR0FcW6hlx$&!2J0+Lb5}7t@yG3~fRTA2p0VcEr5oqd zQTpcN4d=FD`LZfS_KNc{pJ8N<5YFAo3bQo{`Nc1tuyKtaPiL)p2b9GsVwtuu?YgOh zF=sBvA3{TEP9~8vSiPKNy?ghYPmQRdw1WGWfCY4t7CIh}=7pc7TS%06bC<_!%$&6Vs>0A;Pf0K8VN$>{W!uH=g`&jlhJS^&!^>=wG0A)MppnU+%|3_4|18_3h117?h z#kO6i>FvV+y<2k={Yhw0iP&|`3)|0g3kk2DIAVF;hBd3?bRNUEQVJhFy~1}KnyXFU`8+N|goTRj-6Te! zoai%uU9y260FBWHrtA`)pk965PZ?YOV(0p9=Y~rPw0IiT#5Op*h&Tm9<2!h6&KyTE zJ6HMy#{B>Lde}Ja_-e78LlJxb$j|MjO&j_R1I6%=Cz}H&BE2PX%op4P3f<5*=~`tp zc=KSo?sl4agQNE2gk9bxWDtijbeZPYS_+eN&~j$^r>y!~0i0~jvlw-Kuo#<97w zN~|4egOv({=LNmAeS%~3*dN@brwOx6wdXi-HL#?wQd|Ei1C1(dpYD8X0fS9*(UxC4 z&|-7J7nTVj=!jryKlz)cdx%D+beswDzYwv@YH_$S8M0H#wmQAW_`#IZy1#{~m9XNt zbKl;Ie{rhuxdi{74jgNwEVuHtbRaRhcTgQJp!(-^*P_g*IED%~AyLI5H0@}r|AG4s zOg}d|{9rm2I-SG1%oQVX9E@G*hnAI4*&+GOqE|AZt9f3c(3S({`ewKPbd`sGdcWYO zW-18p>_0csy)hq5Rj#f1ICB9_seiI^U3?CWD=C*P=VMSxiAsCOacMNU@5C|78#KQT zPjA%6#N>eCuLhDM-yJ5Lh1o-a^vbfgp>ay1PT2ko%-vh3H~#boNbZbkCml!s*SbOr zNf+izBo)v&f5iRMBj;fv^HIa0%y6`*|M8`-A`N!}4Sf<5brt7;dUJ%!W(zGpgIXoJ zsyGgCLzmI!lGCRDyxoHB5(YRL4g^M1hA&RG%fpOXiRF2XQ>asqo6qR#UC3v)kCAaO zgz?X}BbT{+p{MNa^N<}RXzcCdtA241P2LaGep>DTqptXQmP?`#Goav6(M^CN@8_@C z?Y!}fTe|!^vc&LogI{GU&tUq=t2cFde!Y!>u?vCjMV@RhgPKUC(oOJD)b7{jwKmiz zaxpmV#{!yW#&5kE+<=-NU7#Epx(G{U4;eevg;A%WtS6n1C6IJdHSBHKpn%7-egHQ> z$iLgo4RHZHHSo=l{)*`hPBqkcsa()S72|gqcRf%>spp^jxKL<1OK47WQyOuiYU93A zyz))>ktZx&O_J;krNy~WD19!!_KOMnS}xCXxDw+xPCWfkU3h#PjsJYmUU$<9T9QXz zJ#e}UE!6y}Xf01Pcb~Y`DB>$>$=v#2%Mnb6y{}S*+0}x9_PH@PvhG751(8{#gWg2! zMFl)VOyp?#4bL}E-19Ke4uHB@*eW&31x+;>=FgI``c+ZElOpf07lELjnoYm`pMC5P zjS! zceH*I(06}QI?dr3bi8?Xxa4OID&9KWbm+Arj6E*DmHJ`@)?ykSHOg1>!XyCjqUXdjyU zOtErbl!1X=;T$I)^+R1Hij!OMu|N})K+ci_<(Pyx&F_NXc!gi z`eFHksgLJ{%pYTT=+ zp_kS=(Y&fkH}M0eqqV*}ar`B@7Y*;Ty}mF<%WDDg#GWIqZdiZV{wA(mbEPgcoVTe* zw!8(+#(X}uvvV2RPwY>*8~Pp!tfIWDQ@VjQxnJscb$!v!Opd*sN-es>ODY|3rTNy+l&VBi;|RgwlMg*B<;s- zj6K z?U*j^8tKixOg#n|xm~{|-*gGh6lqUim8(S)ypqc-RGPnx{>&VoWj_{xV>Y?wLU&6P ze(hNEo91+c8OE2=(O$H=52`;M`+WH}4oAPNI*54wjpzJYO48?zx&lY0o@@FJ&|&LQ z^2TM4*SO%1y0_$&wj-EdIc$-Y687;Rj?suT zE%)OJo-Q!&TTtpOYM!gBGFbBl^25u|@irUuv{BPnZ+ES{U>G_Y_4;CF51P2bBJ2Ht z0QHA=?(h=9bV3(yyt((xbO8lD8B(c#d=&MU?`qxph7FCZJkI~JxdWydn&4oRAJ~`4rYI6i0^t9 zC24(C@C=2!R^@u%!8l&pP-mMR8ZK^oc=|>uTF=N&x+QQ775x+1}_gGAA&C=8oVcHtp#{OS#o6bKV@VEO0@h zDPR!w#S3_O$YQ$3^#^eyE|$S)DB)vkp|mfa&RO~Fqo*yXqDIm){0XL`LSp@vJa*U? zeTlOn^9mHB5w|;BMD_{Pnf10n;^l@uhX2z)tY3Q5Q>@vK#v#=BWjJlW;TV7Tq@8fS z390nXpRV;{^}5MgkG9jxIxxuFH`I2{6SW>z<}tp$LC=5e(3jN0|Muc(Ju!9wM0`oCdV{8_hK zb1qbA1?Gi05^u2{gt9vI)la)Izb!vjpx!8V66Vr=YR{@QqFy~JqWsy4W9U+hI_1p( zB>&<XAn=V!ogZG6{FOfPSy+QLJ!{VJOE&3t**UlF>x z@m7r8I5g{NWxUOk@-XTRe|xw37N%?6UR)Am{%Gg_9OrreVNOF+7I6J4yO4EJ47HEN=Z-BE zL*w_P3up(1_Z?;5(%t?D%Lk0gIvLA;?1sNg_kuimOwdwa-)8&XSr{ojk{X^*^EZGz zT&!5rECVA(snj-8EU!{mHTwHfsSBE(@9`kzW4H#Q%G!Oow1^GQNw^#AXs!1J1_s`b zUbDw^c4mgMC+Xg){qyT>AC|aa`sY)g+H!67>d@81`T_AAMC8Nrt>kPDH2*Z1dDpvt zag;lK@L6q4f3xpzQ`rZ9a=s&G9^8N^%)&J zX?p$na0bduW?p)`MF!CcemIG%e8mI7hz{@Am^Zj?pUs%T@P z@g<;D;PJ^vcXUy9tSaNh&1W$?M0wz}@|uQ6g1IV}Cp(&AVI-6F7|!1sJ`)lwIj=cF zs?M{oCyUcjD)K(sE~ts8)3wU|O;$w&PDvfLYAo*fJJWOU7I%B1iT zo?hLDUhsk>>X*Ip=wxsmqIgd}UyH-?ws?Zzuc5!fIJ$Ti1EmaW)Q~)BWf6V??NUdL|r!i<~xJF zT1DOY6zh$GXKcm~=|)38@#c(Z0ERao4_T3)y>bagm5xZOr~`CZ@1d%k-V4o==8S}@ zaiFF!HU?b8_;u6YI>ku@Qy8%K5S;AO-muY-&ueZZLR;ZT5+ZI*qvnKR`KHE& zI>_y!j#Dv4;k0HjEVsHrqMk1*J4|;tPtP7TwB6wG*RMc}B7qmpR}R4UXNiX1 zcNS2adET`ui(7~s_gFM^$^lgl#73S>WCgP6d%d4s_h3$FJ>j-1FM7R$X(~iD3#h*8 z?~QvjVa2sN(RtTWiQ>hs7nWlBCT2rj1N;An zYPr?r`7i_L^~QH>K_u(Tb_2Fzs8&DGXUL%(DD*OlkJNmi!+|B+49nj9~<%C zm_=O`GISq?tYDISUWdqd7lvLRPKi4y4(ne4wLSH!(>#lcG(0szmNa67`kk2+%7!~b;X<@CaiRJAu#?vKlZF~;% zlb_yZkA0VcCW~YPj7ImvoXI^)QyUXRi72fUo3TJkmoAf$W;Tqtl5f5~>JH?AC8keL z+F&v+(>};d5{2$uGT6lLj|K^G!QWyroMDgB!HFw6MAR2IP3doM!_!^8_}aW|42=t@ zFg!D=KvOS@j3l19qfxZ{k)8TJ2a}bFaLQh`nQt_f-$U;_FT}%M&-yyhQZynW8VqKUyfQ z^abYsjGg!?o$=-%5ao;e3=F^HnX;4~FmYh`_TLlN`RdPBp~=pyBS$CHQQwK=e&?B3 zG?_=ew%wVAZ-J@5J(yVIR$v`yuVA(al6N=AO zmCFWTc)j_ToUBSoc$g~Bk1H#31#*S^0qvI<@3&!yPN37eAIEUV^%ri}A4Gm16?sR0 zBi=jso14^}J3maR+uZ0pa1dI={%(Klxd$!WI{K#f(e8iyx+SI|;2%tvz^Kc8@T zG@jn-)-0wex?sox+SI>qi3X9L% zUZ1p0M{KCY{mQ$`J9Y8Q#VqG~EbGz4AGwbVfk7~OWleXajtzCba2I@5Uj#de-Hy!0{&^mSLdZnQ=3<)L{x#wSix~$&fB5OfJ z-O4n@B2JHIy!-ATANOw@n@~&de2_BC&HqrBA-_XQCkygrOvKSdF5~NKyV;;s@ZP(erMjrP74u(-<;#oWLzrIH`V}~SkkRzB91=C zpwOA;JW#&4h#Q@d0gB~p3)niAodeUftUSxqg8;+TH^McIv2RODNpJ=ZjcbNA5`{A6a0?d9< zi%Rpr{N+W3$CHXbpTj89Ww`Z96T0IgwRPbnEVP;3xZ;KBBMlf{(Ji#0}1rT%9-`A02aE z)ORr@D7&53?-&YgQW%*6m7&ifP_Mu5pWO_4R?uXEu=zF)1Eh-uwRPacmw|! z)i1k|fL{cL3ln~pzTJw&#nCd5?f1{mrq+$IjSJiFkiv!rMHIA^_-Hs39J7&0w$^eZntE%LS96(`*M>!9hF97>SX@>oCBbr5#|4-^c;A;EDu-A}rRqFw+#!3<+s60rim`RhAuUtq zaSwucq>Zm0$F4)&Ik|GzmjF}znSEDyB+%UT4^Lcja-jaUfS=2O5Y!iW*+;p7h^9FM zA`Cwfp`<&A%>MQwA~FRt<8PUwq4-DZD}LF~kiab~QZ<2RGInmiCwmJBe@A}eHK(C} z>4&7)ustdbD~mfYx&&3~@8%#Z3-woQX^eY)1ZEl~TRkODqdCuU!6n^BnBL9(Ha1BS z_21%BP1rMoXE-9eMJwt)8bWIae;U1j?#*?Q6Wu3Z=sVr4$of`19eeNA2eL1)^%?Tf z!T^-fWW&Mw+0|Jb`y-E*@0?z!cHarnx>cGk1@wdlsGqPgLzDZ%DXl%&xDVSFbHK@9OG6$PujDzj+KT*gQ88vONy< zN6PlZmAprboEKOJK7auP1sIYlu2^CPJD@4xf%tZo-GhVLX~{5(jO9{|{9 z`zlrEC60cRu2awNZP4&AKZEZxwmv2(UJpd+@=?pgT!fT3hSw*XShD^O@Pom}xiZnw z0*HFYWcsZ$)?cfn<>av`p555{2Ii(`=t-f%>xT=2bE;u7^5c!WPGe|La9QJl%6mkL zVRNW)*#~1;Lh8cvSe=0Q`yu18YYZ^XMz?5Rg!vyt|7{F!$}92Ahq@p62YaH{**D@* zRXkAhK`Hyk{tu|F`_b7O#@R4RZ8v^Jh(-AiB1-08(8eFlWsDG|g&d*nY}KwWnvG~K zw(m^b$#c-K=gVEOMk*A2kzOlTPlodGV#UM9H{@ZUBPi8vd$T?2eR<`<^&=l(Fg^pn zgTe)~bt~nuD(%o8Wu{Iev77NZBBMy|7TC6JskMTH}ZRC7R{L` z3s_&Xk?b6Ak`&DkhVJP+iD&y;(1M5;&q%;m7^a{1Ge~=l7WBz=mv3)_$^9;TpDrE3 zu?>`6c~pn>wHix5F};7y9euWwC)ytHKozwkuWSOs(6|og{R0;;{^v;6YRek)2O z@%^*4AsUmPe=WMB0i`SyS;2ElDRFgO!w?>2W?p+=aKaw^>1DF>P8NCa=&mo&Srya}kI_AIp86 z?OTcRkN07CH`0s1ke<#5 zIF>gjFHiD%|Mz^LwGuU+5+52)9u4Y>dnnv%L1W)6<;!ABQ2#b5|2x-m(eO}N()nZV zh!}il9B=G^=1V--IfBK|+V|5sjh8Y0^6Y1E*-ICmpqV!@9R|sZs5JDq`(er`s&CH8 z>3o`izU&t%J%HPoKOzkrG(7bp1E@S(_b}bVbmo^fd!n7TnTWdhNa)nv+h|Bz-#3(m z2a+ovog3f(JkNoI3$z>3@I#HOpSX==Y4u;U5~1$E8PbVX5?+&b)qIEXbQRVkEwny~ zXsRc2Z9$3`isawBhrE1@2m)`s>iJ%v-;4DfZ|-UW;X9XTQq&ykjgA#dnpQ;JuD%6J z)*JcH^DojvNATy--=Pk(dq=SP>eTN;$5wZ>Lv{IPvE=ek&@p^3$S)D!s3r>zhNUurhzCr=F(2g-Xlr@;Yj8prWjo25GLN zK%Vi}2>5sck)C^$DjLTjLW)%MrcO`Pb!qqa#x3>;pS7>*^SanS|MSc@gQ}%}`+2X4 zS9lZZL?TV~O+GN$B_!y0lokh}{G-`+YCi^GDT_ooH&TRSnB6OPf`Z}wC^Kzah{YH$ zw$3m18;is!B0Q?W9U>V*{ZSV^-J&z7VQBji)6%Vow1uxzIOdly*VOts$^a*AWCm3M$C~JOl#(n4;^RFN{Ji;@u2}JF}DWXR2F$KD4cT}(O z^s-%fK1{hxQo^5WLcFl1-jN7Anoat`Y4K<`Oy_-!O=_sbGY+y;t1q;n1+hUDa}rH2 z9m@Unl&$Y>)E)Y(S|@49%MC+0u2)&(gQ3u&{-l;hK8*L3tKa-K2Y;S#;gEN=gNp9+ zYwylSqra%4*3`o3M_23D z|MOuKTDRYMv2sKUx{bn3e_Lgu)*mPD?^eU=vK_|?P?ZZ6Ee(^!q!K7V-4~>CRjmw> z2T~7<=r3a98Oiu$+p+flWe3oxlrdLs>NOZrQCV`auEH^HYw*Qq%%Kt1la8#jGH7wl z;6|1DV;~(6x9Zi&fccx!Sq1i|pzrPLoU6m-c&2>5Rwl6=XqJ2b*vr@u##IL`23>cf zRxj;Ke6m=dt9hx^b=9F%Y`-kT?OQa}$TQM~0ix&LNm)TM^4D!RkP5e58r6t4wHQi(p^a{il#^g;Plzc8BqI z+4^edN$@MF_R)&zf9No_Jgy&nSKUVKV7OeR?_=N) z^o#B|IS}LmqbXUbj8PS6c--+4tJ@ABckkcl{2Jkz>R3)3MHrrZ*6y`lJUzxUYP8Oh z*`b8RtF>Qt=1APbc=}6u`$bLGX#GIYMELg303kWltDRQsBgc&vFLLNATy(?J_0zq( z_wXc8m6E9>IrMKIpZQ|ACm%ZK*_aYYSAp7^GfVhgh??h8*j(N_p;7){W%tjI;2EAY zTPptafNmZf&v4fRnA=vzprz}GNJAomL%#`#prg`)GfTiTIc4gV~5Q^**)Zm3@pQ@)HUuiLO+YWf7T z1dlLYofzn9Xb^ByEQBfxQN_E28+f|SZx>TKL!tBgRz}wkS};1!(VKO<50P#5*5#jg z3++C7yR0RVfmuV6^uAL~XwgUP zezUe0wA);MX;#_=3$K66?@ zp|;8Fh5QbbN=&>yzCzO}Kpl3EviFK{p|S@%j#RT9M2*2xx95s8(NJCxm-;&j%vGYf zapeXaL#7!sf8S0-;&L;(o?ZuD2EV*&JkCLhK`TY+?S8b#3Tw@+#n6x&BQIndhS>$Q zrTS@mG|V0G8>wF4Lf!sCZ6*#I^%lvpFS9Z@B2krZ-%VEYbI^~je&G7qiiWz!&P2QW zqqX%;U!9E8@MDr+ET+^9WnI`TCYBojqgq0aIXgyyB>G@QiEjdiz^&ojJ~^1Ze9?4s z@R1F6G_U*KBk`+PTs>m@KFD@tLw^mFa?NUABHV!W9`CyIG0H%WKfvGgaur!tPN%hf z!RmBlx=OoU$VNE&o9sVStN0-+prUiLj2j|fX|fG!w?UK7qYK2ivA)O?BPX__DhViV zI{piG4P*QFj+6Mi3oH!Yv7!; z0Lq9@)1(uQ^%la$&{amB@}wcrz7bWEYZR_%Urbz%G>mubgeB8XSz!)TIc%%{LZ;Kpm77YM$9KC{m!YFFdk9`*w zo`&!>3VM0mhA@9zH0Q0wRn-6fOsEuhDH<}ge9nmVnWUZD4ENJJOM5Os=Ryc~lsrve z3c7ezPG{8r# zm`1}|-XXo$X&zATc-dC|7!&+_=r-3+{0bdk?x>4dxZ)Y)j5%Lr&O=w!_ToJe7%%>p z9OHqj%>1Zfkdn6f${EPrW)NyG^ajoEr@z)k4uXXT+z!%v!eQZ!kR79(6ihy1x@3+)YCw6z~@GR{! zH5C&QwD)Lu&v?M=s47}o=GGJ4eh5eZo7+pE9qT8u?zc^R&GtX=Gf`X3`X9Q&WcUrC zADUgzs2=bAH*GuS@6j87sPc$Hf7tydZdg~KHBVmLH-9ylQxnte>b3yeJ_~oS4~NFL zElLlrW4OVl3!9q1?zM%{`jG5?vIc5&9`+hxXM*BgEu&H|H}v_1p`G986Nt}Yu<7K- z`ko#vezZt9Z}Ub7hTp`|uP#5PeSbLE;BhKfBTOABz3Vnu1#`RD@8_nhFYmE@P3eQ|SjFyE z?0G(If6sUr!--OsBp!G4Vf_M!j)$)nul1rit}Mj(2kW;{FsO9@dtV~tW)VIf$*+QT ztut3u{IA0r)&1`q4pSVvWBebj$1h;0K>W<}!?be?$bx3)58rS@Eqhgk&3<9Ja194% zlK2xg@MK`1+gE&22kYBDN_l@b^CPW(53}!?or{e;p!Lt~iCcp6=*RJ{)5l|e;TgAz z-WWGgK_gRIPA7=}{N-qo8*h#!p}D;!i9^FVsOc}0=9jH(I7W>wi_t{{E%e@3i5+`k z{bcB%WtFMs?8_P#bEQ5%W^l>+4M0iOl>=a7BJnG zNVh(&nWle!VG*;^fgLnGRy5nB@fb%h0aX;$hQ9Dj7!dpWU}wZMhl&TbI=MRo>rCiy|O{K6;*h%t`8ZKn?_iJs-Y?)g_>nQ>R_`H0t`6;i@o(_vxy1&a=LX^`Dz=$Te3z^BM^8a?@@cpJ32t z;ro>y+IoccV)4-vp*R${Y{q=O&k-ifPVYIg=`!jsbD^aaeRveL zA7&`LWj=zsd+MD$w+%ptqq^&gwjB%_GCrpaeB1E*F^i*zFX-r_scSFZ@@=yoW~SG<*S$8Xg2$P zW zKB~yW4bw64Pd6J>;OO2@YHi=E2BjXNPke9cqqfmZ_4MYg(3dat`^(cK@JnTrv?${Y zqM*xvm~TkHnqXnG08cFtO?JPranD9WOu}WSUtWg(S`PmN{!I8iV^E*yrHw{eZMB~0 zVCN-?{a4HItqwGNT3fwxIRI97voTIw;K8x-aK6)sPKB~f66}jdCDG7B$s4$`qW?9& zIJ$3A4burEH1mUf#%~bQfoi{)8sfa~KXB8ByghwWXgqAlfBlio_`n;~yJEhtTzG?i z3k>?QHy!n_hT%oMZ|V3uXs}i>i=p5o476=MRs9~1!op+P#+w%KbgNs|1(*O$_<0TB z0{6hE!Tg1a{g^J;h-jJ)Q5@4d7(K1UujW;QrZ@&4_GodUN%x2|^*<{Rb#g5>-+|Wu z49(y<>K&zz!|Gd?8PZm){x~9}%zjqK6H#nKWeOQQP_3;1=L($|imp8ycTwvoO58j` z@iJ3K)44Z9vY0V{bn)gTcB8U2=)Y`3wYjDT?aRbJ(xjJ&+WI15fqNUwnvV5;I(!FS zyt+7eocsdb9LqfYVLtv<_@pnAe6Mt) zKeTY_$EUq#xbdvAA3v17PJ^exovbMGDx_6`tPgbn3a{OhzUa>Z^SSHEA;xoPq@(I4 zr0jvh>C^+sH4KR*?q-nr*ghwVl^^w=RykS{<_JqW z1;`^&0eE)r_;X1G18CNfQKj*kKBCz7@2|$&AfoS;6;1^cn4aBTuaRzq2;uWW+3YNc zwB95T?7dEV-k!G)^60kDp=Q67uL)IAXohNc*Lt!GC~Y3!S{%thD%yQ;O3n>+-A=Vg zOLRlo3?3%iw$gNzVeEKS-j(n^=#uyr`H7APQ7Tpfi#v*e(7f`|?;rt`ML1g6W2cwBzy=GVabXr!c+^hv;p*1fkNSE)Ge~#(!~7}Z{s8N* zXmtCgRo+kpbDFk07g-;|QqRhTBuQUr%YdhJv$@cBjuG9tibLd((i`vATF{iKymlm^ z3uexLZm(O!{Pcky)rU6TpV8o}pPGweB0!ZD-oja60;5U7H(Pe)pbsBjK2dEhLrqSp z$`5w)!kjYhgVpIZM3&C+3bXq^Jcii`ZfOg&9`0)06Y7p~RYb|n&&V6*f7z?m@0xos zjIbWnmA}1!^2T3^?fO;$m5Q!R36BGyfAb^#ogWimDMs(=8Q~^0>M+)#VlM_WnHRzh z=aUg>>Vxe=o-7!kIt6~2V?>03)}tRb9fXm0Iv4P2mtppt-^HQWP?$H&<&wDaFW&R# zpc=3=prIz){=%mH(8@R7eeL5BXbL`4?^?-@XXNVq+s~1M3bJww9xY&bQL60lwy!rQ zAd~yr!46HV&RwuAHT~Xg6V%sc_^rkP(+eATHto!!BZ8wB2{b8`+7A`sfj3vJDiEP! zPJKrN4G#-_-N!z>fBhBBo3P+c3s0byF;-%r;8*BZIAl|&dI9y>ABwZw+JmV5N^ZK( z-y`C)onE*YOc!R!CN(Ui_zEK7QYUk4Kf`QMQbK$QrcXW}{zYB;1ut4uv?njz#`FZ% zZiW9A;7UOw>$=r#ZU@j(X1bxA5I>$S~^cg@|@)jePv9iu0zuc5h+hM8jlRs zvPI}FZz)3LB)@uG2`!(4W6G_?`MA^Sn>f~H35F$+a%dDt=sG!!>3q$8zdE0tit&_| zD8A+HsbIkb2N$8dP)wMJ+&Ecw^bi!FTzvcc~s;;)pTkC__dQIGLRYLY|0 z3?$KXs-Q$q=(o{(A|g+kMX}Kd!&Jqyl;1hepmIQQY^S9fP?VMqnFL;e$ynvj`gd|+ zSnmVV0|N{{R`8(Zb5^1a^jT{x-#dc&Khz-PyDwV@d9FPlckjgb2%{JK?)b>i{A?h| z`tY9b-N?6Y^9(*Wq$`A07heiSe#k%PSNrL;Rlcm7&zygF@4R6<8^KiQ?P!oUt+s^ z)ec>UntbQ9C+wD}(ueD@^IHwH)ReqCn-~ovrugRjd%vRb2Mt5sp$v$$$TwIn{h;|D{U^oDFKBMZbVQd-gna|`S)nE8*xvT=0UTqK zUCholZ#45j;q2h&E2!am#@+;vB|QDU>od|7tT4PaNA}LvpJ*VuTl<~wew6NfG)djx z8FeWs9r}xxguw&!CL^H_U^<$e|Mh)5Hg0@nE%V{&sPE60%Xjy&ATpt=;6p(bjN5ZL zy!TE+?e2_k4|wl}H?AM7n#=s5iSttCYh@io{yu-x$X^+z>_-|>9y_C1>sFJrwOrJo zuy-$>P6wzZ5{BP-tI+%=ldCEN_aIM9s&ktYEe=3K`CUbmry9@z!SRazD_vN5CG*Yu zCNCngNUDHOr49&lvsR~2$Fo`)t z?+_wv5#9Rn+AcIJHO}1a^e?X9oaVQ#_~-wWTX@dOcm4PL;pv`esULH^1%K@DYRDb4 z)3upaPZi17IX-f8Yr&&O&S;LENK!KtfN@sG^4X_lFtJUb?CWV98rNRWGchZM!5yKq zFGDe1jq;H)`4`(Tp6d)TB8%Jl1`y9z$y#rx@g@*n*RsZ#u@5zza8i;SEW_rtRcF&e zX(aR~R%f!C-T&|B$1&`zy}R-j(;=Q;tiDG7NFGtd7&QAPu>K_rDQ6#QU;2)EgmRwB zoy!8^*$K{_d)R?+G^2B;U=*x6pFVZ$qa-X>W>7vEIKg@<86R5f3Iwp2lvX~8h9boE zKg7I5bE1O>>jKPS={u)SX!09Wb^Jrur8}STtg11%Wj86*y(Ba}B9BK?iE}5GznzBJ zp7WL-y8cjiYqC-BtPuR^@Jji<3+w;!W##Qru_wQvmM{~&9ZjI@w@0}SJ&Z@=d7Bts z#hpiG?I{=bv;NzU<9*LbvRGem!Y7+NxpNrbuUx&p(<50B5uQp8%C3)~p6j7|IX(Gc zNQx!Hj|tOTA~K9@yFI@dx{G`2PWIAxkT3`Ik`)mSIF{u<`r}OcFs|;ASn|CNh73r= zVL9FZ9(QPd*%sK|umus$2J#P1#6Yu?ccE@QtGjEh%$!kYxUvGVxpPWKO5xT?^ zr)j!JXqaw%l1`Efk+!fCU2Dc+4zHUsKZ5z^(_EYfjVc)+X}RFgpX)0yca}^juQx+0 zQb+Uq(N)xg`}C7%Zwp#Deq*0j?+-jvU+QRDMJX&wF^5ztI6|G;CQ)hGc%ZiIAy~w| z#WT-%&Q0t!M4b_DOP<}c07|k@ec+bgs1^9 zFcpfT4_*lqDTcOz=bqcY7vSjQEJV76RnQ1k$BbH>0Lw#SJUa zJ3RBCLIQ8lIW*pJBl2-NmRBC=9AjNLErVk?WiX_RXMn}uKNA}-y@g7pU5>X)i(z7` z<5&Nwt%{C77*?pHlnhr;do6&~;mjhu#UX`n|OnPiKU3 zNyqmk7T?Wk=apTMq3LHM(iPDcmbMr_rGwX*aaBzV+6yGL^)Y)z`~S3xtDTL!Dln3> z@A-5UrfWdFUMhDt5!3Hmi*e5$l1M|Lm8X5;)Bi&U)uutHnjnv+qqjYsnGQh@c<`}i z=W&RXsC<~Zo{VOFy!i^7tFiMvyHtmmg#EsX(Rr0O7mN4OY+hX*t53%<)u_E@&MHPj zJAId}N*<#Dy8)Xmzl70JyyUHIJ2W67?zj53Gnp`rd`I`b{fZiIypsF59Egan8a}nD zFK|pv;x0RaM*x0o{c~FP9-81&bZHmagT}-3)E&1ez=EQMq788iTC)`AQZ-6~aP@%3 zZxLrvZR@zx^(uK-DC`&J=#=0ao|~rWNgJ zIurlmPt75c{{SqW*&Ml}{uZ8DbCc19-^F-3*4cE2-`ptD;&bu7+Dz1F=zc%miw?(> z8EnNBN)P0VG7{yt+kk*_l~qi}vH9r8E@+>YLBz<*bWI|zs0H7}o_FOs3_IBjnR&LL zZ`H+mCfl)iEA7$}|8}Vlun_d|jjRTSe`o0W6@O`D5VhtW2hYLNs4JvhqNM|?BP|SH zdm;V#2`Vkgw2k3z$FW{Y{n}iy2^Q`BjSJe&qdxqkU33B!O&0A{i8+&nmfMYv-1ee_ z3`^50r@zHOP2qg-h8T@w~TX4>rH%)BI<5?ap-Ew*(juGbad^I05DRv!=;U zaD3r!@0Ol0$j%6Ir+dY-f6d2Fw%+ z*85+>;775|t-NVK!XI&vmC-_N6aM+DA8TMpa4O_Y`w;Z4eKe9>V}mZg@BRDLkD=wk zPvs@5mk{M#-pgCNcA;W_e-ZVmS!mho^F_uQ%jeDYG*mhWwZJS>k;zjxUzolkSDYjx z2ZYg^wL8!lOw%*u_KY2b-a>+fTv;IW?cYl^_3}c~H(o9%OKQXF;fGaDcG74nnefZ1 z)(e`#X4wp7SW&eU-ItjT73kYlUU};Mc37S&Z94Yo0(^}M{Pg?lM%*KXGoci@?CsCnPnhGVZWUJ@li<(#N03D!!O z?oU3;fgbBs^${)XzEdM@q8#IoQOB)Ll2ey9n%$=7ovY6XNl$U#zCXtJcs-q(r@S0_ zAlv8MP)6zz^>H3C#p)S60o{pj@=>OyJdCzH#A31yf$BR8$ zf#}yJe8P>SNJRe7lNB6C%ilw@{9b?A#0Z!>Rx}f>scZeFLo;Vq_ReDf%)6%+_6pn!Gm#FS;=7As2}~yaPIlb z%@{s=?Ub*Y#~~J2WM}xJ9Bc;Nlzf&gjFnL-i+8i&7NO{}7n)kB6+&dj} z=|m$~DwdSiJtu;vQ&OD!R>=qJLxBqG1J8h5*PhjFUV~@4g!Z3ANCC)W=x<9PZ=2b(WmD=-%Gh{N;x z6f7PXeFsb!etOm1%T4*1AgZ`a_w`j>JF3)USC+}=g!viMx!Zd}V4<&!aOiROhIyM% z*+!?khWRU0|B#TYpO2z$i_^ZFMlt(Niv3+G8uAoRcX?~{mshlWIF6B-U~1~O^?$C5 z_1;uYN%^NRzb#tIVK=6S{!0!V>ttTQVC?gy2&N0DuB~3fWIx7BA?uvfQBv9rO(w4& zmff91qyujBX~a@AY1VagU!6Z1^mDY~XHbWs!hq&)zc-=9!01P*>JBg#Y{sVZ;vJBE z{sgN$BLU^!fz1I&FujV}KPDg2M=PNDoPlXhoEd7UxcVtEm=5}^R|ev3Fy3};cZ8z_ zg^HuQ|BAJiT#lo!{<&~(YYvW)Kl51mwv9NJ&T^&ovMULw@yE9xyPwnaHqo-f(7umn zFg|ey$KK!M_gwJvf(LP?SdoiloZbC2M!%n4Z-8TF$q@ z%wERus~yq^&ag9{C{{%o9>s6P)$*b2uR?%;z)Q3cD!nWlM}q0JJFagVBcX}OY1i?Y zAT$=zQ&-_iY%;?7-WJPk-fDLr)2r{A zI!iP;-1RR$klglr`aN`oB)(`M;?Ud+=@|D@7=E{JrQPi%XB6%8{2rI9JI*|VF%y2d z=imh$;g9oX*rkwH8L~4b8KK;$H04glRkYM+rAi5Jg+BFe^QmYTXgN95XDhM^It{+& ze-3$tCUtID+*KGuW133`Z~FDYvcomzBMkfS^n46d$F`Hu%V^I{;D3zzICcy0>N~)^ z4tvG}^A2pi2mn<;s=wd(X>r#9^#!V^Exmq^o$u0bbeESfy(?;8%(42$Q)pE6;Lv6z zWi<9;GU<8O7}TpM{c6?Q=&y9^s5#`1WBAym+k#n7R*F%R&mlDKU-|#FPe9^lqW4`F z$Kq^(3*nLE6KMRYcWxZtKAPQ_;c1(syevZ=ff;!X25N-tLv(fSqczP+8OUyM1Fpxbb>w3f=HGSWFgfZI<2v*{;2ED(r z@zaSI_+^Lb22h;!?uztm$Lz<;quD0$Auw=XZD9P=1S~68y>%Tu1!P}SrM)Xy9<~2R z?&tmxtgi^=ThfjZ4GeERoPQ`b{@77O`m?p>cs(uNfv(11S;_cXXv#T7)xVUA=9W^A zM&)AogH(0_`j*(Oc)ESJ5AfVugaKZusb1ClFt5HzQhav<%<;{6KXb$SK=cT3CRYW? z06{1B%Xhz@s9Jea?mRp}v*<_<-E0!-O8Vgn23{}_dA#*+SOKDpv(yeR^TFtq+KgSN zt6effcl76HiQi`i{ z!hx-b+|G)t-&%%-l0|O+7R2;HCrz^RlA414t?OHwigr9`^qQHNqC*`tEKW(M{H6I> zFq5{auq9FqO@7)W^?vd^Ojmx8yqV7qZHarVKEK1ptHW~A|D)O&7(UWsCe{`2bjM+m1VQ%y_y+nz8iPS@Ce37CX49AC>&x#gLhrzs&eV@xs*c zCyn~tg|Nyzd#tt`!=q3*ww(XuBTGA<=__B=bVfZ!qbm_S*Ku_)_v>>iN4*Fv2>mUZ zIcE!mV~^FA$RlXMW2YT*#p)3)OffRvKWKe|5&87J=UU#8Fj`Y5u4%vy4J#UOBlrs< zyerq?VUU3}%krPd5j*Ebvc?*3rfk6BVfmc8C((`zn2z}B+NNKtk1+itqKzAQTHFOj zX40I!wDtq}Tlo8fzOhic;Iluz+YAjKy1Z-noe&Hg=1{sG{@rl>IUA2}Cv>rRu}b5s zej_0PmE9V}eZRN|_1L$poQ=|lj`9;t+*bzBz}@d@+<$5Q3F`53vF0wAK^@ylq(n_; z(BFQ#4^Nl=qaR*A%hPUr3>2!)n?Fx;Kt%1+ONW97fOMx=!LjWZ3<_C^v<|1jIJ@Nf zhk&ikYFqP1e*MJT9G%N(h_WnnUxoqdiYX>-$Hjo! zID9U?D;n~WY_scsZp?E~Ao0_4(G?i1iBO|U>Vc6Vr_Z?sEvQ-F@!3r8t4bOjr+S{d=&fH)`Q=e6;nR6*H8F}QF zc5Ei#E4z~!L~SA6SK_PBiUq>^@6_}NVT9>zgWSLQ#(@yH;+TRn2U5lHYOg$uSJs%M zc7gE$)(^Somw(Q>##WRc->9ea3Dcn<5Nfx{eYyu@D%lbBx5iPy@s(|gb7ZuzKQ2Uf z9Mef951ZCmu!_QhM(NmOOf>X$-!{)ZxD^Jht-feU%L3_@f+f8*R@dm6ca)=#P=+oW zFG2MmSbnlz=$sCtxiV@y&{}R0y+KDD4PM;2EEdrS-74#MFXW0t9&kv@Y{T>;s!sTQ zjy|4(=G_NF6^?2ks#nL^p~Vwu7=OP-X$;dbA^^Ah{d2TDA6nwDK4Thy`M+H{vH3O4 z1T^oWba!e>2AUeeU;Y`S#X~q24NdXJ=}e#)2j9)p=SM>|qP(Q73#iALL(Pj7^FP); z1ymmml}9CulCEmCf6xpc$KJ{Qzc6+vsr>TgjrfbTtEWP51#jrL4vpsm&)EDq2Bg6^ zfk8{r5c|<#+tWLkAKM|NuUgTE@n1q-?No?5semd858fV(+k+OR*DTHs>_N@OUW6

    Y_qT>O8|bA2+uwe)ce z_j)>>RC_gv9#2ATJ2tvl+cEIXqNb>G4=#NMm`m!;e}1?NHO*J7x#<0$by_I#ms}$A zs3n?JpW-vIHbqm>)zT*?%Ar18U(lNN5m9S&1B* zjxKAD0qukF^5&9MSf=`i4{p8*Lo)lylLN2A;y35g&)^Yu{U3buU6>Gj^d?u8veS5hn0mFYUNt2Rn zxk9%e(dct_KW@8x98=ytf$U^DzAbd{pTHAz714;_mf%NqSUz8^^!{U>N9g&JsJ)N9 z_g5BU9`01G?kDcPF&NvUFyZjf0DUn?z;Wz454{y)UrzRw0(q;!;XsQ3H0AQ-&irID z{4$HP8f^as_19c|Zd088=~Z)kE;rEq+o)INK*Ldq4p`ZgHdGPr2dxKOo1^&B(dZ2# zp~cGzTH4!EE$pRH&4@(EzVT!rzYg2|>vlK{_x-r>wN)7^s17@9oFodRYW$Rp9_z_vo^;z%M{V=ra=0O2L zZlJ9Xaw7U!0jcx2I*t^F8iMaP>^zG{-3fN0&9~$rC*^+BZhRr4nC@XNwwWsX(5Uj@ zlP5jGI3~TL&bW2-JTn~2>oq?PZVf@pk7D<3fBpq6No!^vzdno>7Lr@pdqU9MD{XE& z`QO;SK3B`?|GXJ34f4PM3KMhxGry?(&SbteO#5>Ot7WHrry z0ICrwR%=lRC~rS$UnKgXTBdI2fhLCY2l{#LD5v{6!C?J`&CfQdp`T6zv!(`EJ>qzn z?04h!pP;8Df24A~BC1+{i-T_pJ13f6_jeS&!s0`f=MxmxR@K7jEv9o*5e-0DbI@ST z)>GJiTJ!A5=X_VxSJYIKcq|%bbt>;_B`vJF?)ch2b~jguIOYdepDnd)gsKzlP9pY- zK&6!)z3jIP!&fC2yIXx%eSW~NNj=k=0sjG;$I>*p(%B=Z$M?#7YT~N(<@Ea}hj&sc zTeMMUHs#>D;A}MQ#$Ty#sIcn&!-~?Icq0R$C7M!X*O7rbcb(YI{`>@5S;<hnNL_d5#$Jq|;=WA&fZ+)^~h9iM+n(;0QQiM>fEq2u77*{9>*xq^?Pt{3Oy z3}R2C-bUZOl*nIbA>qUk*}637DIXf@aTo{6wY6k=Wv=Psd1$<*!&fhR!G_)@s6ACLM!rS3gkyNF}bs&0RuF@UDj)!uIS za0eA%eV@T3OwS)gsqQNxf`@gXE$#MB>-EnNacqsGThnXk4ajnNA5WJDIHrk7Ny7@d ze1V1T*0Eosbi5Bl>C=&2+ZBsyodmoxZ=$Z^isJfz;dU@H>LESr9XgyMT%q%7YwBqUB5HB&7cOId34KeBo0XK# z0O_@m%WKVfpiDEzHqMnG%7W>;@`w{?-o3+d^fn+cKFXaSXM$>aa$_!cb3x1V#tK@;KWR91>m*SY4t7g})S&PaSQ+|HA zxCJfqyIMk8ElN$a_xW%q6Su~XFR%B}57hgh*1NLq9ExR{6Thk)08{CQ*}g}w%DW29 zy=_-;!}K@?bRT@-gM9Fa@~hNo?Fb1LC5-o#%k;n~!9&4e1&jNQ?(us1`T(FA=Jmqc z+m-O_@4t$o76x1^hIJcj%6{6xC4z>#Jls_04>J7TnYCNY(On3U!@FGUIUl0QGn-Oi z1(V}MSNC=s#Xg2{v^zw%*v&J<8&>T{*S-%G_n(ELQC_PnvRv%9h2k1rYwf{m2HUFzOxR)SsA^tSaT;M9NV8lY}GaML@g2^4VEKXxr;>>QD*rGUuX`yRO}1Dy1d2PL;ScfyJbUSW z96U?%PhG{WED*DB7!|I3h8a2T4uP#4sG;;DpL$*_>eK!;)PE`s_1-^q`?PK;YA%0U zUN24t!ZUB8X5u0EvqnL@y1Nb4kT&~vtKm@drKL@OL@g0s=~A++{cj-t)jurY5)VU^ z8>c~L8u~Ic#(j28p<(mx&EMFFFicc=D|jd#X5MpZdzkRpCfOV-r`~LNa zwmGzW-IaVqY;}83;1P^Uzun&&pnfabv5!FT1;_{GkM|M;n!1LPTG7i-kG^X&H zW}<%ANx9_?W;}<{k1dO#-=T+?9(5|V4vqP|Fng%63*{@m|IMB;ixzx}%sFEi&ZBT944XfG_I z9*rlt3)%$KcW@o)&&7Q(&!Zd@-Jpx6`HJRmotuU#JIUkGnDzTR*(;!^{o) zKG$9@?#gJ6xK^eK7S4G*jqGIL(|{SHHl+}6OYHlM zh@|=H_As>DQ1NbT!x^luaY5C&*wRV~$ny`SS{G@kr(@h%r9vI%`uhlbXE1(wT2@(f zu_%TQHlA3T(RCsWmJhwUeO)XR7F4vd`LYJk;%E5%=Fnppp2|Dk@H-R7dh~rVw*wUb zRY0o0YWduoq*VmbLh-nwb5ty(rz~CM9m48_lB_fJrojcrbU^4WXC_1a8lzlw8d`LU z0ko|AOs~rmyyyI> zg}#5SIlp{d9tIyC+1~c15t8~yQ#VG)s7%^QN9wu(3>p00DK*!EIy67@bvggDCobon zsyed(HTPH8k7~-IX@2}cldHd=hOpl|C%h2KIJX$?dmw{)mIL;ds8CRI)R>QWb{|w2 zg%lCFWnf76)dLP?1C;UA+g9wmD2)C&;_YhtCVut;bV>eWx{MjsmO9sch`KJ#P zfu7HbV-^>5S^rcRz^#acMxq~HB5&5 z5hFTqqWo{7vb*j5V4_6gq~Oy&)P6yGHj~*DQG=TMFPHMeqVgA0@wP-XSY=qe{9zFl z9arn)Ho$NU2a5)HN~AG<&omCz1w~3eiZfwWvaL2kQ`KGvbssc9cLq|6O*2RN=bmQ^ zSbRWb-9vSU#C}6Iv-8p8ZfsB}FRZ?%Z3s>9|NMH#+Z#xKJ`xY4@S+*#Te8^)EulxO z)7b3OG3aTJ*1wdC>5YV`5L$1JA6f~w)4Xi_9%wJKLmq9&MKcj;1`#(fKBc~=Plb4? z#c23m6`T3zb&$z6&dyZ8r?R zKjUgsz@rDb5dJ}=)ySqCI%QPd&kQ@loTF~+&3H-Fo;t7kvUSh@-p4RA<}dgsKmn26 zt?dRLA++!#+4q4VhG)E(zP<9I(ht;9K^iGM77FuZMFaLb0cdu5hdAN15|Ca>ZF(G~ ziQ2ZQnXNC~kK@p2`@~bOjLHA{9Y2nr^Fm|goLj;#Z$|U4ijG&utDrgay!Fxfl88Js zy`}TQJ{*%cr^ipu$7pE&6G6T$5++R?&#s%aM3co7X0y1@P+NJUa(eFrM*BWonODSg z1&u9vT{qaQ3q*%WQLf+9w%sC;!ytCTsn-OK<$Q8mmjaz{0>~;~{qJtY19kmDqU^T!Ky7$gi#K!uiZM@3$A)E~%*`Kq z$8s22SdndvXc}s_ZN5p4O@>)zsTcXhi-@?=ouMd+#hIp;rcB}OG#+6xir07W*?ex(F+G5wr4{hI1Jk?9 zcj-KFGs_i`HVRc7ug3UTT)Ez9--zvjCb`U}jxKuq9LIDqt13@1 z3c7j?V~s`Wb?J~VV!l_AcNRML9BbcKbO*|O3^d=4Fyg@yPrJ+6E|oyf9Dy(U6kUIV z`D4WVLbWldmVHau%>Rhtx&HNjy9u8UQSsO1@g3NA=jk6X;<3qsKtZpoWP%4C2`)B^_fapP|c0-n;kj zCOkW9F{`hbACR`b$$uoa3AM2KkMg&kMkN2uo!KH`Fcj?Ob};WVo+~3VcV`F2hetFE zzGU?=4f?(F&Kib2hY<Y!7L$0DBKz?D;Xs3qZU69(2Z5zk{?lh4HJ|#64bE$>WWH^s@qc}1{H(NB%lx%E_iKZ;7w;duTH0zzS z@3~F6pmku+nScgen0jJ(h&N3X`RUUvE{0keMr(*G} zCYx^lu>S-hubmH2cQD!wm6!gyvh3prLdhQIhsUiL`q8caC;CC{Jv1r(d;XpzhVR*N zjV0H0dL7K=UBBg)wjMe|-j+3t)Ac@7?CV)a$y(JvyXv^9y~mW|pPqsCP>s(XOU&5$ z#H+uQ|LqYhiSIc^>idN1+AUkg_R7Ib`>tDQ``18k^|fgmSvn2^BC}pQJ;Z$wNPMRnX@bO2q;r4+hff7trwI%DaIQrf7VF z#S_Zzaul9~I~OfD&(IG-Z$`-9Y+IMSxMYHdrRlYVdf>l(Us|V3!uI`KPT02z(IlXZj=NuW5{Ldi$D)WEF)$ag_R5zQ z8EE2{ain>*qMoh?Zw8qeD8PIWkyiE0W+BX;tKfJockI2QR&f7aL2FWpb)>-o1IudW2t zaKOhORPQ&pufu$*b3)<>6_L%1&)CU{qlsS$W@jg{eCc8QwU10k-oZ@t(CM%5xq&vJ zYSku+*};?yE7OzD>{i{EjdxwU6bkV$n_aX%;=)59J|1WJAjQsb--7F|fHOTZs8_SO zI(xneJxgOIeo4XPE~(kTAvf*>P<=iZja{@y9ro6rP5nZl<)TZx{@E-T=c5`d8CXEU zwrRGnI3u)pE;;djQa&nNyRGg>+(kTFTXySU6g}>MXR}p&!S|ak9}!hbqM%{RG>nM9 zt@ZhY<+~3Zj}g^x)@JB$OHA({_Wg#auiH!1u*)R{*iVIcez)18xM+WV~wcu~$LTo|}&^c|OC= zj|bTSbvtsP&oa)C=FN!D{jDyZ)3X_?lWd*27J28>V>HnCE}-kxBEx%@Djm1F*qj35 zzNzIx=hbo4`#Yu&9SeU0-MowMPj009J5Z~^aV38FXlORvv+1&eGElc{%NC+A@E*bN zrkhN!DM?TpkQq4bXaxPD`)=|E=b?(NisSPQjCDK&r~dKzI0Zy}2V2GL81k&qkb*gN zb7LFK>{ff}Ue1ChlJ9Qy*^8Z9jZb_Y#Yofb02t7{9&D7l2w`R79nm(WDD~MFu{W6* z&Sjp`mb-%MF+S7*2hW+%3A#TCb?7zUvY8fy#n{hx%T}nUXkAQgvz-Ji-)XwDd0Y)v zK1@;S+>C)NHuiF>{}P&+On<)f$%lXXW|m0QaxKgsg)LioQh_|_AEQHHz)8Z!ZKtEr z_7#gG#@@O$cd>>GrhR;e+?h5&yTNE~N&?2?Hacr)EqL!fP=(ysoV_r7+XCf}Qx?MO z7|+G*u`R#y`B3phgJh*ChIcY0S1Q?O5RJ%a?y!#RJ~UgiQ!z<>7|4=uJR|P@h6#yJ zto?@>@^_%_#eM}1156J6HqmL=yxxyty&5y+`iw&9auLV2*YRMNo(3##S+r(a7KPcG z0OvC_dcFe8zNr8BG$sR0-Yc85ZFPoMcO-sZX8R{UCf3z0=g{#ep{;70)<)d=fBVnI z)2}!ZW{-P|NG)LXyY1l`O4_39pw-4BfBE?r=yvqjxnnmyZUEH=HokJ^4E{-|ZQ*!z zFwzJ5Bj#nE$DeB?IT#4`hn8~KXu1G9m|GMnXMbL)Ti z3(w@^|2?*uj&l+g3wK3*;>3N{F2(b#EJO)F8>zBrV>cI@#4?Ld_=D*0Z>xgtZ@EXPwso8&H z*??4SbLm+50?Y@C@bv6r#4~EU*!^?<$Tc8b8y`BB{SghNzb%Xn{Ab_x@cewWcMTfh z*<%?!kKrqk9UKZe6eI#xEjHG&@Z!L!O88ZP?T1?5Kd`z@}F{nHOF zKYIKv7l@T}*EG5@KYelldh|+>QUB`ry*$*QFovkOl9=_unBF%N;>>qB2g7M=Q+E5N zU5e$|6FAP!<%(f>D>J{Zf6#BgOaFiDqVND>AW0>rn)#}?ZX1ooRhcttqudlVpd}d3j;qM!#D~(673JBVZ8T!?Yb|U37;X; z%++Ej2g^Gr-(lL}athOX=lb8Qn={7j&S|9&4&&LWFdA%w#C~D8%suB$g_th);F-=o za6H!X&))Ur*flCR3ybE-_f2C(&?vhBWpmYg=&>KVPCUn0CzAhg1AERN8yMEQu;G^6 zY4i#33#`+t*3HwOt9`@g_~nN%ntFTqAPVM2N^}pJ#~+Tzg;o{nq*-sv&%ft zl=T8%GY`Ow|M^W{-YURQi^iAGb(s8|S*UxE%Gv_GLk>j>!K>m*!HQz3*th#w{AtuR z@3&qDW1VO}C+xp@AH&CHe-iWE^h+3?#pNAQ`K%X;EY|p)`Oyga%qx~gSt@|~<3U1n z;&+&CFxfxCR|Vwi$%hZW^`V|H+vEZX4QQQzD89|%H<}Yx5}%eTguL5U@mBWQ&^AVL z;mm%92yV#xT0dP6MU5>t9+_rhI3|V10#3eku!COJofaj7;xJHA)oH_j59;bIpZ5oh z;hFmjyqcx3dix2FpC_`;C_>+l85usaJeV6U3G8sMVi+fRdQZG@@|9KBC!6oBpOFi_ zE(eGPkCV@fRx`BI1G!}~$ve?t$g99h+cEyr{;l}Q7pDzjTIck}s8?-xHfO%>^*vaf z_^6{?LE7^xu#j%4IKux2S}SEVbk&og&*5Z%<*lDUW53%$l`I6xbHSL+GEZSVRynTdNNvn|4+* zB}Jo=c{8foohy*2^6sYppbDxhJ4eg2eFLqN#=>6uPB3%1aIJ-+2Gpm?tc1pQ;@Olr z4-a}zz(DcM$*CR;M~|eMT!43&hi?WTNpt@n_+(u_L^EUWK&6VRBai&*_&Jk%GA`WI z$8b+6I!`NeN!GAX(EMiL;0BogakWOLqXXu5lk7$N@1maka@k{9N-z=W;hcMq?zcp( zO)5ca_4{BX{2Q-S@^`3Ac`cEA3X2bowDP|CscC@l_U1n~zaNF6V;UDsu78Jt7wd>e zZZP(1K)n4<{qOrw{-esorWNCDm}kCuq;v}-p7h88{ortpM4+veUpweT=c_Lu3%ljVRS^6vbnQjBjP$v}CpmO?h3Me4obnzdFi({tsTP<0!&FJBJc*t7MYy-@x4 z$-FrWin==bD(oOb-VeP!#|N>GviTYEJ>ZAfxtemVL_}UU^SLCO1!&L5DZKkPFx*E~ z`9ks|PQ#!Chsxt3AD~L}IoH3MM0H%x>o{rjd=DJUVMmq+ZFF38JS*u?H0iJro-^r- zb?1v`h?;n4$Cv4=Fq`5^78qdElSxr+_?K4LIWiHPwxBwO<&CtDzx*30NuO67v+=It z!>YN6;%pHo;m8nIgjrj`;&UA5p(@?lENdnk@(#bB@hSR0#G&BAlkgp`n_BqQ!XTpbh&dDY;UGhZmvvMUWZ%gPrT%dvmh+A@%l4R)f z#C?!2_J|%mUWcgO*P1uX{?nTeU3}$if%)tE+iNzYgzSJpWMO6T{Cl&`B9L;ka`1W|!eL|?ejo0sq}nyZ3*@EpG>EIDXEafaTm>JZ_H@z~k&m zX0{|Z7|~6vVeY2$EJ3GT*uGyDhCsEwdRJz@Cc}Fh!hRI;SYtTMleg+^Nbl(LiiXFZ z;|~3gyy=3$b4F59FX2;~URXU2!@0%57$cXqt5>!AC47GD>}r%jGH_6Nx!Ek{jpDqSu$2fNu6;%yyegf zG*otcQ9QO3MuXm|TPxjx3SQ4m_9omwtV`(P{FsIo##|inHy)r=fgi1N#k$bv_B0z+ z2*6CZ#p3;0;eUGf?X|S?`PjLJ5?)T+`vHXeul2X2-T~^pi$AWN^@U$0VtZt-Fy!N6 z`q@fy#9(|6s@rqvT1Sl&he{mnHx&asc_Q_0!Auf*_vRMpt*+7!+^kr8yE!Q2#}tIq!EjfrKl2ot2>t<9?a7 zH+RzGut0LZNmzM?L;dF-^U`?Fpx}y|T)8Y}sPD()q%;F97+)z28DjB-k)|h^EG@|l z^U@}_Il*Ee6}1$q=|79IM{%a#r;T4_0!4En_JY_Cpox6+3*1b{6@u_Qv1d&xcd z(DQ;%#kndBDjm-Exu5(D6m`e-A7A?c;fj*oZ*B&DKKlC_nqqz$>?SD{(9-ZIIR(GDD53N{{(IL~l>E?}p~ zUPmg-yfZwSc}WXKZd`k>){+in-n%Iq9!e;?}XPKNpA z;r~4RO4t$>#nsT8?Qc+H{Ucns$8&Qp3r^tdjb|wk0ntNPfK&W@w#GeI-2gf3o*7)JsY*-rZ7U3Y?_Nf2oK@`%2J6c|l5{dIlID2^%qt)GqiI+#kvO@2-%pf0L3@94`|7%#h{ zbw`8*jiOS{H!U!JiG@Ke<%FG>epq=%zXgB!36Kd1X$CC3XeiI;*vT}^4xc&xUUL9F z|F89g774zs$-5h1Trx5E!TnYBOH9voKU3l%ih z=irn=)UY|xME5KM-zB}BYSMa~)Ok^8=Vg0nSBvq=G+!p(s`a7gbs^$6SM#r>^mrE{ z)Exi4boL9H-uw6QgXl>d*B|c>U0&F}`Sx^giA4_=++qLbe1gXf1{br=m<2TaJNNps zFLN^3ODMf&9bxtXqkcm%7GiGa#CXa{DC8_P7t>Qo>o;hy>SA(lLA2@5u?$8Wu)j}@ zpIq5{l;QhfHDMAg{jQ4^ZMkBFeoRAmbK<$Z+v)W(XzBDD?~K+2L=e4M?iV+WNN}jI zriCBS$m-jek z-fbyyL{S{m?%mm?-7RP^Ewfh|+c!9l{dRSKr2gWV4|FOPv+P{=V>&^ z_t&M;K?KbuN^TSr*D%p24wy?0$IY#RsU~*X+8nh937rwQTS2zX^&&3tz83Rrn?Y z4OMIJ7_!sziV#(hlqB%g6irzz4yz`Np^m`yn}r@uK^ebp*Qb;xsE+B}@Y^NKzuKU? z%g@H549!<*O2ejuh~%}0bRZMsxti#o9Sawpg?3f1=}#V+sOXx4>-&KmRC9PG2hUjt z6Z&oIHctQBzh!%(P4|l7*2Ho*_oUEq4S-_uOSfq&LmUr=x0=`JbI7B?N6x>_c=n*) zbJskF+5XSCmZ^~+hh}q$zLbQU*S!!sZ;kTx0Awzy45L?YU584$%{fEcedlELFD#NAD?Rthd zE>h};1Cp9hBjkLK`4*i&2=!)nyBZnMc}-!csghR%`Jus5Yj^Xf^z$0^yZINd+eG&p zprISZie3BZb^+>69VIkAr~AEd%p5$$ZJ%eLvMMN2_bVN*0g5-w9qV{sg4$SqSyw&N zhOzapa+Xbh;aDzo@yMC`LTTzk;t(?#^~QW_@JzEuP8!y1zh7FQ38H;=b)-Cw>+zP2 z4l3-Z&nfhp$FwKRoDyD|t^cRD9&D>`Uczuu23$N|{lt4hZ9K2#^p00(mM;m{P(Fql z1v?ey56VLS?YCR3Quv`iQ}Xlc)9X<88k21r!DXn@HPNIcUmPbW3tR4V4x@H$#T%?H zfr#SUX}10DA`1KF@7!I!fR^`docJcc2bQ=vWjwE2qJ`S`+^T6-sO9j1VYIP^?}U)URD(W_OXF>=`I~bH zvK-<`iQmy;Q<~)_$;VJy&;II$JwqNJj#*_Sv!YW4(KK33w`Q2&SZ{6MOn>Kv7C60) zHUz66l3KH%b2wdZLG!zC*?WhN;FvFb8=O9Aj28Ax7;JDyX#QBK$9!E2qO$*TxY4%( z4K-D9c(7sT_sm|)=ik}{Q5_d+Ql*eDntC4lM@oYUk+Z^5d+zRq;r)U#xg~vQ@!6X9 z1WtVz)7j>okVZg@%rn(jJ~E!)b8-Qh$%+jAEHtf>G_O*D`2lCH^2~-toIqXfDWNa@ zl@LWXz{t9Bl;OGVtY-yv55yo{wWZ2xngC@<25Oz-k!Z5QS4!k23(A|CDkp^QMC1n4 zzb083;{NphVyR~WaF*Q1(_FyTFuOFGUOHaxXds>SJ-l{?T1WZM+-N zq&PV(c99<{eb%$tzupMM!GVtgHsPp?|F)&z`R`D&m=+p4PLE@tMsD}AfwU)(|Fe=) zhep@OfzT^HeJPHBNcX?ITe+4-zus2-G}13(c|r07tPX?> z){~?cd^i?XmRoY_VrWJ2=~bcZPR93syUez6b`P4L5nO&-`V+0}+nBmEgYo0dvz+qt z+-(Dd-?l+MPV~GO7?o2#Z|G%$V{umA8LG)3=l@gh=zYNBDc_2nzYQ^7ksrR|(D2B@ zu6Mo|eo>xC^ZeSF%`mX%4$~~58WGN{Gt1wSipI~pNIDh6i)M>w5?Nx%h`26e*Y%6T zXomN;hRUaE7?IycX1ZLDdMH18ebsPiIv7{jj^Blr|JEp%yrs)MG%m1}{98~Jl_xnB zJd(MLh*GLNCMg#Y?P8dZtl=!GJ2pJ1bM6hIlqY=hXB9?-_?-IlTlb zO001G=zG-9qxpBO_$^4Ing7Rm{&nMm0q2($iqbLN6 zXB0=B4O_k=0KKCFxA${b!*Kag!Fx}YfpSP#t~qEHwTU~+XMR;hL$(EF#}?&aOu*N6 zi{me-$ya!~sq`$GuKgk5ef%_znRvab@YrTVU?;8X?^p{o4pRMv8x@EMM=H0twxRql z3pxXz{-A~5BG$~$usn%=#RQVj2|+Zzqn|3d^9PQ}P%h5jVpY2in*J@->kuXwM7f7T zfw8`S&RJp4L0=4SXgVZ>%DZz8#-xe9aqkjQ;DY1kn)N3T?UE$;zu5X~B?2P-za~J;EodP_yU-%KFPr3b%7$+h- ziuZl&Ye%G$>fQo31EKHCzNna?_o#Pmy}?*JKdL!#Yx)ei1r4>6EZB97Q2G+#(|gBF zXsICfkb(zd*styOawZ?6Dpy-SRDVd|^0pu8IJO0m>L|k@982D7o{#EQIR0&(!HM}# z(ZsIjzM;QQp!qTn@=$6Gj&1Px;}@su&}67evHx!w9Mi=-epM0|%%B{dBeJnL7RO}P zIm#4Tno^Ao=w(GjopWe#0pmHQ={h~$bNnWbr6ukChfqNri*VDWuI*TU67xvHNH3o& zDrgSYY1MjwT5rxJ^>28N2*IS*aUVle%r_DE{)!W7`=m8Mddq-&iLz`m_Wsk@(Nvgf zMB^b9)HU?wu7QLm>fKz$lLQ#7W@hC7+4t@me>y)%ZxLJ}>m ztgg^QDTncX<`M!BZS)NK7j3 zIe$HrzLLI=OT_B=sqaYkH)QCz&Nw#JAWKU*GMdUsoR4|>2-;jDVg7bM!+x&lIGQ1T z7mwO^E4U2W8vK*Pe^prpF#RXLLuF2TJQQA3jb4e+S3=}#7KdJT#G=6%8g=QlTQj4JW~%M9)Q{gl3nt8d1%*{8<=OLs z{b_Zm{QMCb$>tBLpYukNVOBu?+;P$M&K?|(_-%U~78%rErmD_eNrIHUeX|@<525MH zqx6yNYtW@UyXClV2lVaUI<>`16AiwZzoyALhvp>Cp!^4Py&Q_AUh<`qQ-OH0XlT|F zaLm^u<<^-eq0ZljLscK@qS=F1``dTez~K7cxeGoY5oz0Y%e+$`5V3sE55v4B)R5iP z`R43hMClFrwUfUc#}*RzS*iRhj$=+kOI3J1ik3Sqc`psa=gWENQ6K7nLoLa7qvFfl z(aa?x5Aq&GA7>RFMn92*BJ(FXkDj`rh2xb^kA;{b5^ZNz`l>-G(K&Sii)I1but z#;&U0pmN{$g1nye{C+gXw7xs#u`I+?zP!MUmqJ8-+rb1w43CDYAuq3d2&)6?E0%Bi zd|~zW`>7v&47lr|V_W8x$4MDzaJ!{W{N}*_eIJhfcT>tzcQX{cb-q}o@Dh=PuMi2W zi?}uX+aj*n(Cr{JUKlaId%Y2Ky}oPhvxW)K1_fEkk^wbzKLMicN4g>JF+X5s4BM$Jk}?p#yQ>j*y~nYXo|^ww zPyZfaWzPTX<*n9ev}ZVPVM7DMIB6fZHs71rL~qae;h><(E^J;K4L$Ss`%A*~@dg@e zpFSdJ?a7w3jYIWk_eVVU!|J+PyM=|nobWOif3^VW=1vFCT+cZ^3llFYQIpkT$FovTj5SFQ#5n?+*H7|WHh>WdC?kI z3=?cZ8eiEaGoJ1oE8Jc%CVsYO-zF)4|-?C9#x=DW9 z$aB;${M}>!9V(1j6^%{moIztb`ya}mR7R=dontM`Gl+1?)bQS|yQq@*A?fOo!-(2q zwmj&K#pkDq9#$0}ZvoLZ@4`}q6RK&N=K7JYj|%?gMf6?#r|+NDqP!5PLj9pa1v2KI zsMUAJ@uEI9G_pgw^)qy!v1*>KPN;M23Ozb)^(Tc< zt|WEKt#c1xrds9mMv+H|a&&P+T!Hxx-E5jPn`9EsV9#dawClX>8CzXx3XR{e1B69SQY)hsG_mk zr5hv{^l_}G<1QZ3eu`sm(I@wJ@S&w|PtzPc(-@u`_;csSIT}NrJC0rANVplz2K8uP zE1B8lfMYe?>CpP&BO-S$_KV&SMD<#~CCWADQ0-M60bTJNG{k)D4u6z8G#0TsZD{Vt zaqT==HhljK8oK(yYfZ;mL|pHGFY3w}MA7zbIqdHdE)#NSA^#7h7yC)3Ioh`>t)-dn}ttzK!?58}9 z&S3G4mCMhJ)MuI*?wkHG=KlXM_vc|Xh41??+^h+u5Gs_RNhoPhu1!i&NyY|50}_RV z%q2pIkdh`zlV;6zHP7=r&-1u{7Aj_*DkpYQK|-sd>pzdG7$t$VF?uY0(L z^E%H$_AzWJdwsR%QV!HAH5xW9w}U^K;ntZVM$j&HB}YHi2?~oGf9tg8L65)CbhYIi zw0G?d-+QBi#PH{RqC?IrtbUyH#OsYKRtWC#OXSgomle%Bg^~$=-Oj9%w>WNtx#=P{ zc7B?e+q3JbGScEVK)LB)Xh=E)6QV8sZ_8bfDs$h*{{XFi0>+&VcwW{x0`?9;jGm{7 zb$#NoY3?<#f94fnQs1|JoAe7F=d$w|2p7Fk8-*Bhxr`=Fo+Nb41*!Vf90M z&iif9lV=-nxyB0?FRJH7FyH;_C$m+G3^Wn%bM5M=nfnUN=x=g+YNJPDisra*b>ls# zQI2{u9U?&DII!3koZShzcazl12VzO=&DjC5ohGm(((#sQw-6fWe7Q-9RDfX{Hw7;> z5vcwXe8J4-EY26n*T`_`!C2I#*#~8``dpand#^S5iKa7+owBFA<=$f&qlulg87~`& zDSO%MP!b)#CYD5(ok-gMA3UD@cZbHm5&G{9%HrGRljCvLctrK;%csz6lxueH$8u~Y z^{n~0*9IDVA6cmFiGq<)`!c^!25iv?WHV^F2y=O*fk$7^+g+VPTQ`fPL_@!TdBdB@ z2D@*3^_0g_W8 zXK4SIRoPrr-|`jOI`!03PFg~nq+qUTa2uBKU-xB;{|Iv}zhgpL_t8CPhurOwIA&V@ zlEmP#C+jQoC0e_!F$VXHTuMmHrYB@%%)POim$L4m6WzW{H)<8eQoG8?Bf)`K!y>wk(vMYh@NY$Y=0C7P_b`ffi=#q5ZDdOr`QS%0q2 zXZ{4#akO~5d=t!kTd^}OT?c2q!W8T_mVueM_3Po)#CkN{bNHU%fAACyZZjReCcukj zLKWgg{5sexohY$sX_tOmk6Dz>=eb%yeFA4*XiEij~>FFDlzwa zJK$u69o)^gM7`9YwHVETX? zGG`}-{(h``kAN7a%IzetQOYF7r+ z=5bb+Jlv8%)T_-%Q69~jzk~*|8)R&>fJL*}`jSdb7;JFNi#n$PR0X{>cAx$9^Jin% zve_^vsCXe{sJ8tEHhp`RD`jvB$~)#+7b*9#Em}{_X)A%x%s4g1T-7Ry<<|$B-a730 z=lGsV{qnJJZ6Imc&3k2Aeka1T-r)R_PsCsUW33>Qh9@lADJmWmupzNY6u%B{^MJ{S z2NMG>-=J$qwsJ^Tn#8nAnmJ|k4@`a8V<_r-jKq-p(qY~zg0$>y^ohwcR9NKnf0HCk z8>b{j{ujHOyNUYm36jtene)z2`Q={KZnno*QTFbx+^i`qyrDGj%!?;6EB|tI+mH)O z3QV=}=J6z!OHO-UNvgmSr}o+160~^;jg*T&u4Mfa$46qg)DWn2mJB0fCVWvF{2;9% z*x*wD4VR0pYipCY2{Ax3Me0J9haRzy-HFm%;wpmqEsDv!%V_nvFz-<;Y(RE|C})%9 z58{@={G{c#r@Tb|yD;dJzgbrmQ1_d!%$lW(a}w=H)b+%LEzo`Ms-HXVtlv)Pf0V4u zc<@@DhO>ia0Rx6hrsg<2E+Nje_Z3WY*c)3VJHXV-!{gl>c422l57(gHHylW8bNklB z4lNbrh?MtI*ud*8`*q|9%v>GGUu7i&9qYF}@%N(nxfm5;_0rYp5{?(G`#R(+1%nf} z?*#}G`&opNY}Gn(y0`=F`rKz{>#1K6dMZ0wiiE%Z>pZ%Cps+YSEjv&1tFhA3;hIGJ zC=OL!`5HPy=$%y7ng3w0_JqL=w;Y>&i1=Jh!0pO`^ZnSrVAUuwNUx7H^zHNP>tz+h zd?fX;in$71 zPsg6paD?#wN50|}Rz;XE<#@jI)D=5#r%O9aibJ{VqZN%8yMUb7aQXeII_zd~bLXy> zz+#1`Rd3tMAmmIiMD89ZvFw*r+Z~#V-wJu2vJ^z%_?MMGgV}naeQP0Selmee@|a3J zl(aRPlMa8Z7>U^F?s%|ZguW8w&)Ix_l#ZDEqZ7E!*pvO>J%{cu9 zcK&Wot)tNDP)H26u4jA~HR<}T{UB*Xd6uX%Mail*A*NGU8Pw$Xnb1MToY95)n%eca z6ykU4>osSX-$`vJRrf$=W}IrqTYn_yCyXYVKEc8DB7)|421`oUR9>iZM#?i5lUA?q zfBhr56pf4uY&>AG<%t=cJxgLzOS~F#{WmuLipyz!C0dlJj5f|DU$s?hPE z#OZqk6O7=+t=iv3Fiz~uhVboMFelp}rTVZs#O?Ci*fTEicYR%%Drd79s-<3^Z1x|3 zf#BjRd4dhlbH(^eNbE}JJ!g=_BS6f5%0chZMQ3ZM7ynu$eX9${1SXa*Hb%k3VE+&K zOOi0(FH-)-`7{h}p4)xrz$*H6C-YhrZqXny@I2HXDA|ia%uW{N*N>2xMF%tWFZ@Dk zzr?A@EE{5fDAtN@XgUr<8!z{+BIYF&LQuNeCJ;mZ!0_%A?bkU3emN!N z^xJ0>p*VkkqscGJc1SCaYMpGN{BxeBQ(8-|3rj*j>jtkad+F9&Ag3B=-zOi1%JmyO zsp6}kkK-9BJiUtej>EC_Zp_zkBJq>tT76aM4QwsB`)dxUEBzKbmKDOl=)zj*u4L^0 zwyS)8w-I(SOC7!9AB`Q${+Lsw3t^rs@!b!*m;Zde@tcQ7E3{X@XOYP@JREsYmvH)` z?JEYTF<%@H`!a};QKd3zr)a!Zn0Rtqb$B!xx;K6otTxpp<~zk@fB*Rz$O}p)e^kiA zAuHZUx5{JCIaanq*U*M;eidchiJU{;X52>0K8Lvq_(t@s3qsnu4R>t`-b zf0QYb{Imz^S6_a$vFHhu!O+-VvyIT_&W>JOTM)^MwYPt4KA1${(#czI@D3Ts!feu( ziwh*8PB^+@vxH8C1eDcmoq75ny2eWxT3;x7Woym@JLv+xSf#C|2*b5AGm%uIr+YJ(FdQfCe{Ho`#%qXc&7X z*vh#FN{ZL~ycl}{+qWrJiA07$_Vc_85qY&x(pl*+<8cJK*YL=xDd<2Odr;=N&RU%C zm1_94+#Oo|(%%{Hcfvs{ujNr%0pQ1{Og*w_3GF^(I~QK9!;#}>!uqeWLg(ER{Oh)+ z;^g@lN7YV4SiB-$5&B6B(^U*Qcx+eT_~x^5kBtf3eAU}rxrWbu(4f*8zf4L1%PJ<$ zHr`(eO?@Mr9!~T+X$>hS6x^ISa5!<_{;-rQP_@>~f5=0IcKpj7nN@awzhiyVtF$43 zcxY1mW6>df30f{hS)L1!#+luBM;cxCz~aaikKa}MA^A+C{97hs+>_Ps)OcK{)mcOG zdY>PwPe;NmQvbZP+6_J5zKp~U(eU|DjS0CtHeaE(C8%%L#v&Yj_2b^{t(&oJm-ACc zpOQ3|*^D)Hd&4or`XX-t*$Z1%i=>vCox;+coiPV)Ri-hXF}y9qKMO-`lFvFN{*LdA zOzTJ`ZlwD5?e9GG0Lrs>Z%lN)giGXCg?l&JV9VvI7nWIXv92=A)sxc(=C^&EUUrsN zClCJRH@839JOqQT)55VYUjlVq^Rˊw4Ny!n*rH|Xm-=J4pd4@@Yo+yB(i6_Qny z^cRgKU>v@0jR_^{XIcimS5z6&;%}s7YDYE9jEgWcXYcht%7f6x#zkXe==bFg@{QRe3rH>S-{Gr% z07H1vYWMdV0h#SJrd@L%He3_gw5H zH_654&}XQyj^*QL=#Ay~RzL9=Iz60S&y_yH8bkZlN>3|c@|9|LqN5EgFma3{VSN9ePbx_Ru%ST*_L+J2boJ?VY&pbhl9*#JdAy1$7tmaoF_ z$uV=a!cUO@!y;?bv0$S%e;(_c`qn!A5J*3x|wwrj=0!3v#<&M%} zXxih&dXF~^`rb3Hm^S)CVtQ_=9bQ@j)vlZRIHne1$SON!I>!_W{oB7+&Mm^!^wpeP z3J3K0c)pjbpyA3%%=^!piH2N;;Zs^t%lnD_G_~bDOL8(X4|~sFdSLXGj=zeo|LV6r z<@75caPhrSW9vswaAA&}nB2ocS{$3iS{Q;lNg3E|Ir!um`yfz(%&77tk;Lg;`u%rr zHYOd~)<_ongq^mkmClK_Y3z>5FHVb%LGr`%)$Rx6VN!IZe> zUgf#%B={7&7WPPdT9Ag(%=^!bH;Cet{PyD^R)fU-B%7kAk7ppcY3uoK*9IWwwy2?0 z&K^wtU9+l8q#9}jr-Ym)1)<%@@rzZB4Nml}RXCj03ym{bvRg3`N56Y6HhB49M#qEw z4XbTw{X0=+YIrXs6eoRlC&pcTg!Px5BG<%R#@s(FqLd~Yt`lqc8uy;|?!*{#E(xdb z7m(KM8*=o^1(*`zJ*3_20iEB3D<|e&V8>T~^QuWF?7btmcm5--?gQrSt38dx*|Ffl zhoY%ITD>8Z866jORcZt3->r_iicT*cN_T&7_t<8GHSWJYJ={=-CHJ$pCQKgvKkaX< z++cG##DfLfm<=ZK5@@>jFu@TccmE!d9~tiL%yb$d>bVA^WK()K(({^9Pj`~GxDz_Z zL%&;$b9Tz^;>e*MAoBmyueQYd^QLU_&Ma%^HIr2oc`-ydPY43t*E(^gIP_{#V z=s_@nD-5wb#rvV552`y29MXR%!=0yRkGbz4;<1g>o1=Yu<&f-iTk!MB1sp3;p$;XJ zplt()GV^wW_lq;e2YCq=x7$pI@XmCJPjl!A$Fb>o>#H#%hln{zq(-R z@`d*5Q9{qBTkRsB&<-lpxVoDMp9+PEVaXNsSB>bNBmK?nVsqsnR>!rZUf%Z^yYaY3 zT)%Z1%dvCI^F0YaYx1T>!mHAiFy$8fWboN(==<=K>$}Gdpu~jy`90VM-G}l#LYD|U z{P=m%T3tsP?f|;j`A&!?j^M|ee~PUQEg*f1RsJK_UTCr4xgJeD`OkiuS>18M;1LfD zeQG)p_s#~Yo^fEh+JPa^fc~9f4eK_`7+plQDn4k;2y@>BvYRy=T~-BE2@bdSEjPw) zR(M+tpLzZIoJqAy(_;@fZg+OITR8Hr^6Q)%ZP9!uEyaMq{pITIy zMMw-5YKk|H)A6p67-Ey1$JNJZ=g%31)SQ)^q3gGv+)u4;95OJtZih>KUjQ`Nk-l#H z*+JJHb49e>%e~dGun;-n$!JGnioOsb;wJ(nm9ycubiGK7elzd%j_JT;(}Mi8X)eqt zmM3QE2auTE8@KCRk;l2>@z+;(*piqER<7=7btADnaE`LlIt7%&jax^qX~I}W#N)YY zn%@Ouj&;5lHSXifG!dSQ61$*lf5m1AN*0MxD8_s1krZeeQM}9G*#{G27A#gwjL@bk zYq|YmJa%quOZNLs;!k4x%Dc}*mWmx(RVvKIEHFFZ`o3my z1(XRh7p*?{9NOGgKU&Wj0AtM*nfaSEI}$6ohU`;W2;4{Q^~2kSHps)ko`A)+b#2f) zF&-u{N8~Rjq{_LrSZ5*S+Jk=Y6?!;7xyf(bJr^dX#ACLn(c(NLwjQD4>=bY8d$U_g z{y+>4y*}j)wKqMg1%Hw@3q7CU=ELa0-d*erO7Fatet-T`c|5 z`vz!H849#aFozbCk5{;s)8bh`oo-aN+)32QQ7tR3>Q~3Z0NQ%xS$%=U$8Dmn?;=PH zl7stBC!T_yWW~E*XQrW;DP5ps=W-bHZD3yQ5e)rd4^>~Oo`b1Q(&An1mr#3f;6)g( zE{VzGQ*d!`9f@hf{Uh#%idcK?{E^gwTQKwCbx`^3uP{^6CUJco56p*~dY%1F@H}*v zM9q5?wZfR@#?zU5bYT8D&n4+&Ctz;<%rWw@Dd=Wa-ui=UE7q$Heqp#+4f9tTk81ul zAu-SW{#cSK1XQc2Zw%*1NH*H_{?BzCAXf_dnNEI#(Ht>Zb#XDwn5?S&)wm=^=CU&Fb=M2_TmMK)N(^+3o4BNxNQ4S;06BtWOlxCLJN90I90UY zdqS?pV!~iB1EzIU>8r{2!h}N6&JGni9Y=^)95VQGBbM&>3A0CuGhGBv)O4|YX|)&) zFAwC8g{w9yj$*2#EXU3q5xVQgDmg=B$g(h=aPPo#=N1y~CD!Lfh3iIU!JM_Rc*Vn0P*kaT zP&|UjkBkXFOH6XV0bLU5hvl!)_)T!6e5Z27*L^@COM`$BA9Vi+HhW{g2AV=TN6)q; zl9*LPg3EFYFk|zFgVm=7U}^qY@-e5s<9J77c86R9P}W$G8@mPme)r2K-0lX>iF+-3H07Gg)!G0|vwD~EAdAo;&pAIRb|!NeD%Nmvk~Lf~EzZb-$=3k76l$R3&C+f}@ufI2o`e;mI zRy^Appp^yG?=pMzRRQ|u3n>TB1mNe;H^2urP@}AH>}oxMXJ5wV;PUZ(1yrx^KIY3t z&u=m_tYvijE5YZ`zw3R14X-Ou-&O_&1swb5xNP6F^MuBK)VKEKJia;IH3M~mn!UnP zM4Y}!UX{f$&h{06=Eb`IOFevb1uuTcdn3$T!s+OyK~ z433KV#7CLa@Ou!;(v`SrTM#zNIdxVzt0P(S^2DNi2hMTYKCSRf#EFZargt%!9X zho*Ecnjee9CYv-A&jo@{=`E|&`R6cMwnFxN=T9I9ALlvx&>n|zlm+sR#J>Znyh@t;rWsHa^^m9)j-A-hcm!o!J5U z43~c+HUCKY)AljQpO5^~`iO!py1!mW4Dw>*MRysaJ=5^w7klf*mpnKo8mpZ1O9$(I zyvrLu-a%sU8zYbHwZwS8z6(5i0H*Uqw&i@J*%8nga^fh@c>(OX_egGMK9QGcW6w1H z^(Y=FL&GCl(~DU7ZMi{Yx;u`zWiC>ftbsDynj9=g@VsRfN&9*000oW< zo5p>CCHGHpD#vx8K=54FlA0V2yi~8e(c}T$siQmDepX}Jx9LvT0e)IL14oN5=V+aW zX}detTiFldLTt^uqLlx@56ZJ_;a?=;2-MB%A{#CL+J_v!=8q=R<-@R-joo=--7Oef z`C7zsqB)K6u3qO2*5x=Ba5LyoU>lU5y3z3V;B6A4(m}tod>c!?hmsgB%}C7joWQzQXC6rfX~Tj~ zyoX_p5#+V17k&=lz^+3(`;=GsK=-r&SDAG#RNa}tmn`ovn{8I};N2|fU7@OSCLtDv zhk3$RnZ1U#6P^jzZr^|j4*s*#D;J@ks#CwXXBXB|ea{PXJyid&>Iu^F$3H1i`5Ld(v0Cg?iQx@j zm@B>f?dRtvXvkfj2RDL<=Q(ji-s6u34$rF&AdOfi~VmK_fzdBeF`gW->&>rWPMkmgh0A6b zvps&}xkVCGE&Ux!@|+l8rI)Z+EC^HQ1{Ah^Jq*KZK4>025F>HTCACM>J)D6JR% z0AqV9xY)DnVAOsgFd{(&hCaT$9^|P5HOIfHe2t`E%}RP=pFlxdjIV zr<!?(E)-zGMdou>#cWAeGXbz6rW7G?hNHz(m(ckYD4MD z_c5~9^njYC74!R8Cyf29WXmnJg2t{x`9iEWfO1QF*9DRtjPUa9T=k&}l9xMK8T`_P zS;eW}(z))?^VqUpz?+uWfIQxSGKaaFKy{iuTAD=IZ^L08{?;3-pnJ~?mS@(7k*wT) z{Kj8@?f&v`mTx9>x&gH7B1k`G6;-D(n%)k7(RuxK z)N(IKKQ>3U3#fr1DX!-`Uyw-*>D)HLejBmP!20^RFB;JPc6O6kWFwTE{2FU%znmDK z%kxunBY(rN&g#GmbwH4+N8)J zuNpH@TRI*Uj|*8x%>o@0v8!_NOXrP1q}*qh(KsaoROeSG)m;DD-48FP2zWRAZC6@= zS8xwBi7i_+=P!eIKIMhi6wg7EaNW=F-G#)ubX&u0y50oHjIF7qhc#gGfyPONr6*9n zDp~VOj~-ME=1nb({)8;PtIS2L$DveO?@*xfZb8vJX?a&CbhEAiZz{kgE;}uFe zJP&giHeeoO!0M>aN-z}a{q$o59sfAY?7P2Rk#!HI&A&Y}=x+#%>o19&FntfjH;>=a z)eVBO?51Bi&MPqXvah7h6EWzpOWz)4)CJ?G7ZuzN(()=m-XAliuqF}4NZz77%uisj z(ydUo`vI2Ud2_>M_$CHU#}B={K9VJ5Gq_5%0$wG_)s-3)|aA#8KiJv|B6h*I&%!>|tWU!C~){y{@aEtR`%>`1N+^ z+;G&xAT1HdZyt}2QS)%-x#{`gH!6^oJ=VJ)h4(=zuSC`a5%#R)+dm>X3a^72cA>DJ2T6?SVd^2(6r>0v?#wMbhefq6U>;s(Qa zN%t;5g`(Mk&#q(G$+kV3$D8nT+xY6==*3$=WA1IXr%m0^CB;{|^miPU$(1h)4m}JR zwvr619GCv@>!4pqDYK}h4u@@1tAp$3@#wvK_Rml5!h$D(l{K6jYMn}~-} zHnYq!TN+`$ZS`7P&RvkaIO+1ussX!KrOA^1`Y{u4R^?P)<%W( zdroalBG!=zCP~-UD9p=}>3k$a=)@P@iuSp!w-KhwZY|!3{6J!kQY>&5KL`shgMy-E zn_*_1zPLie8W?S;J1w-|6Nfaylg|d;HTe=7k6K zMiIgD*P3J+*?)F5j2^$~t1)T;boi>chz}Te(JYn||Z} zTn7mM^^;~Jqd_uEG&m@bUjQMqs|`evE5$Cpdtn1 z+$v8hB-a8(B3%FZk1<@-SBW(2YQZu!@&l1&#C*&MYFqL8STSb2{e0>Eswp5(K1{V( zw;tTrMbzzz6@zXC(GBjUBUm%xw>4{JIf=RRapg zS3ynBfhRfd6Vn*<-g7yOX~JN1M9>xLEnZgzzP&5#Jw(Jv&X;e$u(nN+d6s`m;WzX)BvaA{wKFOIGocY8G)1r?=kpOPL+ z54RpYurNc+P66_egY+4=uf&gw=*mg2L7x#H+YW|e(I@chzaOIOK!f`q2?6G z5t2)gaiZzskeF9r*D`y-3%yd$KPz4*=3PVFb5i;E5X?Qh-Hc1}knOLoIK1FSyI=P6 z^;g{o#h|S(*YZw;I#4N7HpPxkFdiOnbZ^gE=rPpyyy$1|+I=&M4YO-oUOmAW# zig|>B^H$Q!z#Ei;D zc3Bdg&S)CrJtr2IT7nO(JKLz6EAAl-cE>8r7&1Wb9hvnHbopVdOMEKqEG@5#R3kH4 z_Dq^S17vxza~z7M>1;#G4Y$`t$?iym9o-HqJ3zr184I3}ss+{JsPcDR(7AFOguf)N9_- zjHV-p6|&#gt{gUlo)WChe;$itKPos?lO59-7mv-YuCB)3Kb6j>Ty|llu=@jr9xdq7 z{LZ1IBo5T1E{BGzQW!R7@KS2e7U*u(uPW1})pwFusy8be9f*Ov_;$B<9pYHClJV8| z$wgRHX#YrEor=_v9n&c~w0r~F;6>&*2?p#TOAq6o+V~HX7%wU#A|uPI&D*m68gjc8IC(+QN9e z94Fs$4qVbbfz97pK2!_RbTFW5e1Yc&+e7G8vyc?aK15=w3VF`hd<{Z>m#<2)T7%6N z8+#~=f;gyTTVIi%0-jT57sC&nhTcP}!e?doW1w)t=L1Kd!Yj#_JA=&2pw#5ari5VH zJcseOy0(IEi8_Sd4(7|>B_m)swrZ!LWCNDU2pnEN{}!ghr!=JQgyWdRpndBZUmTuZ zWnj4B@A`aGyx#u?fxB`wo(M7dNa%78Mq9oR%IraZl`&)tJ_di(!YJ=_>@mIKu&cbL zKMWaC#6u3hz;1oVrdP~2V6y1v&tj@Ol>UrSX#9B{TAxkvZ^|`=Sqt61-!9fLH+1fK zK%O}cJ!o+}Rz|OC8V+|a%{P*8| zX~tbIn9!ZAT6yF3j(u*>qSx|W^9YS^2D^{gukp8ar>P=f?=P2 z-CZ$9)XPzRwX8Yb^v`$-iB&5^9acTTZr4Zoe+qPU1hZA zC>Ol6btSVe{{GD>P!%J9i`x@KG~RB60*$r8Lj6P?W_5|ElieLtn0kKYila+DmL=c$ zp#Fi-i;J<4agl2WtDu_PbhS>@9d%xtE*9e~vAW*11Lo6Z~y8;qxoRnSMh{hua}g zqw#yeSpBXD|IvGY?f;{{?6+@&im8GHmH}?aTF-oKU6LhsQqL7k2G0`X@SDqN-otdd zK$xzW_scx&-~7Pj2X|auJX%T{0J7uqqRfLB|4$7|@(Q4kU+J_L($ zrEU=sH!y?wqVdV-2+7@8}t`~-5)YDNc{|G=@m zRMFZ)qQ&)~W7fNFOgJ`;A!GiCvZf%6w@9hnHwuNm6@LvpvNPtraTV~zU$@YY|~(%(3Sm=EL~g*X14OHa-CQX z-R0IMj{7vACuGdn@Pj{eO@>XZb2P=#`mLJLwe);+9gL}Vx3|*qJ^pY1!qmXMyGzHs zu>4H>$Gft%kSVLz_4%|oE-fvz?XZ0V?I&D!Kh7cQe90_twr?1)d3(IhpJ)4ViMjTZKz=w>#%qcAOXiSRy0?6*Y`BWG zej?M~MCtX&CN;JDpI)NZ7s~s1nHpl$iJ#U_gq}@Ozzm%Xfzn?h+cL)Ba_iNqGvUiP_>h8x39SMSV_<04qH zPw2w2Iv;3bTJQh*sWFbNa33h1`-G8aZ&<%!*M++|Fui4KZ|3+z=x9>DMp+`_AYG1+ zES6ZNpi8{$mYP%t=H@N@s!Le```ovvSJr`sE@6VX_igyLa-qHU8m{HQT%q!Np4`u7~3Sn&-i3d;H1HY&RpFC z(Ddo;^_?pWvE#|yIoz&>+W8o|Rz0Bl z1;;}5h`!=0Lkkjf<9!eAm*l^8%%c@L?`;0s;fFe3OF6;B#{3hGS*NivOjx7tlg2;i zZRhk1`N$q)7*c=jFCZlji6hVhcq{zK^q45&q~{+~7F5c8DwYb{!svvX+&r zHOA=b5^&0yzv02UpZ|QGnYOiC^4jJx%|-pxGuB`vANf7;=Uo+aJM9%dvuy>5y=sl0 zfK?~V+wV|rk|c13L0_$2?`x*t2h*0HVreq;fq{ar&ad`ep}SwGgxCI|U=5ghv-py2 zgof{jM&PtJ-^v4Bg6hTdDc#V&3ZKV$PhxvdwCK_!9q3CC7|%U_38QuH8>UlJp{2<} z=Q8Cr46;5saxv;HG*oynJWAxjcHi?|+ROnk{_M8R#X})Dp_TMVR_+LPTz~(Jxk3!) zZiOA?jHBt@;#jpJXGao^j|aPhYP)v{NP}T@*v-KU|G`g?dMSZil?#2tp>`o$$ynDo zdvDW~7uZu4yjjhIh`$s~6}_#usl%#>B4Mie9UQE^pjxE(0H>BDT+TD^#HqByZBH)J z;lN>F;EdzaYZ0LKo(*{%8G{+^y+_9PECcefc&`I~WSCkTGg_ojS; zC}HF5bKwV}hjpqzHG%^JqePR>V;&a$()*&d#T+}1M&~nnabcH`h=Keg7o;XA>0K87 zfwk)4uPWuzu&3SiY}>3obV{#E{_#-;`t{8h4pGmOm@IdMhQ7IhA<3O?6KT^B?|g7T z{;xw-qMw#6CZmP;Mxl;+m1eKNIFB#?Oa9ww4BY~Yxf(RSR2azVb$;7rNee7XY-{zZW>I;R1Lj-uK=9^oF1#2PGT#Y**Bxk5p2`zU!K2u6?8HAt7Iw+WB9Iz zwZ%{Q!Co@1^714ZJH<<#^hY;gdHF+KdG!$(-a8Q35y1}Q>wMqIYhA;-P9Of~)zdgU z`ElGmFcroXKHA?eRlxW_<^;}L=#EedX&5 z0uSIrYVDRyZS2@%_$MpCkm%p|8zUw_?P*Z@*NNiFG1$`mT<*7NKX=$J{Qw;cW&imA-muros;Xm(o7}deKF^ zXX#O~oiBv2`Kzzhqb!~VfT z&0QLjIc0wLy~u|4(B^MJ*Sz&I<=5EYmZVY}_G6*??cYCgAyv4DeaLsEuAEBIm$N0mXMHu~R z7Zbo!2W^5kBb-85ph>%0d}Gvl=zGO=O01Mtw}j2YMyBSOL(p&MC^wgV8JqphW$@E7 z3>D2 zQ3bU4j2a96YywKb=Kl6-`g*!T%_#AuwG+@*sCYCmxd+oH!oELU%Az~3oFX6XsJRJZ z!u$L*cGN*b=aLuaZ=!B6`Rd-+`%*+<`twm|Hf|d4Jl`lX?-yw{(? z{Og0AGjrmI_v1C4)7`OusFYhcr3eR&j2w#}(D!3W@Sz*omPB1fo&U~T91eC^98$WQ z-E|eTGwL|0#8a_m_1;ir(IZ&S)*Vx;WCY|L`*vyWxQ6|WRI;d>0n|KGlGv=E3;l6_ z4t~CT0Gsr(7^t@nkr=vV1Fg>dKep)8h7?+jEK_A$nbNq;nMQ{DCRE zh>_-5oW1R8mwj$8RGWTj+j`CjsgJ@0m0rKY!TI4(0d*tj>^+`Z~} zkM~C$yB_mew7e2?_ZMvsQTB&{Ed}3p@HJyq<;ENqS0$+Qx&JXN(F0>D4ZDJ>WS}~* z;&4TG0j-_!7j1!KrC+csJ%A)3^&T3f{#?&3VuuvmJ?H8!0R=*Fs<@5N&8UfN8tq^G zkam5pRz+UfpYyQbp;fKdG>u*J<*mDZ=-H^L`VC`ga7S!0?M1p7r7bq`Ywa2<5cB7+IovaG3HASlYN&-8`|5_-_1p?v zx^_FFA{b*LO0esPw(Ylfk~n!IM(e~QcU&ws-~8dzd1xG{f5tH~@z49!l7*Ta+dDDG z_{67ULwk^XSTTQZI0GhxGE7}&8Tse?%%s@Q>zVz8n#Njd=dHV-Z}Z+|2dE?%C)=Kl zxqK6gbW+bH@~+0#!=lR>hiL0OiBToV&?{UF8&As}J*0dBOSg1So;7j6(%pgPrU!T6 zV)YcyD^of>TUh#XlvhuL4D)?UUryg_fKh#KR>fkXe!Tkjo%I3kG+Z@gTHNu>|8y0o z3orMUu3C%PHw4bz;7W&?mb+b#bv-aLcnyHDcAK5m^6BxWZCf19}#4vtr_YB&%PPe$Q_WH5*=;mI>%#^R2_X ze2rnof>$d*aahqK5Q%GaszuFJ~|N0jj<=^uuxf^wBy4+j7IQEq^&+?ubtB<-L~J#=PF>yFH;#*YCGe z-Rx`$xUEgyroP%AvMN~-sA<8@4~VD208l5*>LS zw0he#X6Fxas?*nDXuvT#_$c3hdtadPaLW}~((7c@Zc^L!@Du}NDfpzDHu-{U*o&u} zC+OlT^y7r1>trhj-M&i9hm_|phVRnhjc`D|D`mgZ5R^aFkym!%m4h2{cX2r%iCXKQJu%Cz{Ps#|1BBhaq}QHOr^Wu zidqdF`_z#9{0a^(g>4@Q_JU}agWDn&55U;@G#>#gy78OF(#a6bZcFf{w#6R5a4usv z-F?Zttil0d>^QXe<&5@(HBxLu-1*iW#+fx~Ark?8Bo;wI;WteuaQ4hwmDy1GT2jI|T&a2iTuU)bf#T6Y>3j57}DiVb0VN8)DN_+;oX zeW4)>vMQ73nY4v4kjK4f2^5xIr6o=b&&HNdhPud)_*RtZjkFVbsS55cP^A1;X zGEue~t;N2sru#d(W1)%lowR8Q!4Lbonc-bk27TPHmph`vppV$6YrEJ2Rn`R1oi{Y) z&hC@30=gnx-Jab2f#tkj*DMZoVxhPghjr*_X!E(>_t8fd8|GZ@JYf6|Z44Bl<-wdt z9b;EMY);1q1ucPbLm$rCVvE`1x@|l;K-reO>rK>WsG2#~DjG}EHNw&5f{s1&Vo(rj zajMpT0Ooj)=UM!xgKM7rzh73 zk&iH9ZGGM4_kZQ3f%#vOGRl1a_%T08_BPaIPS znH|Ua7ILM`^<4O6yE{;!r3l*ID7xHxX#pMbrpy|jY3m*~+zHPq*~f~Ju6&1n3Q8d* zfd3P_Goe#h=Ttd)&UzAa%NNI;lRMOy z_38R0(sXq7r#q~4?-5&;Yh$KP<5>cV^vDkZ=QU6}FB1@Qg5aN+>l6*!acr2bU2?cT zNz6MH3Z^s|*i8u@ZMz_q#nJml9sv%xxJzcXn(%TUQwAs-zThP`Z}Up3vcDopkR;xoe{@vmG#F~{j#((0wLZ}>!ij964Diy%?QdgTJQc+OwusCBu zL;x!fB&3|xxC4W0^}41S(y&>rP5WpK!T;M+6?B;VW<5|@752Ywj>ouTUv3ro)9L;G zub;1ROjlGPm`*nlOHwzVt|RJIX}_0m;T3wNxaxo1zV4-4e4KY^`C7~xa%QuuF2dor zdiQUYeSk@xeGyZ7|DjiSE@4SDv6S|I&71Ok&rcS#_DT)Jc6-aNl$BC$rOU@ER{^)&~Y0ocPI(zrO_Ct?@*=Uf&w*UEi8uPF1gC8%`c-67d)<$^@ zf8(G@sYlvz3G6<^roA%P>!0sA`12g3Z#jh{F3R^s)XnfEGx@>X(KPH_VJ4Zw>XDbmnV)m@AH@&BQa#hCYOA3N@Lx7iJ`eh96DE-ZQgB4@T1QM zFWuml`0sYp=!>g!$qpU=>wcPNa>N_)y#MvQkyw*=%MEGk9f@uEG4Hj;3YfEG{OMlf zLHd2vjLnOYqX6pzW|msXB%F%Ry?Qe0E>b>fwE>2~;tKP(E=S*D-$lxe?;M|TVx?bv zglZy|Dv3OgaptFMuUk8s-&J-98u{Ov`@EsmA>w4KZ*Y-mD~=wo_FXo}h-1mId!K5q zgW^~Io^M6BLWa=eQQhtNL_daQ74L|l(@X#V^+>p5B&L%k;qVK(?YPf;*Vv3>Uq^nvEeR&EbF>*N-8_$7nimJG7uunlZ!FZ! zZ#|OP#?qBOx8t8}Jjd>9y#jJ@;i#6$X)I6S%3@^Cgc=DF!{Ez#=#FfXf8yB9y#6;P3bXAZ6)E!!F+&`y@gOUt7RhIMEVK}!a_+k~b-mo}-b6g3V z-Hs1_rqIQKu<3#r#ZZGS44eApw;Z*onMBj7*Y89waKs$X33YADBL9Q z(jCV(b={Qk!>S|Ojoum4#Yb>z(*t4q;Ua91a%&ZQ`vBWMKJz=gVG~rXc*?lq?SJ4L zRcn^bGIBz>fVTa1n`}Tk3De!VHCV~KwMBfbKQ?77-V!`fhsC02dlL+jvEi2P_vMk7 zXxA6&xS*F=26l@M-IdE{haZeR14=yHfBR``_Q<>(mf!L1ynJvy=4hLBA0g4|^Du)! z;;Ny&1`hLISxB7egYmD+cC9-58=JR1N@BjV1>5(iy^(u<3_TO)PK=!mqX%e2CO&Zli#>)FHCMW%-9)QP_bQpsu22*9nK8($-${DRGS00B(Z^AIjA)(9<%mV z{a}5fhf&#iwhWt%a3THAZj6e>{?M}m*R(US>iYEiJW9^r@6M`3mdL^613pn@Uca$# zuGwNE&wU&&y0KdESQlE42>S+=vB9{`tHagCPEfJwJI~=*VI;>1d{{0#g<;RO%P#dd z;jrIa#M;Sm96N4U!l@XDWbUt)tg$ifJ5nM`m}h{-bf_F5>lQ4jTc z8v<37t;cR|qSFh-KKnzfTCO@{@%k(8^t@?#Zj6&sYW_P z-g~HwJO1^05$y00Pudku;4u3f#Fpm?Rw5<*D#hqU7nI)7_vY8&z=5LUgNH58!FWT6 zoyje2Ovv>vyDJ)l#iuf)dt7v}D1>6e-kS`W-$|YmA6hY~qH#Y`6R>aN{h8$#60yhV ztHhi?tuBGYu-zee!t54_=_=GQrxXzPFYWSrBD@zv=f};PgJPk_Q`hAD;SuOAUqi~w z5y0H;_j`0BIkDthn^dyV1*rR}8Rg(`9D3&|VHqSM&)?NG$m#5G6-tCBp6N}n(DmEU zeci1WWu&1sk27rlm>mvZcDwTP@l71R{N&uB>h=Hno=C3Y{qv0P6U;m+l`Zs8$G+E( zI*+MxLsC#6#YIC8r!C%XZc?Mm)6k8JfzI9ytlup$MIq&r&t|EuakGpOY^EE87zUKF7+WWS| zr>NF%3VApJN}F^qNQ2;Yi-fe<9}$ z!OxV=(l5N5%Ma7fx;s+NMq+F1=s8AiH5~c!tA5!_I-G18tJ(0i(MV}1B<0s%PUvh@Bc{nGoAZ) zMe^>Y)k$K7VX)d@?P&~XR^7lEN#Mz+#k5xQ}lU|$Lq5Iwci}D-BhXk)#-@3w?5Grk+AbHJlN{oaB-TnKjWNz@S zr^}P2F{XZdQS^2b&JB*ROt)2~F~(n5eQSuumxZG{xn`C2OF*w%$a@dr3g{kF>bY`- zUT3aLcv0k17F}HiVPAEwsmQ%c$G`p0{_6dHdY6e|0JyseZdIBi^cY57`TFnKBZ`!; zr25c`WUNk_s+dq!f%;I+M$f&skZh9>@(dqCUZ2$~XK{I`ELX?tXa8&5>HoF$wK~&& zz`?QsIym3C$t}KshE0ps)CFOTEqAGzx3U6nDN@V87BNf`3A{Dj-Us=g8}l?)w_+`M zUBe`=&|kZS=U99lA2!JxGVa(C0)>OT8`m{TVu#fo!(BzEpzZk8@aX~MX3K#JPjH2c<4 zz>cdMzuPy{jccZl%VK-gKBqCCHQ<{yk-)|qyZ64feh0lp%smmeIblS{KIF6n4R?eC zIi{Wwyf>hZ^1RKkSphqsbgGH#4Yo;q=aE`71ND1S3}QJi3GLqT0Y-0n70C={V`ccNKV3nWG41@s-HpM2{e3BCQ5P{Q64Pu#U>BbN z@jeRbGV87=LjU?7zT0+$A-UXr`^9|yG=_{#M`B#fp*5j&Ia~icOiTPmbm*g9*WA7s zT4%e6(`Ko&(K$N*+i~@OJYNfxk_sY^h-1f1C+Fi2u0o^Oy8<%)Q@88C?~^-RYYrx- z<7iMuUO1Z+3<;F9s&3K163eV)yNJL3Ur@uUl+p}n4=rT-8uL%SXj(rs2>aCGu8TNe z#cDTS_5gb$eQefRoy87=jvvDH&Dg$jYDJThBklfUzsMJ!2DkCO1aZ|AUl z0o^(cIj2_(yBw$E0Vdk9y=U8*MCgA_xfdQz90*2oWOl?4*FL)EoZ7OKxr6*)^O|(` zp{_Gn#V_(3W*1qc*X;eTcJ}{Ue@UE+wn>U*ba|8i<8|}{^Yl~sU)K0%m5JaUO%inJ ziJ9BG9Eabsr@I6;`#_uN1wQHf(|DKL@3w7l*#ElS3`Kc8brX*NzK@<4=KY=ePkRHW zXS-K)UTVkj!F!hWHP`8`&u<;lR%{^nCo)eDJ$S`6h?BBS?6;TkVA()gYG$1(vFFfX6QrKdgwj&Nkn#r;G>*s zZ@H}8Ig7(Xd#%DU3H`LGH;43Im~DsQ(0$EqibQ;2%t2rHHV}Lby?auEUYN_mgf6*u zz5YAwSV*5?cCrKNuX-LE_aw+^ytw*D?n)SBa=3V1>@)=3$b7ANjL4%>znEtz_=(^M z*A?bTiy|Bk-Br51n-=H58n@~Bu(S7|JHkAf1zy8Iv2^W`l`=T;?a$sR|4vxo<*nY_ zTnd@qdfOkIkc5&`8x_5I2!Cy|xoAD9LHmDxHxkRU50AB9RpQuQ1*Ox#OJeBuz_h>n z@_=`4m2$8xG%tfX?mJQ>7PqHGQ8|9lda$~taAqAYuF#Xy3pt3564LzR;oC?oK~;vw zpBm$oXL`^POK+GK|K+~P`x$ms2r3_~GNc<%6Ya%Xn*{(XypMmpA{vX6hNPCgcC*l- zryFIK&V~~;DUB+Enb56M(I~f`AN$nEryu+h!GX_F-n-3mY3;Q0RZ(VF4UrfUcbcnr zE8)P%SBGyLC$M9vbbs}scx=77%28wW9*j!ce8^&416$ab?+fwCK(0baNMY#BG=`N4 zPrRpSem;qbhise8z7@y640}HM@B*hEtm4xis=rDQvAk45|Y54QT z7{~HLqawX_;bP+HF~PteNHM5T{Cac>`x<0Cjg)r7SQ?j|Q2#QV?BV;VCisi4A4;_1 zId&ZIf!xWdLaq#2egYSs)W2fAP3Rkzv3<2;-Yfx=oo+I#R0>FJpRQSy9;c5}Q(jG< zS}_?4T{bbwZYM9o+^)mV3lR4dVKE?}h@c{vH+%tzct2dJYYH#GiKXzx1AlhI>|qbm23q7u&_wNNNr^N12u+?Oz~RIbITGg`}@ z&i};ev5;H`2Nv+_iS^#{+4}EuQumdNpTegHuA^NljT}kQf^J zd9yuH4?&skV*kow4~@(+GYiZOFk4c%K{vJ#$Atxdj;>3@B73{!CrcH?_ph=^yLHM4 z788G#m`Z-b(L?diRaTzIy7~`;Rm#LX9ld!y{e&sOr#ry-ZsEYTMfj4sPPORe1{hYG z@}FN!$8p&`Lg8YMiT>0+F_pVp0SA_CxFU9(2l{p;KKB@~MQYZaT3)ken6kQ3CiaEE zoz^m5<<_y}#_8uTxV8ISVWF*8PkI@_2Ql$5!Sih12WTtwEq+i#@VL~UfBe*BE#M?$ z(ngh&<~Xt?)7egQ@PB?sm~pYs?0qEHfBY4%A4rd8< z9Eu9!zBV|1Q)pGQX%*JXy_*OQI)feWjwng(cn5vDzI*xAL~-Hp<*d&E9WXr{#dRf| z;QJUZT2$Y5*$=zxEAQ~z(Rg!UK=qyh`(h0g?e}b56HL^vRBYn7#;3%Cpshq5zh_(iKBptr|b14dPcFL@wSrp3IZoj zogN#BZ@-4w{Brxlzc)Zb%gkvb=I=;7E4=TpBOMM22fH8sN?{X$Iu;h{Lt9$>3@P~| zwJWaN!ro2mJMQWbd2I%%9lkH@Y5H6^e$_@%uVR7tE{aOI+*+C>hLcf;zm^cZht)pU zI_0AeL5`UdcV165RQL?m#JE-b^Zg3esR>u>1mlqG71ivkKG^GsTb^xqgCuLwKT1n9 zUJB?Rj(dClp#;txWe8L!|GT|uC#~gybvZa?WINKva1y)g%wPG3S)i+`Xvr~ILYH^& zEGC+rC-QzX&+cwmJADksuAYesudG0F)~1{K*C(<5z&ELoVqTc2ndZFHM#nP%L96_a zEZjN|^=a2l9>1Z*(_nEs&yd%)d$_QNe_E!!5vuue=0EGULAVvgAz`KnI#1Wsp77$r ziDuPLY1tXr#u@#E%E=6$Zf9JvANPT}VW-}HUsLS&J|wWE+X=&;_~)43`-ly04({)Z z$FWWE!2SNb7EDe#8N5}Uu0L>STl3BI5~5CKa2fx?fzde_tuQ)pUFai}Dfb0!m@L8m zBkp_s_R{08Dt5bEkJ+4v6TfV@HodIEc=ftVKRtA?*8>ALT5=HWHW|3@bIqsgm#TFY zSADX~al$D2@+k`~%n>lzl>L?9YaDuxna_2Eu`8u3zU}@&Vmvw@T6}lP5r%EK*1k!b z$F>8-`iF{X_%ECY4VUG*IYrptkF4&fe_0OQvy6d)>IC1y%*vJGnaeAY5@TWaB$kHv z!a1RB0?EciyrcP3AYa__28@j*>Cac)z{Z-hVPe0Iqh;K;%r!#sFz)8Ja^o-BevNJJ z!v3`mba@oW%@ZoFwD+d_y|vKkO}Fq3pr}gbq+BHC@4!Tgp7sNE9Pp0+-1+7qUH{i@ zYyKYQLZ`Fw&-JEdw}ABBcl34B6w|exce{7eJy$kK(rb4x!LK=?vAxPag|K7hRR-)t z46kD3#*v$boGmb=Vi`+*??~5v4)={GzXOPUd}P;#LyFgE@n-C>TUm51I1eUQW$S@G z4~&QQpXGi)#Bt^(oH!dLBcQBPM=fj137kn?8QS`0jClSY>A-9kdi!&H$$jszKhd8( z4;Kr1vaaIz?|YJ)xd?vlq=UtUc~keWQhp&WcoQ8T2`s4Oy=K#9#;C=2t~0gqB!(|> zvMyS5cqJ?vwc9n;@e;;P!-JSAf-hlV>}N^NbAnfORV3`q(VxA2j+s5sQ!k_R zwdw#}yAvC0zfT=|fWvcd-+hqmgpp-}UJbu!x==_KW;1DK=ffDkIGDOPhV!?5>c)Fh zur{Ada+~W7488DRqEuWIyE^#BpJZ&OYro%D@t7}bAlA0Fj%67U>wdZGJi?%qs_Z+voIiwkeC^(Wto+lvU{$h`JDCwB&@ZHh?omL>E-``3m<`qn#PD$Ah5?tT+u-HnNH z8GG~_i%$Cg3fd%$Q{6Q$gAYbRv)O~lxv$Gf3`!YVFWnAe^48I+6NLmn*_?{Ly*DSJ zZ#SanpPj8mTTig=tlWtYy(Ap!l7G#_J&1K2&7o%`E<)GkJ=Ob$TyeN#NYZk@C-hEk z;`z`@+Yg}24)zA@rsI`>hy?PtQn- z$Ir24_yLyjI_1^Q(fH({+~LN#uTEjO_*r-jqsju5mE88bZRZ1&SCoEbC2&c5bZ6)n-oN&__pPgfUvPetOol0S4Wu3)5zzTa@aRVTOgi+v@EMLAd5@PW z%3=8GsZ*`}tDwnI-R_->5BAl`_FC}gz&uMaxZf=zG43%pkPGc1F{EBmMD$(W@j8>N0SXFi+Wiw^ZtK&*AKU%nW=fEXwuP=UEb(RTx znHAX2=rqH`+xJg>ULMEZShY5Ty`wnMV)<(JsUvhLc}V3Mc9YoCj1D~IlOr+o8mmmo zu7D11i{y_f{7@#IT$`ae`FEXg-8?&b5PJqLYJXz$2dbD(BHxuU__*M((_8%u%#1Rm zB;|g@E+(g+T5+GC>lB-phQ-~>*Z z9M!8;6~_^c5N*kO+(^laD7SVHfO>^uv*r}S4(oiZtd;8d4tlSp4-GD`U_XbwS;K>3 zZ2GJ+lDBA!!`|W(Ud!Te&Zc&NGqV$VMXvsQc!Q|R8ed)ceDrNF^t7bDVrbrqLk$)l z65>WA##izUdrI7~`^LdM^cckc$NC;_5)|;+u#@lhx&N@E@Y#i!L;ta#-B`zzbypo{ z^Q>2HGykXlA9i}zc%NXStAB^;(XP`qt+Y5JHYyBlUY~CWZA`wmN*@#UUX)LDyLE&J z?f*(o+4ajS#;}~q=*~~^rE9n9sLwIJYN3BV&*W}P-+nO~pA#f%S%3K!^3Oai#4yyp zW%Qu4JFw0A^+#sqY@|%Lz7^g|@Fdhx;-f24Y@w;#T>7UX4ex~YDiI{T4Xd%nL1WXi z(jFWMqJ*AUqln#JVhYM9Bymve{&JJae4J8mNKdvW@XXWKpG|HzeuLwjB^n2+YoSXw zMqe@QCQ_=l6`mNZgovkYRRM)GohR%Nc*WLD?Zfe)hOp|@f=IpK>=%;6go~S9DNi_c zaiH7OZl`=Q_RM8T$I8e-_K9grQ30BNg5<;bceYQvL3zQqLC1$qkW?CY0M&@Nb6eD0 zpWEbqXjk}|FI2?@?Z z$h$9z1x)d=kHlwSP+=y`L1PsR>J?7hs3P7 zGH~$=VNZ9S7L<}H{D315rJ^3W(bYv@$JZa<(@1pn4$xfv_)_5{vA(x#4A|r7A%@M( zD|3vECF$CkJHPj-X4;v5j<59bZ1>MYboBtV<1?D#tHbZ_#2Go`bC<=wqML-ro0A26 z(5cFQ!8MzPqleGEPN>%?!yVrKeh=*FeJ0_?aW(cd#(?Gu3tN zZtP86@i-AOhz)uLPrCl^@G9Qp?Uq3Id-V|SVo{kJ^gD`u zmDsc%JFCV{d`~dJ#=Hi`Aa5mXFX1@H`M?#(m)BI-{;ui|^>-3gu@a;~%66QyYEkT>ZaQg?!$(63>l3=ds*U%E z@3L{y_2%%8wB+Ra7b25EKQoH#pw-_U^{+9cFuD{FK1w zfSrba$uS^DznTbKi~;i2(F-{fV-Uq1QYw?egi|jsvzpmF#d=AS;o-4l=zEqj-P?T) zdiM_AdXu{i#^0#NY<4n0@)Q48#hiXn7W(v1(Ow2(o;pl)d?ia^kI|#K&yR_IE=(xD z_31Og%P}phP-bX9fMY?Yl`NtvVRGCdC%5ntj(K~CxrlHAxhk~gw`DZ;tdAPnG*8%r zvs_v`hFHe2kAI{1vVFw3_BDI4C_hz-HT5avR5My$45%MedNsQ;VOlx()c!YeSZlz$ zsbGS@Cp1+qoDX3>4oe3vs2SxEbtS(;ToS5Qxg%xu%T>~Pw75T1@*R7-%-{vx`Y}3s zRU+4d7Qe!QHy*Z68h*oA`-Q!4Rv!7k)@LckI6}vI>?*_CO?PPZ5hONUgBa0Lb*wxe znS7en1&5BZvHS2F!cy|t+AqErX!mcibyPhoZ;jot7xx@qN9c)7GP7*n=0xBzhllj` zq*5Aj;^-HJE1Q4e*rW~1dgJX-I4&E#eEC77lCQRKukgkR1BW^*Svvh)9Ide*5qp?| z*jbz|v6kkqVpliAUg2U19Nx+-N$_$IR~sknCdWCaD6_DzE@{&}(QzC;Ze^2adl!DhSGC-K z7=rCA$!hO%{v%$)EER8ld=?9??Qx%Cw#52girU4eO?2-)6kW$xnMULzVwP{L(-p3Q z-t^q*F59;Mc|HvCrBABW5%Hg*^||3sItg4%cZCU)w3`sLsZcCb%!ROu#WMen?M569 zYx;79aU;w$JmvZldk4EViImj!r9u0R$&u{Z?KnA5EhV=)4CnIe!xk24`Ab+l*|9Ca z=MBykB%}-YtUyZj9ck564j8_aRl*4`vxrc%stsKi1KZL{7<6kyIKNE3`_sC(*{ms}H_Qkq7 zy9Mj`9y?Cjw3^>+p(jMG7+(}=-++xCLAzYav#=%MOO~A13!LI8 z?0NK_8$! zzOZw7?Xto3-LR19_i)>|GL!~$_PnO>kT`_LLYgbLAR0$%nSPFd+JqsWP_{PAGQ06X zmPGKIc5Q3ZdH>3W#PCEo%WQ@ikCV#Ol2)!g_-jmBsjS`&>MvMx_RV#|xQ@D9v;m=q z-E7kSyUu4f4C`OslChx#hJGDj*yu{a=1=}7d20WOdqm1}Qu)RAE3l#RVT4??2;K2N z((&gLF8)WJnn&m3){={K=Zzi7`R$ZR=u0JrI!?)*GzGHFJrSnn5NQ0SlquY@j5vS2 zP4T1LP0)WtwI+A#8BE)>`%<$KT^7SFecsi*P0jf?^#z@o|OEmm9VQ>7d)1UkZ`AtZyf!s84vIZfE-Vw?N?0TCA)7o0h#5)nsY@g2Hx z%6kpANyd@AE}6o}k;}^r)eZsGe(``ULoC#3{E|C^-8d_@EsuQ53nxvx_tz`#!&!}| zN1CN?V|o$SwT>e+Tr<|o$POOJCwNOcHPn1>aam)Bp0gm5wn6zzzSiq8={TKx^Mv}% zgD{qOt01eF#&ZCJ2VQIH2pq%-+v!HeW$`#5r1s&sB^yrieNK90l>rM!zj+=@B;uS4 z)7*K+3~T86Z6;qxb2cIy2Y9cqbv%C=JA!|lkemL7gNbg-L}dgqkiVo&P12I?x*{*L z=k0bQIK!gavQ~5f`)@g{&U1CazJm>)nQY(T@Db|v`?1D2>vuo&qaIz*RcO0XTi8b`~|K#q#}CHo6Rypy6%I=DMV3cJs0kTzbW!|~A0OItg0c-c5n zuS%Xvxz**Ofx~lb0@<-KByOv(CYN-HdW*3uv#gT*x+K;;g;jHL!Z5*Xvfpqw`F|eA z5xaGzpZ0u(>8A{pe$@njT+s>|Q;sN?ecg2y{i8UP9BmkG265UdiK}!jLtCO z*n(oTDf=xP(%K*<=(QfF>=t&kQsa?wom%hawUY4jw%xLPzxFru%lY*jeii|}_m>NL z9=?s^SGWW?+?RuO&MWkAz5+x27pxh*(jjN6OO<%>=%G@`IR(p7ST{_-(G_$RdQj@hV!YzogUk*JP&m%9YXr&?$7RV%MP<9H02m z#G{Z&?1PG0Dm&^@AtOo-Gw#vyuZ10f5g3*wXAe@vSDu`$$78oC+Pm#EfNv- zi})`3>t5v1|FfK;H(3mVR%4l>osj!9TUd&97ud-KsUdk zE6thfIUhTTq2SPw>W4%esBN^?AUR_L4zHGU66vJN8xi)_ot)JfJJhfysf1F?Q3><< zJEl3)G;w-lnw#-j3U)oVcDT;S41>1sMSivs`jXkRZ0ceCUvaL>NambT6;AeD2o<6B zV@HO`2HTr1NbT5j%^=K;#QY@x!ZybMoY|?trQ;HS^*fh`-+CU86WkTOUpQ&|BaU@l zczn!)X#e|5W{xa(8lDrAwqG7y|BM!wCo#`lZjo~*{FY&fusqEPPpoa6jeq?90x=%5 zn4b-$y}{9nndTeT^!&$U)58?O;XBxI+T~!%ItJ{G=+z(GP2jZ#xY?@LvJpBF#nwhE zDxHY$H5;<}ZoPgQiRFsd3?B!D>pP75kDI|nLxldV2Vyu^ujS10TO`f=QU_3pZI4xp?Rz5jBjJ`4!U zKXT6KBr(+P*`mY#0^96U^3@*^dB5SFnxd|EE+nQoBkqIU6sTq4T+v^Xh9NIGnUB)HLK076d(IoSz_cjB6pk zU*0+q2kjLK?r&0~`)&jFr;L_T?6E_oR!Sz4E^m(1qVsWgPc0zX^Sa%=6$c<%GTY$d z(0}N?1m%C&zGe{S)?HU)_L)O!*2w3WA69?;gPvM3*5}x^V(VH(-8K@FQ1pR`d56eT2(rT9%zR$5`qr3K9t!4D%^3?t1OD%D*)SxjQaL)?|cUmpQYUbml zU|dMt(OX2no)#+K-1twv7o;!%~L$2xV4GYL}-)vJy!uE){i>PjS8VqW&mw*EY- zMc~s4$AT2R&wl^c{b5?~%H@sMX?8FSjYRHf?R|-TS?Lj+A_f0^?n;vE1?3nLR(U$b zU5|T6Oscdw-UGr{AFEgLfd{a(VuOD`K>3rnErJ?5~GG& z+2NPYn9FKxUid@+Cs{ln^fUj)rrY<C`jn};Y z=l5$#$+@ojsFB1F`K0~$Ny5Hwsp5Qcw!jKTYf!PqAsHH3i+X1d*Fa~oTHWNP2Aq!C zdC@%gH;JiGu6+4{cvyUTUsCBn6Oi94)VlADfhNft3xYqkVCTNd!XqPp_pN=`9u}E& z}~Esxr>5SkKw|l)!(~X7=sf5#gtDfMZ$`PYd#Wv2b6X`^-*Sy(+fLrk=emejjEpJ(zCP|BORH<~6e&4}h6sLKY_|t07;OJIIhOpF|Fz`)&4}Pg6?2gJqYXz9sJOhem4fCt#`Z)C5K%?u(X`Js|f5YH3y}woW zYX@px-VI-B4OZ^1`VYM`s;lB@TN&azZSmC(Rw%&MzL;|lJPAJf@ii?s{g>}Ra?v{H zCpt9$5jy=MFL1v1!0GoKdlEl%{PR7#7_ZFF1`fin&wTt(Ty9~k0r|>{^kqt9`}{C=8jhd)*%t@Y zbFA-|epbni^Is3>7+LeN91)OfclzO<%;}AhuT;X(U8Dl-xp54H8 z|NA1gtX$?5yez$-?s?q%61d|9oc|l)T+Eki2vQ$N`f_Ys)*JaedlwV);It+I4Pf>}v*e9^+oi zaE}M`O5Yjy-#Lu!o~pNIb&H_-evK0O^F|yCVEtLTcpKWC436w_9R`Zh1v4AY2pDPI z`hL6rQ=l%bv!r^vV9D=szK6pVF!0SQ!}m-!jMt=_ul*0dy5mIRqWNo8?ArV_{%qDw zB%3!xn8ylZ-ip_8_D2jchdan}yDu60GR#k3-e8Z9zug9odu=2JtH*wEudGQ-Z+JpP zK7GTXh~_Wf9uoMpS{;7jcaQn8cfpoxd%hjrccF4juz$KhvnODB;>*q<`E;zl!Ct)Y z8*M*B@`0aTF@7r{=b+2}mx-&Oxyf*lyqSgv!0{U{4Xc;a@KG?uXqIPvP6H=0mdzg7 zLCn)3wLz_nKi81zHO3WU(q#@2MbeFxxM#y^V=y~bAIJQiUyqZ0j zfgKG@MWS_|fhs){{-}604jxWE|5`*B^CFw{XJ$Uq+AYgg5IJn*fc34+O$EGbuypUj zdTp0(oR}V!E}Wt1!@xpu>L(vHUJ^sIk|;?Vx6O_V814#$(XUsqN#!fBG{FvBQ4UuX8YPd^-UPC>Exnd=)4-odoe z#+hp^{*YoPXQ!G;tCxVd+HO;pUFTp{Q9~b|5P8qhfqoftMh9${(|>&~>=QISn_bg& z^eIy3*&eLrA>z0FGW$2JIFt%Wm3n8ED&%1y;g9;qG=P$zR~r`K8QpreAm?;B@#R$* zy(QDK{xToT>I@ti^6flW1yj(*9K0lNM@~|+=#xF_F1;{5B&+NNK>nG@WE&omLQ7-049w>Wt zn#K!@?Ppv9Vx^rCB|bhk7kEgFU(3u}Pd*bky0+X5@Aa=~@h$vpa_yb>n+>$%)hGX$ zJq|pE)RTRzTxsr*YNXb6`ECGA|E>#qSO0^=u-NsP{7M$_YP?GM#X;;Bh5R zugT0G=pDl@&1={@qxt$XQ73`z7CT*x z&c$H+n#V5ctQJU~xeSt4;TR~e>|~kh5DtZkDOe@`!cqJC3*1&)a76Qsae**BuMoK_ z)MK%Q=8wR)I6oLihcjpt#+#amTcdQ>M%$+?iYj^zLb*GpaaarzN<}hZP zBnd=sAo7&t%R-Xv%j`(ZcYU2xG=JiZPM37GkTOQaRP5Y8_!<+G-A(e#l7O6RBy8vU zAH4J$%j14kpU0WYZQb#iR*)}g`D1?%jn@UHx4P8TTn~jQ>z%;i^>03+sXLO?jVa2| zomJu&>VFQf4xd)Y|LAEw{TsTKhBnP}(LjzSzNykNjT5=UYn3*SuUqxm;5qbu}# zL9YkaFEt2;0`6MX5V{S;=jgyrVV z`eV_+K6~}gV%YH_(fh8)6p3NYOpc&g7EFrl-V)k(gT(yAsOa*XCz2Vi3A*TTV#M_; zf#W=BFcMjH;1urrJI=Lke5o&jJnuK7$GaUyf?NEhLlqFv#UAaOD3GYOD1r_oq0!jc7?LTe1Ge4J4PCQ z4hGIP^qB>G!S=pG)IRDu2%P>Fv!$p4sG(2xj{u?9HmR(eCAs8@`EifWKmJ;SU2VVf zkDZ{^9b?d)b`||)hj5famH1+}4<`e7GI}0fg{iVem4rVs*t#h?=sMEwOJXdO<26a$ z3riW1rDg*&*fz~gZd|txhIc3|3ME{D(Ye*pu!<4L7CR)gLu^Ql*9Q7`Yu?A!pu-Es zpU#mO6XEsXMy@9FaRWA=8d4{=K36e(| zZ^54v{yUmufy8oY>z`;@1t4=X96FU|M`CbYJWO7-pTzLAXiEL8Fo}&Zhazc$5L`NM z`^1{a6U}Egh%>nFfl<<(kMHjbz~W=onp1pnB<7-!nc;6UFw+v`{U^ErD9wjB`bA7g z>?Zon2l%#-m^-q+vG*K;soW15J$ybSwzlj%fvwDtGSeql@I(rFxY(+GyxI)Y9eKsQ z9(FK&=5_Pd&omqaP_6SM8ShpAnW=dsQ|qr?bWxpEyXqj8a$WdzZD%T+-^GL6qbtNfb5COuiIl_%(;6$^t~}w8Z@4>rZoX2(f4?hD&gzc z>Dm5ZtfT?^e`artZLNWk&>hA_I=^7z*adHU4i?V?+m>DZxv^9Zhd!G>?;R)f$+N3u z4Gv@uVO{hfb&FB}iHG;Z3)#J?z-&k;?QCMko!%-&uhQH+CIu2H|a(w8{)W}T=p3%&7bSFZdXT&?~TdJH%p;+gs=aJ zF*lBeq2+op66#SEQVp> zRr_mJUMF#((P;%u!k;*))b7;r6NDe$w7OEQVc&Kf9^_24^I71_i3#1adc&Sk&@dcw zCfUs9zh7s)Z?MGvOUhvZVjnT7czZL=6Z#G@*L?6H`09PW>gz60J_4=F%+5iGhW&E` zM(6g4W2*2n%d}NWkY8>;>*?{@GD=Pp0`#U?A{YYI+Hc*x!PdZK}AWi-W#<9i8NvFt908o{w{cv|_JP1AcZ=X=q^spzjr-KnX(YwZ|}MR=Z=-1r(P+(+h@Q5X8|wWo$$sI~Z? zdVV-AI_7Dl2!2H3bH|`z%2BT%Ce=Eu4t;)ZzVy1U;&Oif^EtV#((BJv6TE=VwL*8XwiE~T=q`W5 zdXc`ab>QIU705ipaiyhdfqooJYpEvSx1PfF?%}ZXPbEMe7*GBiT#S^&_R}|nbC7x?S)!qxHE)4dAM5WfwgW#4f3JDC zo#0=$zyCA%>G2SbZ08TPEF%1U$8WY>uh^B3)wRnHy85<5Z+K#C1D_)P<`oV3(AEQG zI;Kgs;Zm5QSnVj^OYo#9lqAzC>kOc_F5565q(EX%*PME}A_*xa@guV97V-usW+E~Q z>+WG$&PVd%%0_7VS$_J%{C(`^mOuLUnkkZ7W^NWd3_@zqrBip;d9Xg$^B~OF?65x8 zyx2S=y4DW+#c#{2roM$4s)wnFnj8+UV!L%oX9T8EE+fFY8V8+Uc)kA2io20o`C8a? zAcLB3m(=UztCKgN4U}!WP|haZ@))-hW1PRDp>! zoTHRM?@!<}( z_a$bv#j62hxtj@3oD9^qtv>!uDH131x2*7wD~9@3?yZ=UKQm#(sxv**^VL(G;( z9m_}@-&5P)t~7!Msi?yafAmQF<7=%Cn2uq8UiayYu!TB|-Lr0N1NZj8JXi9uZC&Zu zvpbCYhFv)H8Sgf{x8f0uAK0}^W@#e!sTIidjYUJlYX3Wbd^bRb;zhPiUQb|x?Yw7= z)&&^*a8S-eeu7vp-e*MK1rK6MT>U5v?E!{k!n)t*q>#LNxa{j^1?YSpkpCh?6(*R5 zkz`&IB#VR}Iw>%SbSWR-UBU}|;jvvS$+f;h0H!_CM}G%ACvh1D*LvEfVM@J!-HobQ z=-R%{EKFh_lm2BIhixRb4N}=lG`*oW%3S7B@WS|)S6kCp?AgNop@E+H=Zfze z7-ZU;PrKfQDzRkln6FELY|}+nX+4isOfxHfaykxQc;Qs)?}iPRmF}-z-VM`yE%&x_ zT}8%Rrhux!L+tt%8}`g_0LNNKqpjAZLXX!<_R9~{F+umOA?1)D47WO;EM$sf*Sp94 zD#9a3`!4qp&Z$9ddMMw0fnzvvO`|9LMK@4iq$Jp%l*NfxUB@Q6q+!&-?g6jRa~$D0 z9a?su$Y+dS_5A#GHF59n$uGQpSXL96*Ke8o9$SEW!6G-qB~EHAJ!Bm5l9_ta-3sl$ z`7Ek%j#2!IOO9sY^_uWdFw{9^Z_KcBoqIE@!U6 z5moyxA9o(??Yh@+VLeOtfW&6vw*3 zSz|37Clnlidd|4uBo9Sz=k29H@4R(F=2ka}U3c{49kC5qQ6%;Eps6eN2J;@v_H)B- zN68znHCXi=n0OP`>aV{Fru6pzzU#UUYRtF)dX`4`&zJwWRl+Cv?0>f7_q_89W5EUd z0O)ns_Ftv*3>(+7`Q@MfOJZxXPOO-&!>KOAWsC2!@~0&3HDY@!ZU+K`mdn!}umXEU zF7Z0wY-D{uLrTNQ*wh2MH~leMVSN}!BCfKpDe}ZVj#a5n-z;Ei$ZIB5?<|Se{(XPy zfD^QDjqr`Uodt!Gg5p17{%8FdW+`eK@6O7>3JucFrd0VO$EWOY+VI@gsZ;?R__)4G-n@K4`+cAO zu?T#n1!MaHZz)apz>r~8wE95;r(Pd=ZPH@DF23nHv^0Fc8)=((J7NU_a5#Q-<<8Zm zK-TX5?aOCK;`%_#c6wnB^yKDGoi`sMOD$ zzdK%pKY5 z)xpe9AB$V~Cf<`N;C(Q+N%76kOZGs` zxfV&4E`<5bnjUWSmymsvE7#`JH2nC}W_GIN7u4QEi9Zi7k$6J)P*1;N)iX$JORvW) zRl5UIfuViFt&%YH+T_glU)|V|kh?KVei|waGT*Drnn3lrx=Y^%<^R22v;Dkp9F)I1 zWLNvpiNqIaY3*oX45j+Zjy?o2N`_s) zunxV*tH6i7(|^L%f2<*K^LBCdH2lx`$m*B*Fwj2x^cU37UhJ(qFHGV-uv| z^)vY6+QrIiA$6|xM7M?mR&FO8eEk;J2L zxZ`jttIiJ0`CBnLaaI^-)wSta>l$QQXX!mET%g+oebeU!rq{MWXZgD{yL$$S=IF&TKssUcEU4sB7bj6T&SQw3FIl63)GH zp^gXnlfBTW_yXM!7@g5>spO(Acz^kfR9IC&CQL6swwq^y$jebpgs+WwJwm2`?$>u0 z-En%kj)m4WqRw8+Fk7|FYXZk5Z}w4^6Z+`o+S3{j62owse{t}qL}%#Ah~S&^X2luU zK|f&be&`L5^<{YO4tha_=GBNHZz6vAHe42rzC_%|8M9uow<`+i{v~Q(oA{wf)9tJy zK8NAc_J2hBsF1Zaf+274jKsy^F7t3xDzro_d4F(!DT!x2>G$foDRjNPf%{7a3x5l} zi;K8}yGJ4UaQw*i@ewFgKEw8DcOZ$&!NNAMKLzO9nlw0uv!LAO#wKK|go#~i7L_v> z_#?7@&tP74?mlGcUy?Wq!%NHlOaVhwQ*!LxY3zOTq*lCt4^EUD+%xHA<$)mcRqW8z zVhU1uQXLGVh;zI3c**s<%(K`sds@FbVm0<<>L#@FvhZfuU|o}a$bA=3-Y$NZw3iJl zZ(i#?U#W-`<6ZTac4|QO>zh7DXuU9Ns zH``@0*gtdaS4~c{$8fK zoaIl1%|XlGX6HF#xB+Q3_hTZ@mB}2Up0C!%)+07Yy%sOnr?7#2zIva~4rn*=s@2le z$8llJM7F{OI-2l3{YiuDIXB4ynL;Y zir_c>0_aMPTe)L{4t5puYA|iXuy-?`m)X1q4i@$w`jHk0bbssTw!0OOzPg?(#IK3< z`AM|_2e;-FEGx4SF6ytubi4I_mb)$@C4KF*iw_sl3{*JCPl-Bk%m%}It_+s$H;Mg~ zeRF@sFBstt9Na7U@BV7^gRfrX7i4Q{3M_4D!O5{>BDC+Hkg>?`O=o@^OkSOO7=26* zCvt!5-Ee)16Up&xmMQCVutx}bH-;!3Fg=blP6iD%tH=L+ug(lFJ#s6v7Fs0 z?ju7A%j0vrwnnjh;E>jJA%wf;JPfQ)G?KFqKx*NIrWvj|n5A94vwY_T{27+Rb?0R@ zGHAT#WuX%pw$>%uT|^yv+A@e-5s*Ukf2#-lI7@auW=lba9L@S0{1no-;bf)cIfi zgM-WD1cUj>IJWp(*!5@PB>vM4Cph?LkSzPOx3Y&1Mjkc9zjGJG-+lKU+~2K*)S8or zqPnl*jBu-Wf|>$}W5xEp%0oo_G}654lj!|W%q(5Mbm(*>^o5;ivx~U`%;lFSDjq5T zW7A0kr?FnlzIJA>PlgY4?RMqXri(&V%pvz{4M&j93_NlY&BK%0ZyVWye(}}wI-=~W2zZ%a&`ywIomHV?u7F28U zP-pQ~h`JL1I^cxn;)QTl$~l^a&&#;!h5@5kqEg(Z(Fr3LKv(PTmD;Umag3UUYr48>d)T!%Lw?9T{ zo5Ds39`0phCn0JPTb=deNo+Al{Q53^K|TYT1}u{&r#+yH)U*5Q)2%obNNsg#cES;T z%FxN-k3gjWrl%Uf>BzHRpn^2j*Gl)Z_8}wIlBf3EJ@k${ z>#ACsj|1kD?iZ_TfqZ(EU=G`IqP;#}rj-g1dqT`SRGy^+y{r5Brb{YNDyydKwVtJi zKw{$<$=*4B8oSm9SthW5hmx@wN4H-?*qOKIh{Rn3=%?u)*}9YuN{pp^o;EZAzHS`%0e*20ZKWIwgbUWKB zz(MFPgkniQb?&<{|)d1D=I_|j6$xCpsbd&F)}sY*NKv)fVQIiw`Efd(ZBwe z2N;UPJ)IHu)!E$i!-D5hR2(mUo-l^t+bxIG#R%QT?sW65hi|aB$L?A z7zkwPapBaf>GfF+1Rship^+GNsTV5>reild&*AWn6>k>{pTx?*hKQR{Avm{TaO0~q zCgzs69UpFH<&SXuR=vd1o?SSrC!g0cZ3MILwq^64js=SC1I*ks4l|24?&a*0Lb6G; zyYYGgH#bSXk+I2V4w&7MTSEPsU?NE)&hRYDFCB^w9pjfzVDUbYJlmQkz_$nK7bcbt zQ-5JUqCS?ebg7{5HR*f0?q(c%rnAaJf78Exy*V?Dw+v{f(l*^mFDEU7R?6!15}e69 z^ZHffJra9s;@VOEB{2Ei@Vx1@)CK*@_~d+$v#}QGKiDQ)ygHyu!-h*D=nMG05?(c( zH;D7y4*cKyouPYMtPHKK8HOFGoO`TZ!mO@nNQj6UF@MS`BL<}jeLM0cir^PHLg%+| zo3O!XcLYv6@Vwyi;V9Pq%5Rvdnf~wgK<#$!Eu{tB#6Fjx`?db~cA)MlmMOb&A3N7x zv$$#a6RE3{maKARtut((o_YTFVeP;6oxIo`Lgbr9m1%O~&)#6KRsQ^ww!3?{;)?Z#!APF`JjtK?@0{7>^!RXz8TPZEi2Z0xg<-K{57XNZVC$-; zlpn@XI8-ZgCq>l;V<1=Lp1>%EH7?(<=U^MA_(t|A<`8ii-|o`G_vcvtlvvZ=7q{=v zAWSkvU9tjt7mV-d71~9=>P#Fpe@9>Qo!I{dk0qE3TQrM&n&j-j}zPVE64=#6F4 z7P0P8I3~S~-y-j=jkle_>T((aVt9Cgk3I8E_-o?qb+KjrzA#X8#oBgO3n*3Vi+NIrIP}K@=hz0%g?W~#Tcb*!;?Bbu zTeez!&Ni%2{@x*cz8Sln`C1=IuEWl@Wiszx{y zhmfK2=teT17tk(rXbxRu`D5ecaQQ*us66byO65HAcOA@BF%PEHJOX-$KWSd{BX(Z? zq>}I@7Rh;YYRPyWn;gXUeOfh+j17<8O!)kQsh~xdN=7At`ui=LRKYZqPn}JE6vpz^ zgw$JS*fpOy19>8v|4Nw!iNmS7=@0u6EY|Wfa!gQwK9{1=bqN+QF}%7n+v^n+tO@Jl zxb_l;ML82C`R8%?**!0}2zji1!__5wC=JF9BJLUa)nm^!?H@XyC$PPt>B^o$7ES}G z{NTNL=X?D0SUB-W>=r0+PgfB=k%)|2Wr|`izGG3w^Se2z^+^4qQ_6pg<;M$yzcuU4 zW}YIMt(a%a3`_3}#_crrnZZe*+~uQR(U61GI%#_2x}{L7xKaGqhA7N&ep2NksZHYe zv24um1uK5U360w#Kl|*lGqA)oP2(N3FS-6b=wvKT3?1y+pne#~MYat?Cy|$;SXXY4 zzod=h7Ts|Q>$QN^z&<>7aT{bGbQf8vG7Q~|zx?&^6~GRM^{)=>am8|bbjiT1f9pm0 z&WTP(=x7N#-;!&OO{3=h)zf0oW0b0p7|?+u122wEsuFp^t}E^Q%Ap@|V0M%4iyOPq zzweop$ti-LPEK|HvtlKSrwMa$ukxM-NqzVcA$ zDE7O&@l@VO@REnv*48_c7y5kA*63c*&|BaSM#NkHLcHECLf@t*MEQYb@NJ+K=@!3I zCG!5vj@*kH8x@K5o}McmN3JI3rRcl&RTKZyM+^FrOCr-N2z=c@`FQ#Jbaj|!moiMg zOZeR~ijR)K8(nP4l5(it{6Fzhm@5}3+uzB;!NRQ2@(o9C=8`xDo1E`oivkM7FWp_f z8vE~9`&2(5+Rd;(WnuY*sQU&i%BP=SY>4CTC%wmBF3cy7PK$hWxV4ZM-E~uYzaq2d zf3~koYOv~O`5<(DP7lI|ur&jQI8jqE1A6u$VS<@faqC_MC~UIE>B9{p;89h(d?Z+f-E} zDU9ECDDISqDvmgP8M||h@I4q=B^tJkYC_`R1(Aqh73kHDKRs-92`bkO2}=nc!Fi2E zZp&hvVZPBS`Kb9063^#*$v40TD5;_rE5lY{vwAmkv6d~48PzW-sT)S>=GJ0pwZ;za zpxDv86EM4F(c>?d$8e09UHw}5=7Mo#ufds~$A%5x`XJq+*TMP(fkPi@K56}A$PcJKqI#D^ zBcU*3VsW^K2TZtV-~H&e7svUktG$CZ;&_w<_q7KsJ_ckxNJ(T{`2dUMGk0Yl(g9}7 z$-EqHap;g-x*|1U3qIe=*mFo|5sZGai+I$U3GEM+wtyIm|4ZUnX`srGxC2zFT_4I1 z)QKk`GR<82Whus}GC~h6OZY&qe1JwMs2~ zR%z95me5FIFG$Q%e0K;(o1)}p8wfo6%*uk*98_@<=kS2>1%6R1I%1m6hOdEqqffEM zIvBchZVo!!WAPZEm8_L&(?5yyIQ{JsZ}&o8?vv~ySy!a8U5XWa%-XN8{I5*IP+tO0 zEq@m-8N31dr{s^mNDjsxEBnz{)oy5rm3wsXF~JjP@9R%(s?0&U-wrvN*9m<8#Q&wG zXecs-U!_ixEr4$2TVQp!4w_e6edoF12h3~l*I|tu&b)cK$;|8;)_r;6wyCrYo0TPL zOcN*U=d{jxm`}t>gJFH%k%wI|NPIL{`1@<@wTXY;SGf|Z9RxO+)g6PeK!F(B{x&3= zo<9}zW*8@&pGd!a%DTVd)Yu=xKG7Vc{1}X_y7~bpE9j1!3i_ab!{W2i`Y|{fWccKs zRuJ?tj)X*svd%jwij>^bweu!6NM~JrQM!=NHd$mA)VXyKI(RHs?h^69p_@apzK=!m z*V~N65jBK<-x$vajV%mToe61=?bQ#@SYg4}iEHj_S@~i3xjS>fZp{G_*W_)w84!7& zmiZ4EJ}(L|%O>3S`bGh0=QUgTUUV6>58SePr#=bfS?@d7U#CI?H|L|1?JS=R7+&#c zpShC~%tU~_LxB&J4Rr-1U)T?2R)??FE^mXO)lUbeOP7(jIbYfxcW2=}v6)>G)`>@v z*j=v}e)U`lW3sAJ=d%c$MmL9BlWq1coE-f-ugbm?Q?=3zt)z(i$5f@_%x*4aY_2-; zC?uj9%imrs^nY~7W}j5`((gLJ7Qg?p21V!J;gf zU0-K5G)t`0na}(G-2QzSJ}UAXSjV&Kl_WO*k@1J|otSSDas4uGgtFr0X~Hs3am43R za7YOYZw85{=cQtf`6KxW(#6|bfPlgaZtI*yk?!k@Saon#?sAL(u-Jru>1+mdU0_WptSsaFd7=6AE&!_ITb z9IlLEx!|b~B@Ja9y5g1Ub%UrI%>L+0yP+=xG=ZgB*~j<5=poO0mxl6TD5LPS*hm5n znElvuSb7b1oq6)~5eHE(%actwyHw{5&Qx;-n@7cAuC=0rs~juuh8Quu5KNslF(IOYDVccaGEjF#g2->C0S}e2E*#O46|^~?9LQzTQ+ z_ACh*lU6!GE1$rqea*M`osBr;mK^9C$l^(0PI6ej<=`mR=`43xt9&1uJuM5IcI*Vo z*qVafh6JyBa^C3baC`}jr{8T8F5yCkNxZ+7BkOzvYOuup9b1X_SMa`G6ktv0Q&HGM zDg}}NDSR#4?&TZc$OoOA1V z zt_gb#mV7T0>Lqc8E5w?)2m@JH`QQaFDop<#{#B>45etoy(g#ym&x2vNuuZ|`9!M6J zel)$A2YO@I1<}O#foVm%v_Jnn)NFXONp<$$IrA}j`ndWjto#1g;D{Y7{>IjuciLmK zHAoyQHafg`v=~SCXJ*D6B>Ii4$rKqs3DCfH(}7Rv8qCWLtjJR-fqt*K$kV<%ar9zR zMYOmsWY$}-E1zBly(XKQul)WOueKzo+{g?Dii|?K%iBq8{lWvnraqT=;^!}RC>4e(ovc;X8XsVK zI6+e-XALxeJs$JKKo#l6pMTkIHu+!kio|w5yiq=|6?!|(V&yJ9_mMAH)z(oV)j^k(JcdF-y6}jx+2+>fqfGzcuq{e!oMCTynh}ia1X;Aaq{V%>j`vWG9gsdO7a>r4Hq*%Z2m#}g1_A2Wg z|L)l}1zVb#e=(s&Y~<9fELwJ>b}e?k z=60VGBK#D{KRo{=i%2b4SDok9Xyi%O;N;7^Rb^fT50tXrqV2@T5Af6J{_;c_7T=A; zdFi_A<@O;M+f$WNAk|IctoT4H>=}WM^_@bij-->=ZE}{^taO2DJ!Rvy%LrT=!^`ve zE?-t124z zW7T(%`cA4jeR&{`Kl~PDU2Y5={fGOTGvXkgqqgME8R9-UX4QU;W1SWZcEy={KU%;; z1xm`=rS*ahn5kZU^>|)}gS%~Xy|K6d zdBE2*tUNM_i(7~zuGAPQ9^zYn&l=%?#%AsCX&O?w@`aC1vT%?{Kg9oMtpUN`>3#a% z?Sa8UKKsm#L#q$>%aORJsh_e`Blem9NXMxaotE4$!laJxE^`H%>9+uvx;tFb>#Vne42}pa^`WTP**6Di-Ije zx32`hgy`B6zyIxv`}S7yMFio*KDI~Ngg+5<(~rDza3t!6R3qw^9`hSG5LlFI-$%qH z^pek4CEkA|=EbrjZpk*6VZ`@2@3?g}(p~HCIv*neUHl|vU;>_}!gFYL* zC{j`mQ1`FBp8Db=Oihaq2{l?`UxZebU(IponbUb_zdHlFhfZ&#R}ua`^jkINWv@f9 zJ3xKP#q}Guzj)77iY|nySM}yrN`yaX#TWDD{sh9GeE#h=H_L~FE-)p-pBZ+5;B}Jo z4&2<39}W|Xb!TOoSiWb#un<=KotaL|W8)pB>q-rfylm-l-dmwq8*hcQ#LdVs3ZS*T z4T83pTlRPFW964|MCGeqa0-FTVc1pvX=`{4vq=TCt*`m8XOWCf8+kdr_D!=iQdY-_ zHIJB6LE0GpUPtZl#!Fai-jpyv<$?YOhqeFgO2COr7gIJq5QlN8y(NlIf?#mmUsU^> zIL@4i%J4S-i{n07FV^T?hw{l7%J;{jK+oGf9nMSe)X3HbAuCQhLq4C#n!TpPz8QCy z&y_=l#e;)*KJM$&a0TBCFaEHi;tUMt<|H53DGjB80^$31OJGyvNkj6sO3dUVOsvV>XI)ho!fgNXTayLUU$=z09eF`fhzByYJ4TB%C|ada``UZJ0W%uP8i9@M%Y@2Kg;~ z7U*KI+BZFN@S9WmD(IK^%5AxuHE&4lf+nQ=$)7lW+OMzvT^`U?#*S6n=%L?^F{5|O zh`P++cCxXs)gPd5dw9V3l{gePpBPN$Eyv+ULFP>+gbwS!v5Q|FAObU;eDp6j9f4{r ze`B}&0zNB@t&P)(vM26$6eW^$8bt*gww}7$FSG^8m-uZ33JCsOe|o>!qj>@!QXLrK zL=qu%k4IkdmkUPb!6y}d(!CT<7$05q*e-TE*Lc2l{qeq>{j+j;q(rk$Icl&E0$tG={8<OmxN*7*sjD z)KG))4I8LD=R7?=g3Pl*$Jlt+AmeXW_U>0ytl%%cwErYge;;gbG8w13L#EcwGrfVH zP%)5m^7T@`Yez;V0H7m>N7mP?>&4(TV_BLa6@0ZrT6|fkzhMSuHAaN%u7qQepnf@9^@jh?=M4D$ z7LR|qiNujydF6ujBoTk~H!Hgh{DJs0adwW*Ov-%&a=5oHdg-SxVUF>0}*>ljE z_q($H(HwO6{!Xg=B!V_)yVt*rH9^YwBQoFX@c(W%Mk&?B^sBP=S*SBqIx(xDOpL<@ z6H^u1Pnh#scYJ573=FZ=(I19}!OYF86yL{pki1X5HhrQNny9+V_m?gKMz^t48~z8L z!As<`LrFBuaXx-)DMjex4wbKFV^11{fr-kc^&tdrZ=yFH*_n}t>nyOc z^2qN?6V~t}!D5lkvsf&wq;TBPBXD@5I~7IScv<*qtli}upr*G2sM;q*50s6Q*gZP; zZ5-VUzw7fhp6+DT$sk`?S;hamIrd*3eP2<)$`e9uw#ZELt5A%}+##72N$6hY<9<*m zI=^5(w9goQZu@W->*sge{iZ_1MNBTQ7fW+)!033mrnn{xj}IB<_KDphLL^=pz0qY- z6sVK^{>SXlJanox>>l)N#}@zh-VMD<#C=Ak!1#lO6*MW9vB%r;ViVtaDeKLKIP>J8 z%F+9MIJ)Fu_fi_ci|g9q6V|NVffQe_$04h$q54r)bwm%L^W7)j`BXot0V_G4)|^<+ z0pzhyOFy1IkF~~0Q`|;II7n}a*7|)GC@-g4A13i(_*MI90`1SD)vXSuT8EpQ0c!%XOR-Fu{=89hj8O^Z%PV#8fEN?gt z)JcgDISU6I^FFsmC}0cJKWcvXbaEXudi-4J;!c7pnfQ7Q|5sR=|GrK8=?kRjZ0=tn zNyJU_JLE^50Dn$zVrPoI}@|~rMd(D?VDX4d;*^y8DpNWb4#qdAO77} z(U!R87JSzWd4vysLqGMjxQU}2hWRfu@mD>7123evcwb`aK#@3y=Z+tkbiv-$2GgA=^MT&l_Iu|hvcdPNWs-*-0Jhw{G_#{R?GOljK}_>e>O(5-~QJM_Z(%UQyE)T=>-oMpLO0}_Pg~g+7>JvyU%<(Fq=)#Oz!R|^^LHK@LUrVR^ z4vYnM&ny|#fKGW$_4W(?IJ~`KwQxxUkSnFml#LPj>?UmrN$r*vP<~TewB`ui{OsND z=&GIiNNLFV%ts=0)$6_=ZMD3Y1oZt&^KXuAhK{4fQ56XXVEW9;7pjj`f%>P$la$tg z3~SAH0Um-^PQBTa{Y2tA(n=Oj?=J9yft79+QM&0c?)-X|!MO=*1($5S`O_AQQjgS6 zw##B=ghkuZ;ulE1yVb+NE(pn*&ipU7ZiSWy&oVET4wASY-DBEo9)D8(aFwWjPUYW(+jVcE03EyA?~PO3eE z>#D8dF%tWTx@R-x3)Fr(y?8~`0$e+Zd*GK+ByTa$4KIeDOV@{nk3&m8@+M;YOl{S- z^Q^c9Yi3b1LRm;qJn$_L$*!O-oQwa-lS zIs9OIa`CJ;QD5xQAwONLF^wfV3UBGZd4sgiecd?+GDvI(4V5}ccVW70kFa3PHE{Q< zJ+YR<3>aNXy(?66p;PkPkaTPY_SlJ8sqo!}YR_F?a+L2we>zY2%THlg72a#B%|qy` zQ9qdJ?Dzykdu_xDZy$;Q>P)o%gx59fZ(eTOwT=@rgXX%V^gLl&QOspcL@M_FwUhan z{u~BNj@L)TwjzD3w{}ewi$?{cRkI(vq8p*(RYDt={SFeZc;7FoX=Nrd=iIKsWYa@AW}^z`?<_Z)nPG?N zJ;qn;ygPxit>xFI6Z=VQ+3OO?DB3U>C#qzn<$2#?^YfR{)D~7f8Mn)l}-8XLp$BhoO?X z&}v1btced^%AJHg0Xgc&Gjy?!UFL!0BO)#yS>#yi803cS9*b0pR`>zApz3(@yEoAB zAnu+-(@|)6r0mYEN95^qTHrw7!$4@1KXW3~ECcDMEiT1M8)L7`YVF!_Pxwf=zu3Dk z7g9Ff+P?qhXPoEi+-b&U45JReWXfK6;2(-+aq)3J688m{x&!WnuTqQkHck5|3lf`? zHeFM0HD zLn`rnJ9QKGS7j|pI{gYJ)npWJ#RMbs{gWkz1JAIdXZ5$ob)mp?{E@cnhcOIZ{m8Qd zbYP&FfAHmFUhGI4efIGd4ckZeua7Asbb!Xv9!%Z&&AM*_eVOQm=p8?BQ1ha>>1LjP z`%bCEv4m=9BFi&09Os~^Qz)C%D~&Veb@pnS3D{%G!5!^C2h%d=9v*PeM;d!2ZI|OW z%;@UOD~tM$l+7>m6n0lb`-530-sm_Ir?Nv!>n-$g`1#N6{R>sJuie07gjAuHqUCg!ZZx(FEz)fjvnO!~aQ(_VdkvyH zR8|E(e2s%CM}2rlB#>Iic|ui$xF^?6#fk1u-2*fsfs~AmTX0bH;<_2pm&gz?t$ewX z<%)-bSy6?{AAUoaD2pucK9S6o9Yhla*tTR>w%7BMe93K**MPoQ6llmOQ6r&t-p9k0y|$< zv86VhfyT@uYW%M=fVpz5W0`g%l!iRgd+_BIR#Hx1IOfxVUfUv!?@+#Cd8Wav($$kd zUXfQ`B0}g`%*BUoIcTz+#G##TeQkiH*Mc1ZTX~i@YypP)of9d(Ct%8}d|J(!RcC@> zag&Spc0Y%9Y2opiUU}$#|HA2LXC^Xk?$+A3=`!|zT9lc}z;IayU}2y>}NWVd>SwxV*D-t$_;6D6WcLR?gD3 zfC}qi>-ZoMpgi5!bH$34FC?+QOW-@Bs|jNgYtq~Md2ry?#;nQ-;yfD3+B@gY5rh=7 zvDs!1!Uw5t}KttKHaVU>h15a|z;Ow|Y6yiTIhO{q()uZUT2XK4~{$aJ-$w{=v!j zxr+oe(#cZ;@LiAH$ay1QK=0wC2?C+ ze6I<1_+P(+#L0Q_Y{tNX{1P&wj;#+mv+&+b-KMOeBDItFynICZRf-=Wpr zmCuRapCE5>Og2l2cz(QP-t{X}Ftk1aHY+V7z9*`~&gI1$?Cvp}<7iBP3CF9pGvRB1 ze6J{c4OtiH{+oZLZjHvS4SqY6sefRwX=}#89gDHC*m?yYeUikt*@M3O;(F-0b6~OS z;Bx2}m~(c}{sj}m`S3-25^Ku8+;1)+k+=_)ESe8Kk5lKKc>KJ75<2Bp?4M$nL;8yG zJto0_NF3c9@|mf@j|u4!{w^@%FWs*g!rEtX$S|d0g;yf< zRnN-JCbRI%IFI2A6VFL`E(QPA&Zno&xSJmnS8sX5@#$I9qoVT?YprR6|yA$`IjV(Z!Sp55dFKW&;kcKpLBVA$bl-gAp3%FE65wx+M{Jm z=x>cL;Z}HRy@SLxwbUi}aRbz(#w5v`P=G48bM@7WStRzK&Mq7i{@5|?HZxJow_u%4 zuf~4Ht5Ia`W}_bBm;&mSUuIXL{+(y@md5W3Mxmd(ViP-cAB=Zj7MedI3sV9*Q==Ri z*dLt)OZhJS$9Rq2J8?M*7+)@|)9kgdVfc+z8;ai1dkojlPxmdnENVS-et&{(T{VVKe zUdcIQapRKwBKl(}KdtlXqWUqY|NipXFDv4HKy}>WY3A~AK|fQbH{B_DmjHvXw#hy- zn8d!Vpf0R(p)ViphD(L;Yu3FGx(*n+#=ElkQY7|cya9r#|N800)HTjm8(^Bb^wr-3 zc0e7=KJ@st9X7mSZ#!K_huJqB-IEP{ME)Uc?^P|Q7#KERe|e>f88UPV?i|~G1>>?m z*EU}#;_Q})qbF7bMg!R`EkDHc8&tGEyyU*$6MOV1la%~hNHdz_PfYWNj(dSx;+~#B zTf2XY^MT7Gw)`+6U+pOR>^cQ1+n`Q4bp$(kuD!!w2dOMm9N#2xDt-`_HJILO@G@ zy7f_~6%4s**vPvO`t6f)kI$d`NQ0*G9jjKQ@4#v9E1Co38!-8<;k5~E2FfotyHK5? zfIf8g5ha7k>hG$h{BJnMKLOL;^wZ;nh4|pPE_(;&f8@o-rn>hY^>_#wCSd~IMJ_n> zVp1;9cm<|Z=X&a?m0?Yw%2rQVC3q{VC3(`n04rrS`9;dH@;=Z#T$q-0?jh6{mNtV& z0ggqe9;SxpLtd&+&(t0fn5p1+VtiU|K|3aRcDW3m@yAKcRM{X67Jr7svBV`qzdRUv z*546dB^r+%c23f3XEs2W2m6+0y#;s#*6*IV<-{9B*^e2)LcD^N1Rkwsew#jzkTbMx zgBf0%wMhR(={E{y<+HFdEdFiXLtRMHkuY!j{_p&+8N!Z$x1-Q>e-A!$IQp3n6n$uWBynclw!q%l&T? zJb{6dyw{qZi?K6sA}RDLi!TN}J4aV%f4&0MwDvoEB1R-md)XDb%7e(*d|#g$&cbD2 z{|~p`XSOC#o#}a2cguh9ErJIA?kwoX26A(1XyZbkS8CFyJe9H&H^7uqX6j>!<vP)xmMn(K-qq{{l?`@*#5|J<<8>! zIHQYPWa7t16 zV$dm7?A7c(-sx`-Gt>hc12{^6!uX~hHZTm?vYN)(V#?4ak6U8jABU0kL)-Y>3BStm zq67-LZ25xcP3f*PzqE2SiQ|3s@b7j8(5BtBj`Q-v_&xXCi;4U>%xWhId$;HTS()?M z{m4hq9kJgj+i(|*KmWiybcw)|53Seto!Wf=f9(S#Hm;5hD~H6eKXe(6F1Zbv%*0wp z4K=8D_FVs*aUB0fzxFk^XGhu&D<3hfb{sks^6-*`6!dI+V$ON0D}${v`ew+Q)tGdL z%XZ%iI<{ppYHrBr05i|}^v`fYCuYo~)?zK0z=hZAe#l>X?>V%dUvf6*)d@`hD!6{} zzIvEG9MHCW8CygzP}hpq8p zpC9fe@T@c?(~It|Z;1)+Ui*zn&EhT)+s{NkLKTg<%)Sqh%wkK8@i!hmXQZwF+Sz|o30u7Pe!a)}8ap>DlL|I?;$+5N z-n`QU&$YKLu4>{Qi_Zk~ee1mKWQq8*r$Vq&Z6$&C?fE6!lyi}#dxw1yZ*xcG#b9z| zqw>*LXGn{0xHb5oj_Y_Xx=R@}AApjJP@DI=M`qd5{-f9dq5=f%A+csBIuQ|4Q2cbA?Ux&9PlEhawC-gbo! z^$=r)XIn924R6nfq0dyl9;Iv4bA zn#6xXMT;c07qgChmtDte4HUyye)XSAaOm@7(BH{SV7_zm8`0c|v)45*=uB0?FKx%g zDh&m|IDX~#MWI;|o3e4%)xec7eXE?;QI61Q%r39&ljHb`Ory@Ch8h7RSMAQ~PtJe< z-RiWRPx!Edqqj1zx(q79q*FBOT5)D#^LOv>COH1$8IMHsT5Ou^GHcM|;th(o;*`94=!^)_fwQSE*IsthT!>CX#fe*pE4lJFt5TmQYkVdq-Ml#L}X(XQk5 z5=mTyrgDlR>V?|)1D{_aTn>vaqDjcHO9J3l)~!65%8 zsRwMnFzRUV+4^iA%!oq$?NEthEyBc%LtF{)N8 zC9xgYG!X7Z=mZR&JF%~7{RJF*ctaq>xB-X$aOPg}R{?6(ua9oxNk|^web}P$H}p~8 z6ip|N;izYC;`mwu-_`d}*|_seDlm)1CATKZL)xz(Wr5fan7*TKre#~!0fJo{VT}56`I&_cm>e8{+{LDeHUp>np_fG z3-V#u&fcK7C@~u;N6VL;SfE%3+Bg!&BT_{2Eu=(Yf4F9 z*!Mm$POm2UIGDje|9q^6QT!JBY(L3w?Ig}G#v}fdYBA@q09rt$zwYhM`JsEiYmH?WGG@A#Us3C#KXiL3aRFmwo> zHu@ZbY9#d(y>!KL$pynW6nR~BG}Ii)qx}y(8KP&uOocw+FT746*rwF#>ezv zskNpNPN*GRvNnXkFH~HmymQgrfHP$E!ywJlafLp#P}}}vFH8v@kxj{BgMs&}cbm@= zIx%&wm8McLgwDkDs@oSt{FoT2deH2k-vCw_Tui5Q5qbAs*JT66LTe$}W8W>^tRk3= zIs8r_*dM0HEcH)?RRHzls1Wy=r~mHL0iVyu6ZrdHLr&UHst=Y2MWsc87}j&??9q%m z4I|n+et(nMvS6MS$!2U|6g1wu^azM^5Y4>@i2bzOYe!i>ConPBwr~u zU?hoOnD|K+uK|AS*+`a?WR` zXh<_F;jo5znAUo_`2ojP80kJ{TCndb^qh+@EgaFo?2+xy6{;Iy{JZvzJ$-%9R8rQk zw*L}|Tl-qqrMvpreec@bnubkCdAIxPk3LxOSQ+fCojYx*$MOq;*1M%zn%%3BA%4}A7HPVmA7>tXS=YHi8yTiv z4@9Vl^tQ;88@m?h%M$s3>GYW#<{O}q)~#>3wCUf!;)Yu`zlNzIv*^*qgnnh;amLjB z%d9#WQp|q8@iO!Q%5U4Vhv@HN;7Q9}su!WdG{gD#iF}1Qj%^LJ(eE9^PRrYG4qqrh zOgHZBlX^ki>-H;2P71R29T>VYp)2y2$P@QI+R62%g!p@^&-u6hW*HKP$?n~Xx$#Kz z`}6nw^_$px;LxzaMuLylIKp=~z;`Dyr)Zqj6NFC3n2S@0H0eBye&G$em@AA!Nyi37 zy{#9F%ip7`*0hGo;6!@6ca4Mu&;@)WOj`Cq`w?yNOrHDLM%E}0(ssf#2N#RP+b@VC zA@V+FSy0VZ=pM@O*yh%Uovm?8XJU!-BIeGVdH*C4*L3m}h`#h8=3UANz0u~ZW$(+n#2n+h)Ix02;jm#yp;_Qz-U0aJ&Y1g>moEK|c$Fa*c0JioqUXW@eP zk6TH^pECUy2PE&;QsA$_Hj%<#>-;XkNKTSb=8k{s=IiI_h>!b$7Ic5@*QeS*+dL7t zE{xDSplMZ1?OJLI6MJf%{KLj^0zYnAYFP~Q!Sm6}KM0f9J_Ou}k5Gh~UX4|sH#EY` zlHG=1?$^Ma&XeGvr9nttH1*@CT`7t4PW5T?{C(JOm!P;v$r#(GH`vBTEdu(=L-wjI zUod5H=xcVXn^+)Ve8KuiH*_%n-roL*z-9|3?NR5ibKzjd$xXKtHj|bd z?^+l4bUPM=HQpWzxrTMI3Ws+KwEfTVWowY@iYqF?jGtePn(zIGkMy>c61Hlra}#UD zHQye6PWYG>#1~w5a^-}vHM-;FxyKfa7jsvcO-%eG&bhg0A4z@%^z{~P$3CqjvHMs^ z2d1}Tmv2&??2<`{-KAV^U|fwg%PKdFDBgl{`F%=GMy^D>_1cQ3tAX%CBd1y`Jv&r_ zt#8`YBM;BPT;VcNa-<`PJKBAx22U0d*EVlX)UtKQvD$*|Bz9t*cCNf;eQ-8CMQJYBO zUext;clk${P#99;TKNRJBX{51?rH(F(lVdC4jRnGr0}g#%Ys^|9woLPH84%0ZKama z0XdIuy={(;lok79-drd0*~~k?duiJUzgL=fk2|+vC=A^Vcdk7}A;v{x$FbVDD^Sxo zc_;PR708HB`~7=W5sWQT)ZS3vwO|}(6~@A(Q$2A0x2E?&+cVI|{qvxjnHDsARF8f*c<0f-Mar1%(KxZu5UX;;#uh#x0Z7SG>>G2K8{p| zI(D>MTse%Lp###XD~)hG;h_4v#FbdCR&y)-7r|4Suwsa=`!WLAC!~9~-Z=<8hXgLR zS+)R0MY(JRZ8c1Cu<|;_L6vf z1zQ|@j$ldl%a7HJTpWHf7(rT}i9-?v#f(Kyp{08D>25|AiTkv~N?BeNtm$7Zl_dTe zJ5|3PE+rZc^>3n(Zzy8^4>-=nGKQuq+I;zJ*{-?RyIyeC%E)6o#RWf3hO; ziz*IYx$?kP^xrtfe+pZ2?!CR6c#C|N4%@J$;tjD39Y;RN5^_EVdu^`!Bqm!NgOZZJ&aYa!9>WhZQe_QNSqyag|AvB zVbi46Q)}yM&@|(|OGJmLe~(A-J<4xbjg^|utEtaec?}r+vCW~~t`pinFxxpU6LCY= z+phLt>J<`?;?0XDi+Nx)aqE)Mu`pnEeBipio$zUHIJfsO+pjw~n(#V|q<8^F%ukqY zTv?7ZBV(G!r#z%wGUk}3y(6*L$wnUJIRO*@UeQ=i)a8a0=MFX51(Em;Ne@Ik@52H0 z9aj8>FX*Q?4# z_`FZ7){{x_H!XOiz?1^yxX7?)|n03&Vf&j~f&99(vFF9kTb90#&!UwP>=L*x%(U7xT(H zLZ$oO%5Nq_y{t=Dnp1?0=%4X0>kHM93-#CE)~EeBbs`GpBF`0*3pOF6j%&r*iFzEU zK24hFze?=G`Uh=>&bLB2+s>LBJIa9?`Yr90*%|0v92)yp?<;2RZp-;%`~*9e6(kzB z0rdWOaE|L=T*-Xw>R>8G@XSWtul9!;%MtTNR(eP-h0tvqoNDbX*U*Q~EFM*zId+^a zNp6U_Na(Z-o$|Pe7Z+o{w)nEg!vw!EZM4Y2pwkF)Ltoq<_>cQYxAiuWTiW(m{i{AK zpqN$PBXI^Ew()djtsBTbZJ^(8^cMSD_7<3K_r-<}mz!JiGl2Sdc==*&QK(7Dzp36~ z4a}FN8&@U*kjpEWT?gI(V_lwt@<^yksU~rx)rkI;%o^2nR;^ky!NFsY&j=& zi}Ud~OqZVyYgQA1UKa_0Md~$}Qq06d{N4##lTtFg6VrAlwTyd!#fp) zfz&A#9D2RQ=@45t2C7T{P1Mr*Kfm3IXl_I7cN|S=D&I!pn1q(vBX(FfU8vXKA%MfL z-FDeli6IV;8b1sAiR}q`D>CKZ0#nKVr>jmhiE}MK-Qi0tiEaNmZjZ#4+gv(tE!{K^3)SM1$!S!EZt%boSMeLjILF)N?fFH0kF zZWnS&(tZOJNA;)GN(5dmZ(FcGml>g3&U{<4h?>6+YcGzzJFH9O^*i>f-}}5>9&0=P zs^z~X^u=qA{yxUhTZ?m>1(z*eyWoBUqfSh(wYseJKoWdBd#Z}JkHr6R>p1flaqbMQ z_IJ2)cPUISCG8E(`2;nT*bk}_B~Zf@aJyc07%7`K#1y(F;k3!ECF0e9Q-!al&V`Qv zgCj~<;(jWLjcKx}Nmd6dt}Ismb$vThw5z#lXQ;=>TwA)es zFVHs)_78j^aH@=3>Vl=tCvo&}hDSAd0DDEIkGGZ)e6#sA{^!iDt;TY>w~xc4O_13% z{L75?@4jw&`|Hf>y)c)h@N-8X7k2Sxo1D(whEuBkyEq#DPw(N{CtK1jn30Tluf{!r z6~Dd+W+{o|%HU+;Xv2W8d91GZmQ=VBnv~NlR{Z+!x z(A=*i_9vmOVs|kN={C8yK2UaY7HgrCn~Dk(UfMaQdv^T=Nu4)k-Edg)Rr`iy5_b?(M{Sd1@JZ?>!wSKHdFz z`+ywu9+b~a0WO>h%X{0-Ee6cTx2z2qEFLcVT?4aigk*6N+F2FcY&#=u3=|KGhlGo>}+k` zNyy7E7?rEqj}0y!W$F7C_))`Tdrb51RWJX=MKiAg?3W<9F^u^*?_Yl@nR_pDI1HUv zz3NGUtoRGN&$5S1Y7xAP)}X~RcP)uHf%fL@B|}3`;&1*rp5bZ-%vNl-b70R}@ON@w#9`w|H+?i)J-K>_(JSBu2Qsj z+d~qk|4;E2zEYU@@#%_h$`0sw|9F`o{UyHTst~qvTZCozp0sVcM?y;12l94lQxcc& z%&ess8w};2TQ7LZ4W|38E@bYtg4wWpr_D>bpk_++xKj58p#FUOvc#bp$-y;N#k>4S zTslSqxeF^GRCy*QZ~G~vj_yBYka-lzvQ)}db;^Q%b1?kWu2WDB1;Mu3 z?#at>%HxaJlS>4yWw!B5j8`x1-?{7Kbh&r~PJYmPu}`cL$49k3`fa=i*$<@-`c7|x zXrm1&dD>Cvf>@k9Zu~tKQsVMt-sqY zBuhYdOle`;i*TsctT2*3tmd~6M%->@CQuEhB z@>#>5nytCmaz{VRFT;z(cIrn-dH7$**Nw}6pnL)P1g?;GAMgbF)^PP^egvj-g2Iir z>#_6pvtvEFB_u9AzR|BQBcLPc(MRp&n%Ft9U7%Y05_asl>%C)<8_}*gXH|t$7hqCP zI8(E+0{b%CthC8a(C6*H<-`8(z_{nRJKnq)+Rt4pc#8kxo40y)SJuA)nx$u1=cECQ z=(pxndWr${!PY}x@?CIX+tfXI=@6K8)Ld3?^bTf}8`*ja?2sOy;H&xa4>q1)v*)CK zUAQlBreU?a3G>0Oipw6l@(%5IW zJ1F)$ac`}0afuH*GXqR+2`w}Fan}DA22$r;PA0>w|LfZ>1%w`vSB-<4{_EQ$K2`Gb ztI34U)8GDchjz`}B=#c*E0YUN(@5Fj)**HG0*S4P<49?J8&>N)TmM;x<@baoc8lL~ zejWm5|FNxMOL>WXD0BGl8<%h_KU4lvp^wmyWSFh-bez42lf3mkt#a1D*q}EScHtz@ z9m7Uma^*l;*U#@~MynxZvDL}iwpYU2@F7?IZTvw$|+}DLvq6p!wkFiq9l&UG31$Xbx8EFZps zq60q@5}&j5NJv~Bo+=vJQqX1N-S1KZ z>dEG%y%5me8eiCm9sqEG-GaQW3mR^*m55qUtjE^`=0aef~cX_W7lnuY_ z`gL6gChe9f_n!O!Z3ji3y>V}buD^XkXZWe`*C4dL@cbE^D}Jo}*@F|?9n|F1P5rRs z#@6B|cj|%ecJcIyp;)A8E_?1aD+psMf!d?E4`=j7BtISCL*_wK*9V8rz>H%ld$IT- z98V29baB)g`*;U~_^2#A3k)#Ni4EN)V{S{C_3{cGY#jJGnI5nLm}7?zFQv&r&&Aa1 z-ma^#yLEG&w|zB^HP}%&w|nE1cEzz+CxXX1MgM%w(whR-wMri>_(@p6HqP_5L<6gq zKYK5mt^~b7=QxG$E{uaI+{=&AKKutyl-KT}kW2=UvvYiY1QNbMT}OnB7OiT>DL&H4 zv=CO@P2%Qrw8`Z5{h#xqpQGF4n#??o_}c3D?q=z{135;_jg&>_lEVXODM6yuDF&fj^|K}Cx zUo`dX>^B)KUMuvm^JX!2cAeIkpRz}$$opOUiwNEI;Z*}-yEGqS@u$`1<8Q4=?38CG z8WR3Om*m_T@7sjlMD-C$o1PRG^w(!~Pka4=VYij(D%<|T)@iaeiO|uQT=B(t`N>M4Mek0RSxV?% z&xr3oHNvBh6?#@@{$7dy7ZeA`Jy99@#d*PP@DJ&F?U#g3gMPQRq*f$s6x z*SJ%}EIJ413jVDp&b-61Xj)qTt5eXHYWJCnmyl|_rl#|J3U)3Z4#w7(NN)4ck}s-* zBFdZZ!utQ>{*w}Ee(q*a$Trt~mrme(3Y0djEtOBhigsP+)JihYpJw;DyIsY^*DUtb z`((}|rc*A&yzXthSeI4K5A`qN_Lgzn!=d^L^XS9Zp?R?9MS~!LYwb7HIPMch@UVLQ z*zauCE=Pt>({6Ld3h3`}j;I?hgvP*KU5Cyl!*H9_*4SUWfqM6b{7ad8IC5m~_=ntQ zIHnhvRQyc^TjpDqZtVYeuRDD$a{F~hpr2{EOid_(!MH~UlCE#YN!iEkC6cFrdLuOP zPs|{`G)@ieiRi*nv(3fty9r$~W~5SZ;^KT@9vBzxBC&AaNG*QIAzn}1*K0h3HDQI%FPKpg$63GpPJLT|Lw~(^13tG z`w1O5+UJKt`>Yee*IM$v@JB+=DknN=(Z00bNK@hwcyRkO4&JM7+bSOflSO(wF8dE; zu#wcZsYV_9*N;?z@J5#J5lo#s#asD#E!Om#+h~0z>e=Kc{SziPFThCoZHwbF2Vn98 zC;j`&amXFe+Or{_h^OcVlIwSFBi7&0qj`a4XNd2o`nU4?Ub2MgCc#}DEvw+ql_E#K zAPb=JFfISld_5ma4-Wcru1pB+YJ~pjaC7zsR{kI3vX$8U z+@_H_e}jH*PaTQV)UjV-l;BHNC++^`o6Ccg0dC_U+dk}xGCgnM2T137$@XxEB2*bI z9aR2J=wDJEdWxMSq^I@?r1PnBr>2U*?1{#{Xs9C6)B>BP-zJ96bl~@#L;C;eYgc#CmFb>+51R#LPJR@824vfP zs~YovLHnXZtzqB$hhXmg(@9fDKP9~9`&iQj-y}kzrPB7x6 z0t!5V=`zMYRpyLjt|7_i5eC@j|H0P6g_YkUaWB_*)`<7{?|p#v{iV8NAJhs6+*bF+ zHD}8??hx~w)KRy0aDc?7U-UKb{u+GuvQBnDaXWT5WLgT6C9oq)?=xxbLkths2$3yW z$lvU{v+IX}3vu31+hQz3ISOH5uD&U~vjdv%*@D=}THq6piTaBq8)Sis$z5G z3CwHP@i#o$h|MSW$RCjXj6L}?ZSyA&)k(>QT+9v#tK=_8ew9 z7G=aXT0pxboe|T#jX%LHa;n^ymSPk_}ARmjCatS<3%iVytrzhm0^=M#7H<=1qd!&1}OAz(1MclcO?Q(mtBJIJ?mV>)s z;PTRdn;93OB2Vk9;Xy(tuQOIIIDLl$P*CT{q8xXi(g|O*#+jl$wF(3drs&1rq1ek(z<3g9sDGsck`?3AEK~{o z$@bLhk<~5)KfK=7EoQyoTc9{-Ufy|O9Sn{x;k~?b9d_`o;qh}O+BY(Hc$4f-8K@Vq z|4X{f(y1VECQ@G1a1;~c`_lT!$B!(YJ~HFz8hG zcHDb`UMF^>mA+de!@9RZ`zD8A)6E3Fda(QGamKn+IBp)jNA45h8`hmdKai`t_TPDG zSvbYB5sOq_rZ0W)gSZ!ruZf*Hn-5iSfnn)=Q%EbxyInTR4x`p;!J{++M@V*vG+dJ3 zgX8iE-3G^`u|xThp6y%(bgv(;+0|o#Eqf&o-F?K`S4o25TdVjgyO3_hFkIih@xOl$ zi7k~}wOnBqek$}gGwn|!)l1SyX?y{W5`HIK%6ZExN#g1|5G^bchGX9gXYW=|!ED*r zUEDPjFuw9ITiYf=H*Ge%=AP-vJ=iDH8RA^5MXZCJC0dHtd~oU*<7>KlG!)1j;`i)e z=`WJl?+RAaGXt=?`L3C_#X=lc*R^}n(-RZ_*1wyqgxqeVsJ|I({J0OlT@%wed)^hP z)}@8}U#TM9=zW^K-5%^M*SP*J{dNZXW0!W}fk32fW?vV4=O>A8t-v~wLLCxYpqTj3 z*7Lx4JaaYh6v0Psyr>uE5|@U};`9`c$>VxmE!k5Q-S!@H=|*o;HBRA|;S2QQS5J^y zE3uny!;aLZGJ$JL9|Nr@>)XuaU-)Aj`Ca4OFi?eNHIyv~-?r8p&&`V-Ym?abTQbH3 zqM%afTBL!CKZ#9fhqk1>HjW&lw5RA}ow_X-fh?(J%M4l-Y`WBdf(QKgmv@>$WdCA25L_Fq;Em-7zc zEXUa_!P6&^CNY^I@Yofqx5{#r$-Cn4WsL&I{92^#_&B;~=^rS)$2eZ#M({DH_X~5w zwpqhKXW60iK;W@Qv$+h69d2W@$Gc#KyNh7B{p^_5ZNi6nU`vN?b$>ZDzBJjJC3+9$ z>q}o71ha6*IP+mfxn?Ss#6NPW1sv0XTDaZ*?7S`1KB!p}RYk-NQwm!=7A?CC&5z?8 z1>Z2S@q*gQ3-WugXfj0hg3DnTzS%I_KlKzS-%g4;QiqUyAcZ6K5QW6yo4-9-!-K@$ z^8CAo-~yf%>-&deUS^mqxkh4pd0&y<i(wd>>U zw;qP-RmvLKeY!9_r5$d+)DpULf3CmzZ83=>qLNF<`*1=AXrALJcr*jlnIG@MKyvz(wUZRdwaOt9+ zSqg!#uF)Cj`(61NE5Mq2N#Svz*aj@?7Rtr$+s|*meQynOx1*K}oIgn7z3{>MxY`O# z*&-HqF0v2O*Gy&nP>jVXr%U7IO+@~$u7D%XUcMY#E^LsizVC)3LhE!GYxZH}lO3V^ z`-pi?dH3{2Tc`)pGVMO@@nO|VNIXB=rw0{}LgV9*q_u$rpMd@$M$DPy50ohlfpCQk zBtLm0Aa3;(YDKZm&%y)?ymGo;xTll2zPSfVp9;ZIuiI{+K}k5i%_AtX*A3}k5@srH z3j=+vp2BvuZtM-rR|{5tP2&8P8=EI^6sxwzZRuG};6>@`1%nNZM_JD==N9B!zlhM2 z;;uF8-8%J*#5bioRri>ve~iAqa_61RO%kuT%iZP0ia2m}@YU%%CL}hE?DAUw_1K}& z&;0AK7bg;3KOQR+M*0sqHJgkN*dg9}qgd@<{I)82ht*gGbbe6nn>ury#Pyed=zGb( zbC5A^{C;gSG)K|bYA(5gi4mGw8OT>B2o3J_Yd4o?VCUi}Lj%cZ zSjoy2dJZNU%}BJuJYqFy6q?z zLiloaEPKr3eA*lC@Em?nxs%0*A+d+)hiyEQg~PdJ+S&%y&=;}cv}VA4pso1hsWGDh z1NC`YoTLDxRBwMY)>jBaJ8Cyu2pq@Jqa05)idgXmHXYVjOuefGb))&G-&|wy2$A+# z*VJEk4kng#5630lg2qx4J%!lc*dwnFRk0so)U&2a^X>&~7izVMJ7tN~+dac;$(x`{ z<3*wT5+0zxIkrbvd=XNYd|K3c+7bp-a@0wpiZJtbDx~_88+N?9aoNT<8@rF))9|fu z#J=)ri~i->$WXf>qG_*$MUEq)&K^22y*qe!Ci+$a z7g7CWLu?;M9!~63ct2X92~~!}`42azA>&4Spjg#^)UU>VmfGKJz;30v@ko~sKy!ZR zG2z#N{dbZ!Sf6)-MychCf81=wp_9My;?j3WPEswL7WfDZj=fFNoA;9hWv@4CB(n5l zfqWy!Y^fzL^pszJE`8YoX}vcs(j_h6hd1M`_(ow8`!e>kO>0eYc=}@!ZJQ9x*@Dk;KHjC&=h&<9g>B4b5&k zOZ2?*kYXczSY%5V)MqMPsxBmS{pfx6Bwh7U43+Udw&~CVVm=%6)xZ6??7!zZGU)dl zJ#R!K+1m7ze&YDQ`6exJx;PQ~ID{3SCQiagjMk4MQrVEXd|KBxe+)Y2_sjaJ1VQQ5 z9cuHz@ksSqJjZ=Z0eZxxL?iDw4{LX*N^Tcu zhoWU^AhkDz#CEhSOux4PC(PXX2LI|~@5e~WRy#gmcq(R2OIky1!4-P=k7%fy7JjkD zpU@xq@TM=hXKE)_9t`Jpo_&hcC_$yCs%kj2=9=u$4h3L-)ey72>4=kIXH~dwN5ep< z?;C!`>kKvnYuoVC0yyBp^^51?CZvj8-{Rm|0|SiernhnC*s{;+MwhJ!PUTLBT{7dt z>Q&X0j4&f$d=M*Z-m;LNPj3p<<+1yLQ;qwZu6oKLV`tq>BcY>Eedk{B`TfNC)SH~r zul6h(sD~@IC)rg4%_QJDwJjAoPc*Olp~O+%$MVjGliCDf;;N8Q$ zUx@KD6z5xQ@({@h7V6h;IAU*w@`hbGme}#9@SuZ)57JV$r#OWy(4U0ACZjR-Pi`Vb z%jHb%_iSi$E^f-{IEMZGx|;Ui{=>H?FuHpuQCG&n7Uea+x5lvk?&oMZFx5-=izGk) zYBwcl4-H*`J}pnpAf^STO= zx4_Q_TwAW)#VO{HwrQ@~g7&vL;*a(nUC=A$xZXKQ4w-xMyl$VMVW0awnfhoVUZ#m} zbpPz*J+UeSyR{IThIs z^!)rc{*AO7r3=>~c|yS;)A|U}PY=%j`MZ_yEvUb|($u^6Ba|tGAABs#ilcBMP;?dD z{v{MNm}e}{A>#JI)Puge#66I@_Ya42azD{t%E>*;YX}_q&`i|W!wqL}U{RR1**M`d zIgw|2^RiVI#Ji-Ya4p&l6ZWSMN%0#%YqeeS$@>#9b@+JXm3yrE8g~6@=Dbd7LwfOb zKIuENkm~s+L8NdFsK#@~+Yg0eSNOHp$szpEc`HSrLGUe-m*?I7O(ODrlVVHQX8S(l zgwl>{yUIjipeS19qvjm+3hQlMG?;-iwI}J?%ml3Mt6MQ~^e*;m8Gip9&f@!#*x8r7 z*cI3UpJmojf1WIYa?d{38126}Im`UGzHgj}|3_D>cN1QYi4LgYn62y{8O#uu)fc|Dt^_pgdFReV9ToPHg?U&FtWM>^?L2_OUc8Uj<_i z-@lQ6{{$(cVvPcPbwC+(`>9>_6H>JV4WB8PW7kK`D|)wDaiA&YPHMa)(s>2Xnem^3 zv5Ru2R=XNw*@HVyPLJyU#YY|al;xWL=I?lTz-4u;bK2H!eY77s`fUp>gVrGJ%Y|Ru zC$(Te?ZN6}Hkqt`o{*f!wJ5^>$h1vI;ef$7P~~^4vy{F`*S#g z$h&twU%BML{V=E*?bBPI{;&Nb<1+Wz30zU;d4oF+^JUmNqtn~%^&D$At+V9x(!`L( z872XfJlI^bGqp~&A8G5jZ;q$qQrlXi4v@H* z{I5<95Pn5;Q~s7uYtQ1~RJ*R~PeMoHJ=J#iIZlF4F=_oeufNX>$)`50Zu8lK6YDST z^Mo4g?9Hf|>0gSOtLh+`?KRRn4J{k8v#@&KPr>A@e{q$(*b0NQ_BgRkJdMLw1P6Hc zDO7C`#12wdk$3n3Bx|m;8y4G#xJ3q1@R&H6!`ox{Z)}R6Ay|U|(aL-B%EX3(CocuO$7 zM=!+ai8fYBi@14SBzO-MNwQrpk-!fRpS?1CVc8f^kBc39+V!uW^KZwB+;c>Rs@%>e z5n9014|5U?Vn<5or(5Nwm+m%MWw56lX>}~Wg43tRJV*BudK&d%K9Layw#YDjCeEZZ!uW$5DGIcEI8iaWY^5}T zBdY)Nx#vOHP8{EOS*+Kr8b^h)H0b@(B#yAB3dZrqSlweN_sT8-Mox$B5%V-fGE=5H zPJa%1P8se_cX_h}A7JYOd<9KupC!^&IZ@eY&0P`lxleN2_ zXRrqj?FoqZ{NLjP(+v+x@`%KLSR()HpX4#r}KUxHmGC;Nun3*zZ5-y9I+1&mxv6 zM*!n^T1lip3(W34Nq)zWfp#fs+^=B>)X_AoAKqyi~Whzq|P!cM0MT1#}B2gq>W-`z7JkQR2 z%x>TNc>K?q44E*?aA^*LtnBUgPtLXj-Re{Wq0?xUc(4dg@^`=Z{xF zPA34Lz0FxWwGUA!g-Nl8Zom}BcAA9CA|iW8-k-RD@o={{4R27Y+KxV5xkKt{C`1$- zGfpBu9ksL=KTt{)Lo)}(v}9vhq3_6Z3fIsf=qKr};i|#vSvt09cyi0xL0{|+m$2IB zXz1-ao-k)|M9`CZxn64!{;24@-AS!QlyS~OgEuvBtd8mj^zE5Z|JF*OV*ZV&0he5^ z%*~4M+UiS+@{MSL>rzlzCsse3VB?>eFm(WaauREJCuL#!lT`L=m9;Pvv_W$IUJsfb zA93=yL4!)R8`0Ot-JoaY?DV3ZJxZOkwLP~r8qKlto&63*XlBub`&!yt7#nK2f5O?E z{{6#h_v_w2pzCFbvhC6IJ6=;5-tKeS@yuI9S?_s83l*V`vII3b``w5fa@u>__y9~F zkCQR(^acFR#4*d7Qq*E?EhQ`^l_lQvv*ab@}p5DcB;t!aiA17imy2_kBCRk9AgvR zi012Gh!v&CA$&;Hg~J`WxYcO~S-&0GftJh_7I!KuAbguzlbYOm==0hv;o8@PWAY4G z>XMp=Bz&DYE)U!Ek>zmpX#*(w9EJeY+($pK#Hb+}dDspB#7^!p%R$h~D;0b4&alkqlN z-H8nQS5yCLuA>E;U}i&Fn%}zp&^s#h&V8FPjAslAt>Vf6@}i;5z%M!Iip|)VMa1xC zrlx!!2U-Op>QBEbkO5>rsY}#t^nY#hp5p9^(Xk^@5sP7uUnM zo~m)eak|{YoZY1$`wyi!X1ZG5A`f7|1B5lL)h;H&9L?A@)RzH=7S%>dGbtEu85TbP6t{a2cI`i~Md99#i z2|Z>%3=i-`WA@*+8*ZO7)be#$`WYg9OD<20_| zO~&jGnvLt3CyAKfYsS)^qkC*8%qrtqdB-y`J;0O}GhB_;(R>wlJf#;w|86uZn-x>K zVG`>n?J83RcNY-C0#br}FU#kl`58X{a>h%d$bHqGl#ESEp9g(+#`JNd6z9@Dsl zk-Cd{(|^7f$9ynWRqjbC>V05qXuzupg9b*6gnq1!lqhqyaPO0U)Pr=6@UGc!hnD#x zg&7hdX94M0!`*=l zIzJyw?&}Ruy>9^%C+sKnsejM)35Uupt9Vh1W81T5(%vv4Qt>dqya)~MA8)Mtu7h&3 zpFBGjiP?XBC0kPlA6TQg>=S*zub07u|By}U)EM+lXP#(88mmu_*RsB7$R&cOpu(W$3qbD z&uji!du}xUn(0c`b_{2INN!p5KvDu)n&vTor@S8smwOE?D|&%^N-HVy^4wp&+g|uv zcr0p<_%7IZ593j8vayS~GPDH5qn}hpFY}>^-%~k*PfCDLmphC5cm(>>Vs0s97^0yy z`AqvD6o_e;PX4@x;lPr~-k(b@V|g0{?RQT*BfepBQ?01jI--dEpTH#YU2z_!1zvxu zk;UvQqR->rR-Y44ICGNci`pV`%dS0t>-i3-)#!8UKSO3%U!cIp{FW0vV$FP1cx*w;6oJRGJtDh#Fyoe?r zv{4#IJW=g{=yJ;HR0jDcRNvTDb1fP5OKaYjs`)$4Nwq22U2bSrf=bXAWkEf^H-9LZ z{sUu`wPJsa80&`{f6r@~k7Itd1t$Y*%cq$*RvqCGFH_9VL$G{##C@p@D9)>8cRVK` z+K;v9m15`r`8}7(J|?8>>T@(KtDP^l)*ThxK7LjJJwU?0IM;z;d4Dw?V`W)q>T-LxP16@Pt0W;ljvUQ=;!}9-|J%TpZA|)=@fH> z046qWx|Ab1^H&aE*64M=Km_W+yfT6yEQhmfteXmgdY6j4Lt8mee_V`_*XT2tS{y&V zi2H(im3?>n?(>H6qj$Y+m|M`O8dvRmSFCPl%0%IL`Z5m3M#{C`($xTMdOvx@oQ^`y zb)u=KjRhh-93K_zB7waNyG11PF;u5gdx02Z{MRl%$7^eM35Is;+Z?(06~@1$C$#nZ zA!__}W}$rpIOe4J>HP=I(SlyWVC6=}x}BdUst$o^+YtFvk*fToB-DMa(0dc^J5)+) z$A;Tt@yPBdG5%ZY=>A09YN?yuYs2YvC6F<=@6`C>9HJ=C4ja$rp>I>`1y1y0xc{V{ ztgLt;Y+Vd6M=tLrPNPP(U&A_{Eb9+_oXw620*E3>fAV(4kITYE#uDo1d=!4#`2M#T{2YI56sPQu)wtb>87?Sb{unV^Rb!VFCdo?d0T1McdVWC zKe@b;FQ9oJaWxO0Ke|orBE^$zATPT0z4|w-KD|hAto2AChVMq>aF{!tVao75V>!l| zCi$8$>>%PH;`0kjl%cP7c47sNc~#Rr(*hfqCq0_Wv35oaFZVoVeYpn_`?snG+OI*B&1qz- zvH}==;U>~ixCpah)xJ~0Q)uzq7N3!hd|0@D{P$eRD&*VvnCIkomA`VJp%nN!8qIeE zesWW9!Li=`a6*rth+Mch`riuX{Qih*N)Uj{)Oy_u*g;(Ahrue_(~#Qk*onf#9!tf zGD@F^<%@)03n|TLNqv3m)pg%s%Jll*Avc5BdqYcSu4)0H+xB;pAQnFenY?(1^=7PXboqw)pBoplSGNICneZs!tRHGBv*Z3v^MK~A;71Wt81B~6E~XbR zdg;7m2){dk!(jss3NB5D+|$SWVv}K(s|Ei&L}atd#i)q^H2Qq|LUx5ZTGWs^#j};6 z?iaqBX{uLBYymRq^rEm<(qFx1F}=|}7RsbIkNAlSqt=l<@wUn5pnntJ`Ddc-XhEg_ z9MN?GYD_f0$ZmZDBgdNL|72F6DeqP5a!&gqGEI8Zbjk*xx(D1HH~0jzTf06z2_)i} z$zB_5({<_Z?PXS2c9ggUq#vbmswbV$pyVVa?mC7yGygbcjXnp^?{7?36MJxpZkOPg zI|ZH}x+#JtQeLW3YB!*%(9)`-SFyNt-}i*(yw`NU6i`C?JuLZU(P-wHLr1oeA*oE@ z$o0A*An|C-CedyHWye#`wGKGwKO20xx>=OokN!v5%t6Q7f#iP3g2bwX>UlOqxUNYB zxI`k{oV$zg(NA(doo<3D!DlkzUgoH#wA5qvs02{%<9Oa>R{=Fn&nhvHA)lSzueM3f zw15|b*gRi+-Q#_g9^XWbNrxhCr(Qx`kM8Q)t!V+OmEVs;O8r0v@p`jP%-&s?zlw`i zq1VyE%*%Cr(V-YlCzZ33geUz~x-YdNLa+{-GhK%DnF8JBCdizTEVecXD ziO$tbSy_)bjT;%f7z-rswJ&>ou(~+1qt8yY+tvT)|6C@UiOMJWEvU68JL6L)Ru9(M zSg=@qoPpQ(zt`7pubDEeY5CXt<{KRstp_C-t`8nQT&}pAVZTNTfSfl!p3q2A65q$4u*Yyu}&?8oF?$;7sZ6Aafw?rwe6Fx>^Cu@Bbr zT*dOMvg(Faeyxj!;qJ$U9k*1W!tmzmC^b)P{5j+kyFB-!VacMRwYsgL7+&=<5f!#_ev=2TO)?enmY3FZ~VAb&Kxbh1LuE|D+a~q5j7XAwkuFTYx=fW^N?$~rm|T^Q^uz(1=yy8G?F_5&8$%@Q>ROi}Sr^?QroG!(@J zyBrp*QFL-Y0iXH-)oNG8-e6vh77Mudx_MkcQ_R|&n?aA|XNR46H|pG>_Zxf6@a3~W-d4Nj-j+$AUh$MW_aqmZUB~i! zu71I>iA37QyvcwO%DKyhA$>p`j}Vx@LC14L6Iv9>lJ-5&d|^m*{QOZwIq)d{&we~w z?z{ZME%ZLr8{U2=y_RXkah@GN=wtj*AN7f`aK25!>iEg$0?5N=+0eB{lkarMit7hx zMj&%qw*zPwBM_-#<$Zy+gK)gPMI4b!bIF5EH&(o_|Hss1?6!+QdF;tzO6grp_B#XWHd z2!m0Fy>CnR2g9_i{PvPwQB>XQ3mb$sq1j(f${zjn|-za4#`W_ms`7F8E81h6h zJt20|;*_8P^t$x&O+VCxqUKjp+ji8UzD1Uz+e6pjkHSgaU%bXZ>c98Kf7S<4I_39T zB)kNY?31n1m3TB$p7|or=o!i_SH9ILOxH_stgg`;H{Oh3c;8R2;`%dDSU;PMCc8`R zozeVTmij*f#~>hVzhK@Nz0MAZv6|NQ+E~1r)_0D{TfY~^kEocQwCYAZcSRK}D0hL# z#4475?gG?ka;M0LC?I^!wwv3$o}mQhjL+ngme6~d>hn{_A5D`ULe5`(wlATl!l~*N zyvaRaxwQQ-ZWVs7?`?~NFd$wv|LBAe8Vh*iTjkn@V=fA6)zYTp0-@Y{u67?BF`VQW zZHe%}B2(NNhdQykv{Y;#WZua1Xjcr5g`o2_{JqLb?1(uzXQto&Vhupme~JWZ~=0k=yiTDrzy4hr&J-sdx%>1D(0tY zIIf~8yNj=#(BcC*mz@N{Up1~nP`ZvFInBTP$)n@_L1j9y3uc|>{gz;t+a*qLpiIJR{$(cCiMvH9Wi ze)8)?I!Yy1a#cO_!1RmDHZynp4xr(VAGe&pV0AkEYXsGQ8(ae7=;PA;yB_|l{44}F zw`~l<^zMno5H^EMd)#XN6Z=kXrsuh!s`Ax|M=ol??8}3n+CzS#vBvtA?^*spxUu_Z z-5wbf-`+;dMD(~D5EK56NCg9L9uO}iPZp2sqV&4P$KTHWtG@-cOGmvtp7RPV9_y5P zX*i5#b${IXIrA1u#g!+{z2AlE*W{cpw3Ej%ai>Zr{yB$+OjX6*(=}i$fyi?%AerI) z-3FFyg@Fq&ASo+i$^8D`-VZ~Fg}r^r27TxKQc!Pv7|5%F-=40;_-(^4Y+Dv)3Wt$T z_^P^Jb*Scn?S;wiv=zUn|7Dv*&qFnsDcxo%mPXHygxap~_W{lfem5XEQhZK_Rj=5; zs{NzoQF;?loXEeglPd&f?GF1W91n+OQn0{!)!l%v3n~25e+S6jKbA;x4E2^Ua7I{@ zsA&)R;gy`hR#HGvnTt#MEP#j=R-OSBm>x~=%jPQ0+l5jU2A{3Z*$lnDKAdVf!O+Ly zlhv}#0chKFZVS4}0RdkY)3dkpubu4EDB+69<egz9t7gbpq;@|fZk$LB4}SJcN>9N#&c$Ab@M zbYV7Yv(DDeJeU-bz29@C4WDvEb+l#$!va3P z^`;LW(8ihLxud>9+KGq;>9eb$LHdJ|=^hp6*zLSw`_?At;J@LicIGI=Z6ZXHB6Y7IU?h-SHP~WGmMe;9MQ1Nx0uN^-~Xw28JLvNc6BKf?P6U|IPcJB5Clw$rrQ8fwkrVq-J@ z0R5^rjG8PT0oi2rw4c2-j7A$=czm-O#yxL5S898V+5<;8*-W;eH$Ov+6}LD+-R8YB z8~x|d@@Pt6n0_PRZFjzz$nwWAi@y^&?})`S2DS`XriW_6D5;Y4%#e%-x%ct*-(<1= zNm$(de2NU@gk`z-@(d&dw}h$t4mx48t2g9nb9f_4hkZRcdq3(IS#cRj>nTO?m<1~ zk3PF)#sY5#L0rc1a+xh=3`f=MWgK0q(h(TSM1NQdfd1`(h$5Auv0F~Vu-7vvR-bs)#8|c zwOxPVXZ#LKUm+Nz5?TZPajv0#+(BMgp5=7^7*qb>;!5vFlaFf2yPQs<2G4EzXK4aJ zbP!gtuXjNU$_6`3E3x=eos@^~zPWEO{e8~XOII8bxGhNg_&T7db*4LX`$@=;o8Y!T zv=L>*r%8|R700o^*u{Bhx)Ajo7u4q~!s^Us-rhZEWpx!s$uAzU3jD)AzMm>()sOK4 z5Xk{00kas7d7Gzd;!j|v0a*U$Tt>^%EDM%zu$T~&1nq=y%xuc zduTS+!?)7-GK{yUHAkDlwgfbnwEB(k-728|EYGjbumRe+lIopF9cWrmlW6Q90Mu8N z-}*MnAnLaH#pWmqnkWyryPXxQ$C{J~3-`H?@rIFO8|<5X?f=@*Z@+e$VtI4FIl9+x z*5rn{s~>f@b{ztO^L#c%c@Pch+r7?{z6s-$w8NWT(Ru%1M0eCK_w81gejRksn-{|c zBe2ADJ3FzCai6lgrr5?d6uSZStan~naU50Pw6`ztqIHA(d!oB-q zI(q!_K(s{rF^E`*i7SzRhGTiH_oFC%1CHZBpUCQQ%r0C!5H&mY_*sZUfr8AH9+g)gfq**Ipc@u*wX z7g0h6iQQf@=Lw)+dQR3)&Q}1D=g^7wFHKNaRnM2<4d-A~cV4s7je)m|VI9Kqy3 zV)S;G=WaT03Y|i|^yoCW&&Mzv{IBQKsZ!!ZQ~ zpFeFx*N+hCYs=;EoMqH^H7iPCOCyXuI z8n&*vpNZkj4d$b(`PVW0qQ?EZPFsg!c=w~tW5wUpoc`(Rq(f!)($_%Rf&Y6@ocn8hwd%m74 zh!2f_Qp#z`I0xh|0rrKQ8o+NYJ^qvj!wDuIbm46v&*Qk%R{4F}ISxx|%U+8O4EW=? z)qj$R8)G&9bsUPL{k3Zp>3LcZ^E%O-yVL~c(9X5VFL(X7-;ZW1;}3kjEe6EhOF}2& zv2$Vgoqo4=mLi(pCKmQJc?`8iTSVRalLqA71F{xLc$m2r9anP%t=Mj>vGSpw+h>4q z^SS4Ob|jQ{e|flxlb)xH`i~s-Q}Jj2*YCu;TwHZ+VJ{kJ-P~63+4^7SlbyC8m1fGA zpGPeZv+u6Hizp(!pA|~{ah%&<725t`!_MP}kxZ&REHD#NL6!^efNV~MkF=*2K%Fp@ z^z|r0Gnqv<2Oh}+F+faB+HxIgZh63TMB5YfhwO>;&pM5o_==sK#2c}G@Mc`yv{My& zxLC#F{INI+EyOE+@S6a`c#Q@Q9@toc@xl*#M)R;(Eurxb&m*QNNief6xqf#CqrE)a zJ@;`bIUedhdEM_+jKwj@+TOcu9DtU{4(_Y{>Ha}9&h-@?8mGb>vGRRm+KPG|jyb2c zWgBh}8dA^0eHW}p)L2g`J_v%>z8w!WW|d*Yap(K@CKWJq#K_I)OEFsB7G3wPo-W6* zaGpaZX^D=j2X!TfpE|u?fF8%~FUxVJsLmiS#r-&r;X8=B-#f&nlA!_x%b}FKV!DAoZnB}AC385mkwY)&G1zj^N&kN;2-xl6m4EZ7}e&_0Q_ew)#kF6+o zBX)rtPwjuhYyw+DllV56W0rNIeESNt`<^w&BpVp!kD8{b$CLaNh(*uSr2ond`D2i% zqi+>zv9)@`=7QnwQtD2Yv!4`)KfL}fDyD*-dc47QZ&3d{& z2~C-L=bzn{0JLBC#1nb9AaaI~ddWw+Ujt3*7D=DoaT$$K*PIs(PDQOYqESXAya@ci z@RU8VMug3qo^QQEx2I93PWQ=Fk!947U*KA>c0Z17NpaoJJN~fT{aZ$0sv2UVvLk*4 z66yU(xuPy3uKvgau=~DvKM(6a0ecd0&W@cZ=2D zXA*s1({{*W$l%}R2k;Ip+{}f~(2Nb|(#exSFxMr#>2b(!L?XO1*SlGZ#_x}ka^9w) zp5D9o8IJ_iH+AB8Rd+iY7+4&T^C7|P)-N$4B2K8KEtquhh7-&v>O|bxJI63@qi0!VzB3uJ~cINA+5mS}!0dC&Q1H0*+8qm(36^E5{zItgp39GE|0uB>)wG2;-9t2x)?@qwE-<}cE3 zHrTrZ(JkhK!SZe#>zbW`&ezWY;le3HzkRJxySghUwY3Y8U#q?;6`ufdh^4#0WiQ;S z3xdKtvMFdfDL|raksb$y(Q=VTF}$}>&%JE~!QwqQ=4Yb*N5Zi94umPFEzMzdiZmwE zvSRtczxHiSWcI}-m~D#cJKtZ67T7X1a>H(-Ik|f&dfD7)=E{wsu4Rl@k77bTaTRsp zn7R)pmW17g+QK=*3qDx>(Og*2AzUV1ABDa)VzoXftn3wjPa%r6Y@vS^1C9fZv*JPdy$U)m1)}IR6*b*q#FOm5g3fK3qwfzuc+EySM*cfb z&HXlV>$y;5dsEJ0mF?asm7^L+iMP^0b>mAr3npq?`8q|=~;V-D@>Ecm_~ z&2p=5IV}7N794?dY)aiwu!G{6;l}`V_(foT>$#!FS_V|6Lxt@>= z&`kWUR-r)~v`Ds|DbURlqlWsdN5}@!L>#x9twQRRBb}#W#1j>Y*;*;i+pPh6~y=Qp9CU z(L}i>f(?>RX=w8Of%_8u>tM91dSH7fAB_6Pt<~I4_shdr$AcFC?Q7AjtwjFbCvV)<#bT>nV@lGACRE`M(-JdE*G;^i`Z-j+|Jg5NuRb2npn z12d)P)qfWE;aJ>qf8V}c49A-bObp$>ade)r4J%iqm4;!yPq}wCtJ4xC)NWmC}GZndJzF^nqR+TV~ zlSFYWONR~~w`0&#SNvZ5mw^|5xaoMNI1ax*mT9X=|N2}M$u)jmk^|~g>e9>N)kS$3 z3LH2GCfusW`p1SUSWyNK2mfg$x*d&U<6WQQu!RN3Y;C`#SR$06z0Mpdnf@Ofh;~dV zNvWU0@caDE&$G9FFz~_yHSFw38aEyH1`Vhl%H1b`;ePo zVa2OUqH&^<08V>5xRD0z^qgW5zW=Vo5&t_dcbfCWvDCd#=AY{So2hZ7_J9!ey{O?j z!+R0Y@JwXZk&6uXmwqhfJCPcSmLudh3-UI?xS9J&_T)=2VB?xxsZGx-01}U(r=tI3 zwD@^-M1s8|Lp!uCt81r>R?NFcCDBDY+HEeQImhh<<|48y-bZt-?C`DaMNXkNqZDtx zMU$c~>g#(N<1fqm8W#fUnI5z1FHKi^K7J- z3x16Pm1E$#VDB{jxsGnLq-0|TKO9=zeXKLa%^szE)%fOW`volw7e-$x+l^{={z z(w^sWEJlw9etgP8Gft007G7a_9#q{7>AWZL*!Qh^G2!IJ;9mmLKFhsPKFd(>rT@Z* zv0o5hBS+|XmjN_&(Bu5p3VA1o+@9t4p?X$H{b0=i)Vt${eNN9?sH@mIXRaQNV;xWB z>(W+3v#IeVlWdcySv2vosp2e}&<;Pd;VT=;<2l^D=9~wRO*LzLg-vTULlaH1+M=5iRoEc$|hl3{2A(r z8NVB44jMGm^Ij2g4`n>GXc?%|jqk7T#_CE-HYskrDYP9;u2o$6;fmq=4;(n}rTaP$ zjihXu5>3P6O^q=eteQ!zXzHM;o`(vC*OXdx=5XihSVUvmzRvQp8IJQt*P`gs5eE6f zr+3!5WN9E$py#q?m?{i*2kLS;5mIBH`ZxxCo7*Jtmgx){I$Ux(e$WvX z1U}^6K9dJay9SgpvnhR1M)Nvd+ zlDpFQ!h}o6VL_4%5H?q&e{-=xQzoQnicTNk7hPx1isvxgPc*U_a_^+`N&r=6W@z2_ zGW2tHoi%lN44bF-f?=0$Y+{&C!ujuaTUh9J8=C!Aps^@Vh4IVo@6qcHHoqQd`(C=ZJXW5^EKFi0U@EKgujo^6+D)hrfFO?{%at ze}D=4U*1zmyN%^h3>%dDum&>L2@@i(E@d9Z&Vgm8=P~8eVQ7>jNBV$^16t!rbMisD zegrcaDNm)I)1YSWeSz}kG_<^_#w=Qez8+C~ajavDq$c#36Wy;Fy#s<@*xdF5iqOX^ zlVozq88z^g1XixO2E+ULAGNX;!9v{+#j?5&*to5&7=HT06U{tL*kM4#@Eqo!H*u>g z??&X~?0ZC}^D%kvxmWaa*bIil{OP0{7 z*|pnH1><3CSMa(+Yr=3EsG<`S`P>-q)9|BobCE6L$SG~Wu=O|t&m%TInR+IXePaK* zZo0cS9!vPj=r1B&lkN&L!S=<5BH7Gp(Jq*}SbZn~QBc3$W{Kul3>UJqA=gQ?VHsxg z9Ygx-=yHMMNTww|<_m(kXX9GL^&UV}<_H?t90H`lXPyJM8S|XS>Vm^=xOD*KW9*g3 z+YE7Q1U|j!iPr;K`(UC=4nzGFwqK0~sNSr`;(K#m{%1BERRFSR{SncDqi9-eK9efr z1O(ZdJq8MEVA$b z8m*0QltjzxHgAr#nTMsZe1RcJHZ;GNmKB%x3sE^r`Z9VV5YdN-9HSWY2v}m#b7ptL z{Izp9*DM*zzuI*`$)za{UU~S}4jV8E@8e_mz0CZkQoi>;cxiPbvv1^?pm($T+#mQG zz|1?%4Qp=DpyAoQ>l;Uu(1^YHEK!w(asQAru;{*l`K$2zD3KznUQio-#UP=00mg?u zP-OEM^n4g2S(^E>dO?A5_?5!(^-y0X_w?vf`hJ2sIr!5LIAiOw{>?ez9@o{V;-)-% zOp_-XdtzksJd6)bO!$3IO=j@tdVCu0R-Yx z?gKEnbZqQl-5;0}In*J)J>~DdRD7qe5RYgfn~&HC{G&cwKwpZ~cMld`y?d3lIum9u z+}ZWq;x-Tqm^TJ}v;2E5yJvoC*tp_6(xjB@&6(98(L+V2>={;1HS2f6-H5;WKkrBV zEN1a>_tyWLe9-ToI%&YqI>!)4f!UKsTR$^jV7NYVTy5#{Ml~4CzjcfCDc#cS@II14cI0ohhv-jr!Xu$A6F@X9`r(u<53{7P#tSqi-0G`G>>doR z1L6%CTh{;w)NyK=&_-HOF9zDz8*z6=?l4^MuNllZcowVYA!N5Nt&Io+{IxBa@^yQF zrrSDl(WD<0OkEB2(pecde{l1BkJABR)MFv2np29cLz?NO>!P?C^i{l5d`r<6pqcEx zBznOHjo)2=R^ur>UJe6)lKY)caiFF45zz!kW0)DaHg}Pv4kIZ;YfgnfLm%ZaPE#3=A2J*$Gr-qv5@L0+r8O|9+p%>#yrEd$3OOk(Kq-4@?eN zxr4v`+=hm%2#*uTG5jh@+r6QKNBDrmH&D$pp9Sd-o3`wMx2R?^F5yqk0aTr1vaGaO z1Lb9=wF~lN{&B+bPhu{E7!LlRqr^cKJIp^uQ0OV#8FK@rFb%cqPVb~&Zw=CZGx}K= zHJHDe)>{7!>j!6VL;99Fn0l*p|D<{)wx5sZX1qyz3%R?Wi`~C%3T<{r0<+b+u;*|b zTu&cj`wnr=5bqT73g#c;?Vr7Kg^7}J;g8CfRviDiM6nOOEwxZ9$L1j7Pl8_NVivZ1 zhI4u4=a(E<21s9EF-cU)1;gh6!WWI~kxDFYVi0agsRm*=SCrEWA#9HNP~Tt6eDKqX zcn~miB_;LFx^+NqHrw(uJq3*!DYMU?X2{Qn+8a;3v*Z0>G;lYc1-S#JZC-T>RL8>5 z+8><9f3?BP(?_1=!+qE}ReC8jsFiL{KyQV3wX#qZEdFTriKjh8b*$eGEhl374>e)_ zrLnm#@Ttf!Z#(;Iyr^jKMSj}Vm_7WP?usZmW zi|)sUKg$-q?QKHPBMXbdnia6r7?UEyN7rYd`gcuhWuFNQ#F2KLWPgXP$1yxtVY8o zp;z`R>%zjR8}8v44-Sl})i=0RX21+?dFSEP3qVM``E`aI3uCVoDrrINs7FGmoV+Ci zidi~z<)eis@Bn4~Nt7GH9#HEq3h}CNmlJNU} z{>pbqBl5{2zs#NAfXwvzw8?g*JniF$8K5#dqZ#+S1CKxCgR9G?6R$w!WUdLtYS#NUC2 zex~<;8lkv3zTiIe=p+f7OM9Z8b>ENha-Rb-PWsaai&CH_ZEO&Hf#pe45_)E0?;gRe z;woKp-zNY&SDb`pa`XOge9CniJ|2KiuyWu}I|XA8vJXuM(DPbga>SR!wGP9#`!&1x zN!^M89|xwthWpOn4}ftat4+qP6qxZ;w3rpoGV*HNovv2bteL2Mnz?R_jB-va%Ge$mzc zu=|Kw@1}xFZev`bm{jP&`}9O~%8jB2A=a{r&9y#OqtSLUduzU+zyIjaf>3?v$@B9o+ zq5jv;lS`HL8~8rMn1tBT&za9)#CiRP>yH(1OePyqhwF+sLeOir{jysY#^2Ysy7(^s z5JSEfOb_T9+`o$PX*RG+3b1aZA8VtmM!xEh(7+wl~>-vQq zPPd`HSoZ!8p2I*^x~Ax5hUp0;hyC##wb>BfmGyZ>%oBy5xRI*rKMC_ow^y;pe+F8d zlaPDeT_CIW%3NMM52F?$<<*IaFqYCI^ERCB-@>ulJ>og+HwR?fb2rLx_Rup~<<}F$ z4uiS&=R{^1^S>r#b{yfgbcVi<&!jH<*utEak*p^7U%gN%@X*5>+AtC=8*9_ePi72Btf}m%Mx57t^ef$7Mm5fJwfreN{>p1|0QZIpXyWZ} ziktO>zHWDg*Eg~J3ewA5cSvxC1xK@ks%l43PulKRKKhgj$D(9Yt;s7a*Tb27$>Sy;~i|gXNL;5ZRj{rJqZhDw<-_r zb_UwrtWo`a7@l(PPT|M3{&moK(O}@p!AA_^H^Emxkgmb}C4}x_@r#^0vFC3J4p4Y2 zfslKpsU+ezOb7qw(vxHb0(*=7Zc_$-9L%nYt#>;$4a9uo<5fd+enP~ciG=~XY2xu>|JK#-gN*-cch$m z%eed<_l{Rr5A#8r$>k$3k^a-O*J}&*eFiVrtiM6m_krj_L-8HF zK=eI2DsWg0YS*r5yi8-@Tcr0RtzL2}3f;#Gjs>s9{4FCgXUlN6{((1Ps9X@uME@S>zo~XhV^bMa@kHoM zOIV|c!{_)#W3l{}xj|fY`k7k#|Lvo+o=;nhfFQg3R^)jF7_L0qqsWZ$F@BNupH0%P zg0a(&Gy-a{Joeeh2nR98>A!Z2LAl_IJy2J?mgR1)FzV=;q;yw`LC;*rSC?H_o<{kD zj`APMe1Lxye{ajOGNOeQzqQ=o0aWpA3J;w2qHc+ux|YsYVd4DcyB#eTpl#9ZcTT+w zYCL_(jK!3ZA9FUlDq`m?hWH=K<@~%o-tQ@p*95FObM6WdN2G?@z4rqC)Yag@vOd%y z-bIN!gYj15x8&@RJo+BStL@pTL02H(_RgKJ-49U7ce14=(Ug8)%cOZ-&oCR1jdzpQ z{$$9@gZWy1xi8L24E^cXd)Tu4{5a|~+l&+U-iBs|Dx3#vf8$o&Wna3#RSc#sO0t_3 z@}MCP_orQl{h_yoPj-n1JBM@jaNNPoEThtsR~AT9*g7U>p1%A{@CyMxSyf{oou6lHTcL@wD1pBnnFn>KM za(vfIJflB`YVzuSa5cle3HXqBq2Zx?RFq*XU8;Q!#=fR}ziayoC`{MGEi}##S6TDV4bX5~!60 zFFYO%)&U@$Hym;K$WI@SX=y9# zuSxzeb@0HxIBg8~r(-}kc=0S|uTL$S7Mh(ghyHB&IQ2W*(2|(VwETcC%pD<4y$!?g z=xVIR-!GlV@P{Y6)3}r1C5%Om%w|6xKqJMK2mV+NGxT%8!)4W%KbSo;L;01Uzr6=$ z9@tFqrEGy=iET=UY6Q@t-qM)|C0kJS9sjjYMlPYD9#*HgxPQlynw6^l?sIWJj3#Ro zjyhugf##<1$$LSeK#ED5=I^aX4M(rnjYhP<0A=$n`xZ^q{0iT`oPxzg8*Y3LC3RzY zYHte$*EODIu&bcg-bLDP^E6ZyXH;<=c>i}Dv{FQ0asny;+Qs!Tmw;CHMdjzIQK04O zQp~?&b)Pi3W3HRjbI|Zhwln881&VG#$!3`UV_~26P?3rmYIMy@O?rA6 zReK9MHFaS4XQXCXp|ic%^V8|@BwNZ5C5ZN9?)9bDU*@w9t!(5&JW0vxO__D~$W2Xoq&C7RJXu`mMUoF7q_%tVth^w2%Or zhC4MfGLhjq>V&h=kV?tl`-#w-LeqgKd?{i#!u+nG8J8Z6U(n3*RmB7kg-Oy4_S|GwK9e>-R##(I~Czw zm0n*m*%*H8OS<)RH^x`qzq2j;Qi40o58l4n8zTjihOa*G`i{xQ+el&F3^|6CzJO9CAQ& z#BiIVkItcHo9{tA&y1l#ZBY5+6Ez&O_bm6v=`hp*<0aZHX=r{A`4l;d?oUDZOUaKl zhJ|3XsQjd8KRqsndh~B{f85;h| zYIlq+nA>4-$o~BG%7u@vK;H2AgW~uYz;lrez4z@~ab1jlJe?}JDV|a8FFha`mB*pc zj(BF}8(3XX&Fbi%zcrQ^`t$X$=q|RGcK_{qWO%;OobtNz)r$V=6`%W|&|ZDz2}8an z`swStyQ1p|j0?JL=^mIv%N!C0@c|f)Om(Evk+Q{ZD3-bN>1{Kee-*|nm?U{L8StQ? z!KL9I+o&9L`;IAY_1%EFMR`MpeiBE!A1A&n0Mc{YJNpjN@mbN(fXYnc zLs|IdiRa?cV8C6*F|nDs9Msu|{^)xizQHSqihe)#gDx!?lhMs19PNXCfvyG-rI%>x z$@7+Vn&D_kDxvI2p988^GS{-Hvx<~l(YP$W2MiqLPg#x{BEQ)=9GIf#>nT;Cg zb8$^B^Fr8Y0sbNoD2wIoMJwWD|9|}pU4P8>?r5Xq3<5#0oKsN{i`O;h?m718T+oWo zw*-CH{B-&*jHPCpuX7tiOKGF&OU6m4m-10P_PQ+e-H-{JBDx^rK`Hg@SVJhStf}TT zv4=@hvG=*vJ7M&RFT3LT1JDperm^e}fdDZLo%+M{{9zpH+8emv`xJn7!zyY!?<<&z zUf?O>KMmB6+&j2pRs+Fe|I3XwSAle(^!_)q-%#`DyJkmX6dDjV6}Y&;5nAyDC(Yl+xHKPHBmp`RF)Jucm zg!l(P&4z&xw}r!e1Kr;Yi^AEgi>J<`!Ew<^*Ij~WB=q#f#qqajaLXIB!^f_`+>Z1s ziSmIk#j|#|H|`6Jznu14H0OX`;;{Ib@Gw}-kngdDL2vMgUIt}Y4to@C)iLN-{Q&}d1! z^*K*b)K^8@HLBN#3ZQe6`S(kd_iBrAb{zrr<%gM(d9ix1Wq#g4J*IF}1M07@vzbA| z4c)2Lt+mj`$0s`XkzS{X#!kGgI%snfM)m5vDz?)7+i32Z#p%z=sc3|+oN6FF0#sh> zM{s-=J%24PSOawjB2|kCNhXY=Lj{eqky&9-jPOS>fTSAl83mQj^9zwTG8Ha%L zS=9N~_ufGReu#VEBI9qn9|n#~6>m*(LDWLY%nR{=I?oNb;$MG7L+&s3$*tcGWL<@z z!Dc0()YT-^S{_E^UoPKjoaz1ysJ)YG-06i=G&NOJ8bkivVjt zl)p7CAGNX$HQOl(g6FNoH)}4EC>{j_>8XsQ(Wu5Se)Z_`Zpzl zxa}1TA3j?x^>-g)R<@h?a2@lL*X3!Pa6Eex#th^8G*9ja;-7Oge0c!m+DosaDUCta zLfk{w2X~<#SFfFyq7PLHy($N&E~rgioX`BTB3iHuUlaV@4N-?jCB#B}QHZG5)XmEb;_+8t-`ZDJ>8)srB%+18v8`{l6SLV^i z&!isI{vsq{acK!P>0VX3tB(2GMjXH@!C(5X-8r%OaQ%w)B|sKxxBr$a1jN%WPaI0< zaXBE}z5cOB(hMdd^mW=5)-pUt8*9tD{bU{E_4W1dE!HsRFD*8VUG*XTug}+iJ%1_V z-!_Ux(et!WPPKl$aD*vPj&D@Hc5)2Hm4DU?eSeMq6dOFwwikfD+{4N_eTityyX8Vp z1Xka*5UOf)jM@y1F3N%1UbVsROt$XWu*=vuJv*ro#9|K(a;nz{oVK7Ll1s$g_NOS1 zS@Z_$*emE%u()xvITyxVS?>v6aX>Q~>vpeyYKEGpzdpNsH~?nFdp8}D3W6^8dmon; zJ%Bd6^MK_g93sWmJPRX4f~ToXQ0=Qq7&HHV=g)>j!1Klw6%YSweJjIwj0-2TMTENl zZ=ZvSt2YL$6fIz;t#*w;=>4BAEuUn z^|!xSfEt@{is>>Ihnwo$r(NJkKc}H2Z1=}^2N-e0CQ1iQ!Ze%F$kj+==|>TW_Myd3gAj&KnP+`e_}=l|P|dtjhP7WaZZpd9 zH>FKUl62fDAdbJRGFHIqxZ`isO5`m(W@wL`cl*(tF_w?jzEj&y@%v9i`?)INX#mE{ z*LO_ujaU0~hUbS5>DUCsG2qFe{+p%vpxlk9!n%2xJ?<51H1fD_ed@2C;(z<9Vkup| z5Y2D!Y1}DG$adIs>V>*4bhcLPtefUXkqVcW^}DLk_^aR}>n~`b0JmXd6F(wQdt0Pb z@6ll75c~SAX)@5G5JQgqqm7cos!|;K=g>%ar_+|sY_#Y^3g%hk0@MeyETM&OQOiPg z#pV7Wn10=UuS)*`v)^A4`y&UvxZTm??6X4ByH?b9EnnmSg=?}n|gmA*9-gMk5SLWw-CjW>V z&VRZkbzKN3R%#2$LX3D*vz#f7EPZr+9LL=J+ zs(%$7f_6jwi%urFC>U4#U0z%e$Ml#z^LSMZP-gl^fAXJ1^Ea|r3CZq6o#&~3-e#sS zZnjvQvV0wgJgQW+okFNl|I)}^z6&sv-ZCb7LlY)%_Y6Jn9R-5W!}V@D+hF2cXKEX_ z)?d9snCH&J1^AZTG7^}^1bEB7xVAt7jJzm$p+0H=1^EXfGXF;(e0*b+P52qKo1M-o zBpim}g55WR)P5pT*of}l2dh!<`Uc+8kIRrV_GFrtdUc3H&@*V?=v5i3oB- zqoc3?>Z3cnx2e7*qhD98BRge0(0DQL4?S`b^i}&fh|aIU4D%WNT-a+1 zL1zziY^=Qu0~UvFF4ty4&!noYxnT;T@dbMnzx;_x&JE`VCDNexw?Mk!0tvNdv^FjM zOsBVp_pw4_9|X_{%Z1;hJy_lzu|8UsML7dC`f$md+Ry@Bk^*<-*cfr9N(HV~dl<1p zJN|PuOQj7=?Gg0&5X^>(kK;dx8YLs@lEs3TEdH+ylqHmn6=;+i!1!)h?g8(UrG907AWrzSr<=!3ZjKh`TQPr(q1`kC-B8WumL zNtYM8!CZXF#TH^UD*JO540s;FxakP_LVhml-?^h->Av6J^Sx$xPx3GfH*xJ({gMX# zis!c5v6jNzy{OU&vuGe)jTB3JR}B3rO23%;vHf}K{Pyu1fA+)Zp-!ID9Bx2TBXY(N zCSdFg3-R3E2_W8lW?tw$0KHCEQ*xRxT(CAv5=nC5|fMB9B5Y5JiZJrv^q;)>c9-*t^_ICzbuA~dsx(kNvtatz za5AJw8=Ca-V7w3HJr%rb=5t`eYRPZgJR6#w6D1wJ^b5!*N^aB;`+&gG_E=BQ6$r%r z@y~*|QBQWZQRc}x7`t2IEw6;tyEp4aa*f)n0de=UhobwK(MWQ21-Duy^joF8;Ekr| zE25cz9?|{NbY2c1NNsw3V=@i~H9Xv;mkyxu&^^RcN^L+6%=R`8{D>A0WnQ|VZ;U42 z4c|OgBo8w&rpCW2rP0CeN8^ZoaND5Xj~XC zD2-vvN9$R;)!V954aVlQ30q;90$NlaSP(yEUQO<1#+8@cbqt>!8R-m@5aV-76 zm23E8byG^sA`58v>bJj$6)=9X|9prRra#u5`|@+!AYK2$G2>G8H=jxXyt4GLp;Z?Y zd+WWsTA7S`$+^CbZS|;e@F+i@3=C8&yFdpmCAaB)6EqJ_9JoJ9tnqP939^dmadyD*cBWHqV3MP3wL!;U9|J-w|UiAe_3A_ z?(ScJ`7{2?W=3>8C)6WxZN|+_4`wsi?~5WVuc6as_l5OojP_=?@2usH7nonJ@8jX~ zUL^)-SiyV!1p6N}U5#!$Aa6yKU3dK0WQZi+s?D0$xGL| zBzo7u6ql;k8zDt>jT%yw-QI&^ancn|3d8U|GlIl-{`9RwqehxPpWnsoY@(GtD>KH6 z0OJbr9hdg_phib_Hf~Zb`b;ZHo7{UDYMxiMb=lhk$0v8~!k%{?Nte$4dHP>QAcJF==}I3{B3|-2WpP z1^Lz{)F*f>{!R$mRiv`%?>;H@=Ha@=7ce>-Gr6tf4UAg2eW_|kK)Pk>B%srd>iWi) zZ67^_d`?d-|A91UrNrK`^caB}vB=ERdN!#0#L#bOZi10&CyNWcSYG@{lYCSqD<-Ft zfwyyHO@6_wms~?rRuro0!MCzLR)r;Mh-{36RZTv?GZ!oB^j7V zbncs1d;-&#{O^YBrrTwxMe|IL*Rw-^*MHcKZMKexy6;&k`^}B03m2ZSSb*Ugm#tYP zs~1P-!$XVA282gndQr>sOXHVS@1uHa+h90tf{16J;_&?-Abxguz0REkqvGjV`PYIG zu`6uXL+dG^t@r=7a4ZFWv~n!IYL9~{;}4{Rq)kwBNx^fsFUH&QO-;|t;YKbZ?NX;C zb?Bk?m|s?oCDN#8Z@+7G>JTbc%6Ra0u?O{pUX>iT+J^eS^t~P9dE>LSpHQ}6Zm8c; z4|)xe;u1yBpQ}lqDH=^uTQKvDk6t3N-b~uTQ_)5`C4m zdVfF99X%)b2CrYs7)O~%enF$XybMGSgH7AMtDxbjd(ShYFg}RsqX9y{AMJss{leO& z2G~02ugUJr6g&b=*H1KQsWd=!H?adhJAy_!4BupEp1?8PY^pbw_>P+0G)`uEu3~ua zfbPYYE;d-b%GdYnbJI7#5UXF_Eh8e3PrIGiz^Q|}UP^eF z>iwP9WA}6QLOKv3nZ)^eJ?1Z%i>@dTa+`sAt^t=NL zTKUD(v${b3;1gUJ8xA!B?HjE#e*k{#iLHV=q+rRdcj>Gcy`B!rHhV=(ZB~PxrU?Hw zJ075!+~A8%IS#WSCC<+0_P~Vku;%dmJD~lVHGNXJ5gWe{l^O-J{jgkPmQc7G!w;Z2 z=tWHQrUF5GU{mFmE|_3_smBq!BF_X0+bmN!YlZop{S-Bh_W2C~!Ha1fC4U=gJU+OA z!0C-*dZmtD*tT+gO^2g-R-*FhCdAxP?S>im(p{Ztzk&3V-*yLo4ouh1u-(v{0K(H9AI`7B>Ojbc z5|0=6{MWdgIM$?Nr~jyC0VR4tnVp>xx2{9UeoAdR7{_FLX3DUm8I2r$QkF1Y54j(2 z2egKlquM=6KeXf>AK}=XM%sxmVr8ra0j|*G zVg8nVJQr%*xRnbsj{-hL@#_v>jEA|NyZ&-wsJoFUQgz0L1sn6luq8(2e&FIYg*Ny>BMA z{EDwY)s5MnsgG4ruK;cfb4>&qK!NhwB8#X>`|#Xu@jH;edi2AKwIe7sZt27ug^viI z{%1j{=rK&ax7ho#zYA4lRl&{YUZ}85I_JlAFBn~0I?eau-M@`9P&=Hq3{PPC6e)PF z@y|D`j-Y4yz9cgfBcJHNfgQsyrBA@z;T|tv0cj{c%hv95m7$)UVcux()PqErOzw>+MPqU)!ilQ@$Kka(Hss_4>X+N0ueX5J1AhLHw=dnO z4dw*)zAO>3fgbB&4?jJB=s&M@@|c|y%$F2f#Yi*M0b_c@_Y%Ua8Uz0rB6CX>8&&)( z-Uf8L{&}(Y5yO7=zx@jtCcmor4(~w|L)|wAH{V4wUOT_1cI|_(f`y>_;#W|ZT9)Vd z+M7UHaB;tNkO9XD=7QggavHXw-d)EH>m?Yz2Z*tIE4I!u;P?VwYX?`bAchxH`!=?B z#x5F-%qj@P9{)%F#CI(pvg|Y(P;V15%S%GvpTvICe|-nFSzcxR_=~QOqUGB&S4V(8 zZYUwlMtJDf7?k7}>1Jom(%btQ?1X3e^b3voM!)pCNw4=pi|n>#b>A_5&)E+q0g6Hl zJO(f+J~g;>aTdt+@*gc94?>;wyh5FUEoAc?$iu!Pi%$+V!5^cuGi4% zOu$PPWQ3lec(Y!Ik;>`i>}SHEe9IQKp0?|F1qHGqR-7b@H4d&jN136XogT#4FE z*UMqbF-$&v{0t0PA2ONAID*=|3KqDvMxZb8!7&}Sqo|_K+WDMbKH$A1oHs8m!CdlI zRh1D<7_S`Dr$t{w_|}mVXy-b>D;3rG;~8l1jq3QLk zn6WV$IIx9tThvvcbnQIl$hip`Hr?B1V@$`fLL-7{y37^~b=A;be{#Qylm)cz+JRqu z5`YHy@@cQaqk;PKbG^V<2EP=vFSQ!jhx7s-cjNG6f-KC~4f25ohDSS^GP-!V-4Ttb zj{eY3nujqwDec4odLBROS0-7!2%yJAq7=F)`jznpPE@#Id zIu6m7)+`*Glw0xn{v1Lzw<7n7^IT)HJlm)d27OtDcM;RS#_gbAelcG%s}*|jFKqdD zxInAqjZ<#7R>bFr={5VGPcg~DNTP>aUEOCO{~i^{JF9|Yk`gxNdH4hMM%pnSV%-Z% zJRghdjUCX``f_H?ihe}A*gW2Pw-4pctQEYY9S#JQg3G*&VE#5Q(HDBYnbfIo7j%jJB`OTNNxK;0Nol6;-MRRHc>vHWqpzWmX z}viNag+m)mwPLb_gxF3*c7r49Fm4B8lcN z*K#Y*NGgKXrRBk3kK)cMy&Z@YbS(GJ ziUvVhg^ayCroo&wQ()SruW9RXif8RM6^4@04{J{^^3!W>J^<=<9 zfl^uA>UpmW^veeY`U<;4-xf<=%DxupJ80(iD%k3Oa821JXUfpGJuE4$yvjgbT~=O# z@!SD%Olk1aDtn;4xVU#FI1W{2cZnbEAB3ODs}eZ^F&u+h2Q8KjgLMBqEIt!H?bt-u zGf_i`K?UK+XP|xAl($RrIn0&#m{=J_~=Emx`{bhm1kWa?N#N$gzc zHeo`pg@k zZA(|-UEUBhzuLj`H$MrwPpCdL`fJa&|A==pXmbFsZ_5MLBa(+u;ObgSpnOm_nhZ_`{3_u?v0ur6 z-;!Zue|QZv1m&&XNiBwE!7IO{F1`f{pD=g0cpa2l+YkuP^&NN zOy}SB1Kg^0<6$megwXsCf#YF$M8@@>@ztZ8YY8oIOzq37`l}zxA3S;662lct%d%g0 zIQ}EUbNFA6Rqqb1$j{4gKgI6LK=K)780%X)X?3j%+D|`8h|D!c3xPk_c4dsfnA8jL z$Ylv2ZFRQ2OJ>lgQIX$8r-Max^#2J5s&q71vG@_$@7UFyX$f9+3+eN(LTd>PbsDO#*t zI21-s7-&78QUsdE)0&;{BM?>Y)b+-&78vKpKh7!ihlYQG~%$QL3oIU^2`R@*6(sdDHLmq7ac*+*YL0oSNsr$GPJYDL^eWs z(<5_A#{?qWJQloeyCsk~N*&9%V_;^>i-KK90Lc20{=fJ4p-RFPs@TLT=sUWuXxAMf zG+Tb{g=GN_H4RSRJ<{WoR?>MM-(Nev}VLB2hhUP#Q%&%Syoia}mk`njV;FY%*^2ZUVJWjfy( zP)}b!KXyVNh?ycJw0Q2X4sS{R-wIm<1^}&cN8gtinXOZ$NcF+`i{?AP`$8nJdsAApSn987I*Q(W$;R zF;&t)6wVoO=rIT4sA0bX4fC6R;W1yxiFpcm%bBCJP^?~wX5H&2b2J%dB0IXGs_6DO zYF6IxCiD%)N8hu@RXo&52gVGN*#+P20h*9n@+}<<_qy-pU7y>hF#lnMxZSMuv?5II zWan5A9!8`XEgtgTGce*8=o|O=5%lOE3wAJ%grxrs zKNqSPY8XfN)1%2lW9;p46sD+J)0_U{IM`!AcjsCl-goABE&W6sUe#%yNR-; zao5033e4{E@Nn0C3Phrz^X^j)Xy*CbOF|<{FdJi^Am8*I&G?JnX8!#g;dSrlaS*N{ zYQYYT>-S30q=*!^=uWyH6nd}xS!HrC5>d@{lj3DpQwzn zzur|mAhe-@p)LqJ=Nzb4kFHmNQcuZr17XH|((h~4pDM21g5Ct?$3*saAi|G<;*+8G zQDw8|?S)&WXncq8yQWRcs7IjjiDkh$)Sb`E>tE}QNbAIFcAhRkrTHzVy~2;d$Bjm3 z%sq8b|E>1}DNl@0)k5O@YtCNOU+!{IXg@#dMPtJJ&QGA2_2cQ+DLQCS&@0(J8KLq5 zb>tPc3-y#3nNYQTP-Vr%ES+2tG#=9NCup9CCSs54ALc1TJ-WUc11aOE8*gtP=`#cP zPp;>l);9sAh3f4*_+R6L(&de*O~uJ~H>S_bB)u7LF^@*%HD<+LVrwxu%wyunt1Dm_ zSJKi*r-V2?PK4%%HtQd_ip5tb?PJD^!J=r`Jl*_CjV&6q%{YGa{YPy6-S6DpU-`HH zk1p(Q{Cp28Iu+vg_!C3xM0O^(Z-cfA6>y;YU$+S@f}`sTPDk65WxG z=scvTlc26+k)e-<0!^glt`z{mRbQorKN^ONv!&ziWB4S5q3dcTbKNlRN?^XEQw+rJ z-JJQ)G7<5)U~R_CIiO10-OkSpgRwu8Z_Y*@Ld6kYPno~mf!+$sg)Mu9(MYP3d5G33 zm<*hCIvI=Ycda8|h<9xrq2SW9lfp{XK)B{8AEfaLnp*6LiIZ8-x;jNjfbtDiDJHGXC}S0hbh&#AEBk0#TU!YTV#n)^7pQg|N^# zCU2m6K3S_8b03D|uMzbp^3dp(dbVq>jCCOyn+iOM1^qDNzNVMk*b1dx9h2Jq2*Yh( zN>!DOwogOjwkbOIWILeN>3#|Io(O8VZn<4n^Eiwg(8Za?9|!85hi7W30yrkzyU%l= zgRQgXo4WRC9{``5yiP@I6`GAY=n(rqah8}mlKmq$AZ79fE#`J2yu9|kD?1*eJRZeg zMVHS3xgse1VG6)>NidIbx*!l|Dxc*WFxbUF3m_kS?Y$3W_KSqM%y0wV$}(qN?l8hr zXYN{`_=sAvj;8)jTv0!!w=>C1`{1^I2c!O6tkO$<*9Qcrv_LM^GSqyA22sx$_5&bt zH>XlBBJ^9`{AJboY1E@~ufX}mX;|>LUO%E@1haA_5;k&eFkx$`ukxT9h=mKY->e1z z&+_T+oBMBI>cDYsTHC>oj0pgER|s}P)c^@Q8AF-emFRF(fO~wbbPTuw+Ltr z+1$^*xI*9CUwcPAFkZRQ!lSD{dt$gkL?q=g@>m3>V_Y`eJU;`y1|$a`dqMc^!ilSn zrt>jC&+V5-*Y~ag3ZCVeqj);>aL%mde!3s1tv6=TsLQm)Cs6~a%R0TX zLvx&`&zWw;^sh1NXR>a;;$iCa6Jxtt%#NDn1;Dp`gtQ}dmCsx-{8y44-f(e57L5iA zWWifOR8qlzj*!RTSB4R%JZs8cFBo7pIC$HD?uP>W*lBw)8z&gwdLg;)Za%|vGuOv7 zrBY7)t31z#J=wM4Lo$j|8XLItJ`bqr%RO7Z?Lbq)issG78(@Mj?)R!1dYv1P-rIkw z{CyHgzmK1eJrV`e?FQk=98y3gxhV9Re1JjT%6GMwf5Kv)e^NL9-*fmHYyCWPCrsYn z`6Rzl6%oFL{WMP6f=2%oeo0C<$YqwAqa1vX)pgRkCqA9J^bn?Yg>9UC`U~*(Ia&w9 zS3&=~?#>K(MqGj}|HJdWh!3@c&O2_WG@vae)m#nV4HQaB!>XE9K>Lyzv9GWZ>;EzR zwfx?kK)9!Gf*xb@G`#Lh>A9nt&}X_K1?~EYVw{YXz0!Sv;HJwuXL}l^B9oe9&tg1` z4U;3?lY8lUB+Qxy+AOA+0PRQQ#VhaW{sY{qw5g&eZ9??>+AQzKY4&lUtTiJ$Gx|+o z+Jk$e@z?E`oN_-fvP*1$Uf-@_n2CdmC&aEzt$!dr>hZcxn9^ax^w3(BHRTV6Zv)X_%zcKhf3mrJO#31?Y#?7-iBW`6eouLUaOP;R%wxd3&rX#MUkY<`Gi8Igi-?_=$b)NgXm z9ESy@swbLq=C5AC5#sk?#kx#PpO>9BmWadZyE{us73;gx0RQpPnKepLF#dje2v7-vN9SwY8E__n*`YP4G=gzye58D{l6X2!W zR`CkF0AjP#>Bc^WIzLor>$Ak+;07ZfNG4A2FJC3Rrv#y%8GQK6r}6LFhm*?M@I&HwwitqWa*hyr?8JUM;S7 zd{hA?onti ze_o;cSMS}mtLPGj=S`MkUor_iM_^3OM!xBz*R%$=o!4yaON zLz@b28=7U&ex&P5*8|ZkAvjL;>0f(l<5l)1g9S))NbzkSrRR}C58m;y`7#3zCjIv= z-LH;{<=0@zJkasSn%h9(sWo`|gMoJ!<%Is~VD6uX_VV^xnIFt(a`iWzj%F-RX;dnl zmXUq|X5gr&BI1MWcfda@wHq>aUJdFRp+KM7Q5{HHkH+IHD-B9pTVGF}^?x}1#^!#6-Xs<0wBI{!Nuh!|Br>il3wh;-Q z{T4L{mS1=IfA5`#KFJ?dL31C0bVOxmLGyZ;FRFa0zi$H&h>h%)dJ1TUKRit8gbOsW z-q_+g;q-TZPw+MC5QWJ^el8_F2HrOMd#nzaHE!f?z`pZ#v_L9}h^k8?*HpbzL*Ew7 zaUOrN2!ebu#+;%68)+MY8@Lf?bY=e$4*m520vei)6{ zyP&l330g|I?|tUm9yFi0yIKF?H0rHK!<0K0;E!hgg(PbReo+|ts{B?`3)?RVM1h)E zL58|4Xfey{+faq!JrQ4iP7r>xjGDNcwTB=a`Z-C{!%j?<}?+y8CW_3l~a6V>|Sn@}~KLBA2@P>orVYrCu8?+PMA!OrPJp zc9i@ZMvI?D`if$H&n_7WE?k`!Lx1sRTgN++_v2W_PwJUm!|Vde!Pk#QA0xo;2!AL} zy$s~7K@>$xF6euPZ+~^4j!OZgNsWT{lt1X_fhX)gTd?oy*$z^Kq6tVHe(V#1`J>U~ z6D?6bMKK_KJ<#C(lnS&yG0W$p=NS66U~P3aFW3(o?~z8IX5l?RWZsi=AgdGb8Q$Jo zE{oBBU#xP9$#+|Vp4j7Qqx<)-BzFvQQXSZR>besSk=m$NIFDj^dZ}z5_zY|4b`M$_ zV^3jnJdPUD^t;E-p9abqmu>c;_K0NGd0=+`4IGmeUh0JkC(OKf_wAvZ7mRE<7bo@V zJL(yCZx@ND=Mg|G$J3x=?0qn4>^pLo# zJ^qTDRWgtq&D!|MjxZq?OtacgLv>P%mF858XNDRVzr=Uk7n^sTOAB9pmw;d{T@iKp zHvAA!HA!1LgnmEYTPvn@4pm*e8>1y)fSQ6j&7+O9fF>LM{@3UAFn#=ac>YOV7!Q{% z9PAW?X143OE=gCR_cK>p->fu3C*XXhguGpy1+8plITS07{T8RVB>a(lI{DHV-Y z)SMOcru$XV>?O9|q@-1-gZAL$yNV{D-J(8Hd+Z9svZDLWdnQ5Sqa#6^3SXd!mLiMQ z0%u`fMr2P|aWE{IKI7yMW#Hk2&YPYa=XB|L%&1OK@^#eVW*CoT^*&zV2cw5qDfF%h z`?qxfc;)S-6zFK5hih5k1eFeVa9YzB9lTA8WkKezHYn( z-s+z(5A)n)80UD$1+@oz9>KupkQA?v#!$s9yA=*gMu6*A&0j^l#tzB3it7+*+3=oek4S{8bOShzCsL=Bl&{LVN;SaE(j*;ljK>tl9tj^y)z)27)B?KJAu;w)HL(9raO-uvMg zjzZ2-lXJw4@E9v=>Ux-K#PKhDAADN#A)E!u!*?B*OV77OA#lw@ zXr2xJIOSS#sOg~jUyH%h3Is%svT(omzLG(1THdU`GdFDwq|9dOvZ*3=zI?e(GAQ z!9PXo$6-MkU2IzjLOo-j1vyH~QOyZG-}KHp)WqI&nC6e+$&MV{9d9d~1M^)T`TjNU zA-yD4T(E_1=i`{W<&@eDM4_Dd#>2py+n~?-)#>46`4!j6+&zxa==TgfsL-zScRN(M zk7~5CHFl&}K=}u2l{@(}m03W3GxN%n!+?E4E7!& z-DHd}IeTlu^FS{=HvgfUaKW$X`UUFuG~8D|5(-1bL5{ZDF#CXr^JIG*gy9;}&KT8% zGo?Xqt+Mi=3(7d=3U5(f@&fpunlYHNw1bxB+=G{TFnl8XF-l8Ld>G8hIJAw_v;yJq zZ9%7t*m|b2m~+WB6~LIS?<`wZ5o!tj^;VeE05$K_+`Cp{5V}qGhfccgK|M($1q)YO zVd$rk=;KGr&~nK7Rji{d^pZCQ){Ys&qU>& zDfBWcjPHS0ClDVlxQi}q0Sdue`Z=B<4u}TDeSh(}V*b%lq36Rr`oGa&?&{wbVf1!U zTf$66RSxFo9d9cLKcE~0#PDJML#u6I=Jmut(iK-k+|}lOXUje`IJaR<;sp*EeNy;& zo-z+Z3j~9C=3JP}jcME~O0Oe_0olU=Pr06;5sih)E9tjU-4iM`so*sfk*rTL&mV=1 zdbf7z1Q8&-P#j*z%W!U?t{AOv8l3B3BAMIzW85J$Z@6@aZ#xH&>Y1di%N_sD-?rU? zWAwNtM6MA_$iVoO8QQCC%h&m+hvCvtoEx^^x5xa=L#f)2Zawd3kZUq&=6a9C%6y75 z?U}VSM-A+|(_%TaFYBU~7tI^n-(Y+v{j=PyZNUtA=+MY(Hv1#}BQ$Ck-!HqJj=HLL zN{n9(gvmrEs+u2$*Gf`LI_7wb!4HF`GdyZEA7J;*$^Q5*+ob{n*6YorWd9LAVasZ` z)FTi5VpffVsvFS2C00M>RxY4Xz1Tc?kK5QHUPH%G=oebctI6Qz$~Dzd<&R~~?}EkR0n)*hzgYM;jZzwlp{h? zynF^0BwFny=COSGvXpPyYd>Q=8G{dMdiV8X@r8-{%okbZs|Xo=``QpcDW60Gi=AF8^Yk6Plgg zaq#5zPKI$su=$B><<*q>aGFQ%l-Iv7il1Ay%ke9{}jmS zTZqFdhCtANp!3TB^H&TN+wGOZ{6JEACDH#qTw zAEG$rj$Rf2drsG+SuG3FU=A-&_%$O5#Eoku8{br+exD;-_s=c^Uaabv6>B|EU!L!} z{LLO_jO`wuQkp}fJHD7%-=Wt(=d!RC?2sP9;z?bOwVGpWr(x#(_8zHIF3`{+W|U-y z@pBHks(Q=>|7&~@=C}Tg9Ock}z7)yA7C{C-I>Y_6!)%?*wsd_T|Q8Gpv$#Kd3(YCLq5*YeOjIA_hEE%w!NKF9Z;R?1LT=8|IEnNWV6aNQ%K)c_jR*v z0nE7Lw+QjZ0%`51TMj1}d3R=?tSjpmas-mOLA2q{dZ4XVPFy{A3YvI+pCA}0!5|aI zItY6MlpmLRgSq%n|69MuYsapmHkIG|)rBy9sb}`CxE(kFC9l!Du0kGWbBLZjbMIhA z?1jC`bv|sK)Xfc6+hTYjbNVtOHOedUKf}aN)u2d{3hE-SNl!c51EjLW0z)MX=V#W2 z&?g|jB2Objdz4KL9VePDVDotT!XuRzcc7=4`h)uCBIHdg#X50F1GVM@QTNSfD1WSD zB74Ib`W9UNJm*mXVrejq&kV!=C^4K(+q1-vngdc-P0u<4k^KnQwj@oIo%3+)y6g~2 z<66tiTeuaH&yasg-{eO{qw0nJEa_eM z0~U_vZt4!8fR47&0Z9%Vj4}nRI{Ef6nq9LHExRKh^=%g%PGq9{A<>AZ^PTrKcObXv zghoQK3LT|5H4J8%ZbA5VDe>Zl)B+E8a$3DHLJ3KR2_axaSEHK zh(%E|@%|hj9T~g4QB4h%^lcr#)94K(1&Y^Z9}yU9cePHNevZktY1`>RV>uvynUp!G zvKrd;b+tBBQm}Crnzg$&jYouj=LqC%2IRLcBey^dXyq-vT*^-o(YL?#i}EKJe1Exf z&YOg~J6Rf!%l`uh_}M||&rxKU&tJ0RC6@1*V7BC^!4b<4G$H!!%=tV9+?Ey3u^w$n z4>g-IRKfn~#?qGyXj}SI>NBii}NnWG@z*JxoLduWi&dr;fUB}4+uA-4sZQ?dU;($QLi$l95ab(NVwTVynV2>Cdc{n~2+B$b^{j*^;_PoAE(>gGZ#B;53Zya+z zGJ?o7>n*2|e0BWhm3_iUZNKoq`9KZkUF12f5O)g3zuhi}&92aabshs&oq@Ub9?DLu<4vxJKHwNC0;)cSk+wVnBngUJO zuYKbs3Kqy4&O198fNxfDX=KVbR` zjy)s$*J-?7jDx|(X&zy0{w8RR8Im6Fl|uSXt<4<u0B705(W!^GAx+7X_fif*hFXFD1@8ckbyi<~s6y7A^Zc%R5`K9kjKH$3i-h{T|_=;sX7{Ur0tdqvpYe&8w*}D{b zf4x*l+xiMy_q-Qbr}_&AvxbkzYMMZY&gjMGA1#p?FsxAED1#Jc-pJNxOQ3p3zE0Lx zwr)JcuE~6NeE>;7w!h9e_WA82Ipi%u>KWHP@7g@kLDRQi@aaNqUKa68PxUu4zg=oo zQX%%`4)V;Im1Xy^)9aMW@C1SDq<#3-qq2E3R+yBMc#H`DY{YA>jpI9s|N4Byy4kEQ z(A6CQIp4EFAKu`=UHp&t-GMCKuuzLJ5>Kt1oG2GzhZXP#G#%IC(9y$*?^cRMYHrx2 z=e}_vJp?Jjxw)rf2>r6IJ(6jUe(Zz3fq++D4;1j!fu++M8##HVTgppg#s5#et-h*X zl=l&+P2VJxuxyqX|Nb2MA^>c6!+x;uy7 zb2kz9F>R50BdBo~Of)@7%b7bx^oz>wd0dJ&fE@4l=Xw2KKVVUjk@2n`n0i!yFJ<~F z_Uv1^O@WV>Byh@W$7kDAF{@cTfWj>>c>4*q8DSIx-8I)hExI6I1+ zHt%(+qnohr=gQ99{i-A`i_vRav=3l^5G5c9YOpcr<*>1G8@7%fO_nI!iXDf|-F?n? z)GBpV1C(CYkT?faulq`y4z zyi9r&Yy0X_49xZ-({fy9*<4YfkwHQ{3`ZmotMx9a&1qrI?$bt6Jt=Q{RS zB^Sj>1whvgqq3&!qu8I)JUgtyfSOVzWy_O9odR`TLU+byA85}yu}Ilv04r|Xjq%@K z_|JJ$_-r8iQr>s0Uzs=6|AQl6fOQsgU-~U2>M{opjC{7>l0?ShPYaiCJB>xWT>ioj z|Lz0QziXD?(8Q)?Yt7GWD}iLQqo*%R{K2d*rB05^ZAd)l>VAFV^Tcs~zo<7?mSL4| zOI0vwKeEE&cBn_HVb*Ha-6Hk1*fy^-NdL@VJA6%+VEEW9R?Ve_?F%+Vnx<-l$44Jz zuF_vnAjQT}B3Y|z|G^n|pk7>RbGM6=SCSz%H$?ZBCU)O=EVP!V1V_j!#U&5ovFK1C zxYw}l4XAB-FIFKfgOl9yr#>r+;&9jsTI}$D@Tp4eKR<3Qi!{qc_r5EsBgObh%K@eh zHXc}cs`S}GpsxJ9O3UahFfX0Y^phfCVZX2Aiz^yPb#FU(<5>svXJeXpN-i=xg;+3m7PK?58(z=%qr5-?-56HUIQKNL*JmgF9bH!w@ewU$vbB z{8TZ_-uluQCR#ql1uMk^dGQ$6iVvw+Ec+mFo#q@!n->{lSFeY!E$;Xp9NPd)FJ+M! zuW;=6SQ4%uwE+en@sc*5T8EGI=7O5uQEV&fu1K>Z_$7vN@0gYBDTB7ta<%;>Avn9Wwj5O6=!w z;v8tHZ)|HXb>b+a!${X92&tV5Or_4;LZ&Cx_qwHqY?RQ~K-YZ6CDX--hjR zZvE?PiG6s`DCWe4F*7U^&iSgF#lgFTxs76i%aS{=V@t(%<6I&SGHCq{!ygiOOW(yu z-fzt(Aye_Ske&%QHZR&Rsr7}hBgv1p$mjPE{>vXTuKQV2>yW(OYj%>4569+5+^Q=# z;8M9jRN(7DW1U#<4tObg0}4<@jdM`NeJdTIG^MQD9F z_`>w;5*&S_g6)6pM0(d$+gEOR$ToSoUQ>Dqna7iNAO1dqeInZYht6z<`Uh!`U|~Lv zjh~<094rdMtrr%U9Vmu=xdRG%Tx+2B`ONr{76J!2aC`O6_xtA}<+GCNr^rGi>lc=G zMrOmryYp-JeL928ckRzSBZqP9ctDbZ#TfQDA0KPc*@iuSM!G*G7s9BJyY)-G14z4l zzH#Al3XUD-+Rc2QjRR|b|9KiV58EPzb4{;hkob+cRMuNwf%mHy6>+x``GRq&{Sq&S z7bE?4oTg`(6Lwr#UMZQaiceH?sYC=+0O0Lc?(-mwV-*dyV493QlC+z0t<+t8{#;?O=w6T9xT z31Tusu*EflB(TCcW^F5hccQLxTeG2+igf?Vi6{GWkPBvYPZCSIqzW{L9}M zQId)L>$qK#@)?^=IC8RresMjg4&wNc{g?E*iTX*JxY>oHI|$zEffqx?rc!`pQK|CN z+a;i`kJa21zX_NET07455xjQf!WFu&v)z#qUcdC6zb}knrpXeic~G%@$9FHCAMo|? zPnmVSt4Z9x#fM!3cOV&TA3WcX1Qd}?ePJ88p_d}~N&5XS42n9sL3TO_ifw}8^mm)X z#QVxld+$i3^@zL-*xm-YVi%|1AB})tZGX<~GQ5OM)Lyl1k2e9`DCPaxSF$jx^F8ru z)GHj!_wpW1KY=4p*EhL;iAKtuc|0EH?joK1$?@c;HY_;vO6~aGmq^#A!~}-*0WI%t zroZ+u4h-yBc(I`!sUhTw)+OoKsix_itVQT3vu4(|jvf07?cWd82YQIW;KwJ%JHBX; zc$ZqadM!`KVciuU-sV@}aG8I+#?f#b_cj+kGcthPh6@&FefWf`#- z>+8T@zb|(2>vs?T?iW+TuIg$?Tf0%CTvi#Iq$&CUOyd!TL3 z+fdsF1n#CQ;h@&2Z#R~ERBfMV?Ll%~`dNuDzBsMyO&RrDkCd_Dmwk4L5j5=h_I6jEt!Tf>Ix^m*1z*RLQq<98*;3V@a2v&9%K$3dXk{{{9Roz2wh?ugp+a< zNnDnyFIW8f47Jak2d>$bV1`~t_J)K~Y_MOKqPBAa8YaR^ix#tRIH33S8_E_OI+4SF z;`VE3dFLfG+q3|3L<|NKE_jl-QVcR1e^~>A`#ZCVS^_1}2l!hK`oVyfq}1UtC+w5` z!V_Hc8Tt-8S`8H013lBYL)qOBKJPoW?P$qw?A*{5A$egnj=GA{mzg*rjZacB>X0ao zI`O)eE}%nK$LqN=XM%9j?^uKC)&IaVj_@bnjeCxr!isgL6$yKOWO8--N2NGOYe?O` zw)z!}+Puu+Dp-eOikqLji+&2zFR!_DPM08qzH67&;#{mb-}uN(MHXvBB3LqI(MUaO zr}1Kht$T{}&KpA^sbTE*efKPkWK=z||Ni3BF}G@Q?C8Sn*F$TO+Uk2>Q12NqM84gM zmO2GlFK*3uHYd)%(Myfn(@HqZYk12b9{H3Hfc0q077O zyA5p@(%yP>A60#g4PnATCpEWWd+qo6qZ{`@dEIiwF7011S{=>X^T!)#zR8Oh>JV|u zWXU(Ts(P)k(|YN;U+2=1zT}W{&Ixwj6SMwQ4sZHB3Y35&^b4nXk=bD-(`ZfLB`CKB zy-v%pa7^2#@ZbPZmsQhI=Q#YZ>|gx=?dlTJ29y@eYAUQ6liQE9;FS~)evb9Q8TT+W z88m*(iv^K~K2u}2K%>s@_a%k=*ss=LMWgj%gZH~1XXWPL;G3;UMvGPu_xU!j>by-6 zHV-X7Si3zj^bOqA(60YO|2`R)T5n#~p7?2|ZPc*scP%9i!Mj?@USG z!4Ui-E3NeXtGpk|eADBo3;v(JVn?93?YAbPE`d%q9uLqZ@RIM-ZacMIBlL%u!Lx*so{$Syh$`!4em73V!x_7srGf`JRr1tLp0tW(b*faD(?~D!^ znOR|Ze`Nli@9S8s=VE@)`|Rc|wQV1;X5R5R8&n%0*W2|AukKf59DP&%^$WD zW(~9s$wg<%_hH|`_1Z^{31Ig8c+oJ%Ft&bA7F!*`gY`QuKkuVl1e(hZrJ=RPP-S-h zQ$+SBXuCk`B|EHw8SUB>2Sr>`f(gK|c4vRl?)GZmLeSZ{LrrGlR_R@f|fVZ7%Cw)@)Ie!l#T#HaP{ z=lOf~*xs{p>f?5T*LT3#G%e*Z9}dU{%1RmW6UL1;@E5=lx}|jZHGB| z4ciXokaZ45K#lZRLcX{SmWX7wYOeD}%KV>>CmvtIF@t{l=to2zx>H3{#M|T^7R)_g zSHI_AZ`fo1`E8KFb+~=eOQcR8s@`xnnkh zB1HXcpP9j@y#{Q&75v^kKWpJ4X`H>NDdccEShD&Wtz+282!wRZB`Coza42Z};2pG5bM`M|t43Wp|R078*<$ znvZ|yfudQ>X9ACUSz1H)s00}X%y(Zh_vXk`6YVGu(^z}9a(tdBeBLkZGBIz;lb3uJ zenH?{s?N`UB5QCQM|+BFBi-p(R3*&SS z1AG2rk2^ViPe2Ybg*MI|oJYj(D$dW#2s>|!%tA44(S2 z^Wo6eXIU+l z`H;k1R}jd{fv1I;`4-0;EsdesH6!TfM;4BD-3vNd?g&#$)l1Wdov|S$PkQ_c5j`2CMO<2Qea$C! ze-G7G4nAETgo7rZ?deqZJcjz*5zQmD9WW#j>K}G?E_Thkwn^jlX&i6oTRK)EOX7*n z9_FnUNBVxx(#P?o@NwVV122vpLu&uBJI%pUc(blvV6%xVjAa`vuiQ)EIH=(bNp`V+ zaD0KyI$gaNNM8{3(@j77pX+<@Rc+2Ytr;LoF5J2190zX)^rc0VEeoMw#PrOsh25*L zcj@No*N+K)-7XP>xlffJ;p9QG$i6Of>=h_^<5psel+r|xu%K4#RdM_1)-@MtCBnf8 zO-e|sd=a>>hYzP#?Cn0N6pnREW;d7{aQHRs_D8z&<91l^;-ErL^_rw4?AdCWzvl7> z#JPG$Zn_y@v=BKuSjmZfeY3~Bx@&xQkgi)7<7h?b7m-&T<8Nw>gfB;GOJ#ptN9KZ8 zyvslJU}u+?-IxRiUI+UxO>So?y}`0Selq2@mN=ex)%f#M52SXqTT3^w?P+NI_U_nX z(+fBdLEfG)yc(O-e5;Gk%z-LudR^sWw*8L1+PgMMzkG`QvYSO7kT`UCv3(!Y|M`j1 zbncwnB~f%j->ZLaA>A#7(50pp;$jhTWt`yqO*$s6g~2(^$#ZqNVYc|1t^8Fwj$hum z<~~T_UNEtxg^H zZ@Nq)`M14sdEzV%T@r2BF!>5t_a0t}vbsi^1A^OqiVC3U!$zI3Y%fR_7>*(_Z$GNs)dSd{H}_s|1fjQ=rZy{^V%`KaZ#S-W z$JudvX#aVzEZZ*~Mik8#`!fkVF5^C*|KRHv*vq4vKFIt4{mP@I_OG?EFpI}G?$ZVA z9pbk)-hb@x`lR&Z`j^=8)oz1>bv@G89~aC_CiuWc68Pe}NPZHbcg!5>+e~lXI6y^)+}$ni*Oy>v(bz{;F;`=+*t!AFvY*()kjp5H zUjUSgnbw&p!1o$I@->t!MfsGe@HES^ByOg&};wEIC-@c9cgI%u=Oq)37?I#5PKk5qQ z*83Te1h=aAhtJ%__KG-lw0ehN)EdyX7V zoJkz6D7lBUlogIzEMufzzi{r0iY}55|NPP)dmhL3bu3DbcfjE<-|t0@#X`-INad~Ey)%nO-2ob{jm-M3VY93@Jj z_;=)vnzOs(`8CfP; zYv`gcaKz+fd*AjaK;6|4k{088DyAnf$9377X(#|VG4>|yonCpSrK+quKyyRX}S{XL1Nf9vOC)h}^&jlvxt!xm^d zad&L($?rI9yeIAciCO6Va<9YreFhF$2qYN{Jb>Dw{I1_JI@qVdI{I^flV^;6o>HLE z;*Eug=QJM6&4cOM_}a zjv8uD=}z3nVf}*sKWoond(f4mfl0zRw9YVJZDkb{r?K{&%ZP+WDfNosch5rO1O89n zg1^D^hC`3rzGg!CZQ9_6qchm>>(P&p;IGhLpg1=-;|`8oSiEqrLoJReZ<%7PX99KT zi9?DNk>8s+Hvgjef)J$bJ0XH1Y`g;wRr>qB-OjEPft;qm3(M^|bXz#ab97hYfl)yM zmo#ke5PnB_0`eV?pGZ2(_J?6$)v3MKPY+-_u!e(g-azVz!qRO!N=V$p4JieI@;G4d zxWe=Q@JTn%y>;SoeLBw?U+)-gUZ@>?@veXO5R*I~bw8~$Bk|HJ7yjYe zkL0|}vr#Eh*!R=P{pL+?tlD+P?MwI)j&ZJjBv@{xzZ(Wq*L-aB;fVkJcdyv)dD>)M z`z!W)O3ov@FeM=z%5M}T6;1!2e6+Fc7q-6x<2=h3N9k4|-Dh@m;aPJa!2M*U;iy@OrdHrCHcyU zL}I-Jbgw*Q8Gvon?*80qPwY)T61GEsJ$4msKOcRS(C^4}&-^hy{Smt^a8U}zN|15t zgv`iBU99{u)LtXE1qai9rV8ovA>H`HQhTpm*hkHkn(!g;RW#qO-EK}k$Sl(69V;^< z?1Byt^R9Q($ALHNL%vGXG|F8R-QFiXszyld< zoYR(cxR$72qzW=GR-tFlon>pGWHazQKw1s`nK&LwY(cK|? zp~*;V6H8?mR_i$R*m-Tj_E34<`5#{Z)lurnwaY}_e{9Fbd8*R{4`i*R?%cO0hyw=}!cb=M>Z~72 z85nQIom(}%v3`rnIM2l%9Ne5+ww&aNWbYlCv9M*!90q%g1x$Z3a6o0t37Em}I=>#ZJwbqUHNrA!B5LN%Zog z#JWqeJfnXi6gmrTEj!un0E``5WKRXJ<+zXf2C**_HZjo2!=qQa)D#CsX>~VJ<&mBm z_SB)9@arZ^?aNc13&E&|bnE@V2Dm2Q@FPb|8kh@uY`wI%Aw6;#eNhw}{{w@I_g;i; zZ2ejw|6b0pym}VmU21A4S9jomsJMdvwrZ$$-Sat1RpFog)VD8ZFrEJre(0H(C}Ob* zM`eSuI_Iv$F7A(CnbZ9wt~Q~?rxo7e^e2zi0H1SRQY5 zMQ^VtiSNvz^*srySa$UEmx&GR_!CJWeqrUUhX--!w3FQ)nkfb(7D0uyzk9TNLzOJjv0?Jlq;^=pnE(72bDu5zm8qUl6nVM zp^a>vE1(P9-95j*4o7N}uUefY@T$`T+a7xgnqg&<(Y$-_{!bof#k_+{Bd}}rMXm*9 z3vq1O)X9gxZE*Cs^2tok1n9YQ#KVHm4jZmV#%z*5h5gsAn<}i@3l1%#Ql35Ln00VU zgfErNUjwv@L2921`$)Xs7zo<|S72ER? zq2UJpiAj1^_%t4VmUQfH9k+yJnR}~vSMSBqxEDNkb|+#@^aV#&r6$zgBVE|>=AZT0 z#D4QCrcLgW56%Rx`NZ?s^Y1>!TElbJ8OJv%8k!HCL`wUJtLGh}u90b*^83=8w@?yd zaeqNSJ1&aUZw=($geO>L7MIx?o_8RCV_#}Yx{Gqk|m;(+;)IQXC zG|kZ;<0P$+YEi;o{_$FVq2g8!J}_W5*Zf*W*?r!H9xo6N_pd7k#hY}9|dYKXbD;hJpMwzG0N1^;rv`l8EEHt=fUR!Eo1Fa6< zN%=JsNU2~P*El4ORq3}bzcTs+)u!veH!!@h?ZFR$jlFB3zpQT2oe6?Zwt48nqZ_(A zv42tb`H-(lFf`qlHL}D9eki6NiDy|t)taOFD^7gC`b|d)n@2t&WozhMv)UWT_+_-( z%7|Tu4XhyA=)o7J*oy07$tO@D`Wb#+IR*dN&IKGv2_KZFBS ziry}g^4J*=Tqlr5*g45k8=7C*6ZNjyw^Fk|p79~^8JYd&UpWU$C|5%nOW1xpkVWy` zaQasqd*wyv6AH#IPt|6rC~^F(q&yLn&CVZ@=7ulj*Qheb3faO}Q%(1QX~#D;m&+Vl z8Y@rBq!D>+R+8m6+5B$o8NMnfA>j`#4ZPi^$p*?U4WUK}Rp_f#*tIH8 z82UpJAGhVG;ou7Q{i}9WLrS$WnH4AwtX$KmopHv{|7U3E!9EorE96{$AX9{Sd@F5k zeIj^d>Ncmno$@F6{Cl+WHXRp`#8J)n^Q0w;F=x?&IX1ui;rBWI<;pj|VV%|szw@W{ zvFF>Hu=6#V82-{CTy*;*sM8cQU#4FUWCQ-dh`ly-J7}o+MTJ6ZB)>f_Rr|{pyVu!vtaLYr4-10Ls4-nIpe28HZlDZFDC23-%HNkT z`*EOD+t5l9SJ3{53jw9rPYE}B%w>vIlM4lRFw!wPGc(ZaXcT73+)#XWYXed>vssC1 zcVNu<{LN3w_hBM?mbC4*4fJKp=6qRbgu{XDMRnH~ljaooS%0HOL)~apje*)I3|+a} z^IgvX8|6PP%HDbqlK7WJSgdU$ahFOiP}{FW;(O2%8@J>aFw6>^WOJ5~cxXp%ekt?D ztkeK`1BrLQSmGF3uH6CD$Ow^?MSfVlwDX1SrW@G2H)k?t#YUWH$$s?Jq7F;M@wmmM z6VO2!^rW6p;@GD~tq+9cJ8|el0Ns9Uk(3g!tZMe<5$j!Bl9-RlT5rp(H!>e&DKJHmf`ewcbP zU>8=rSfkc1a~)Y%Y1)$VudwiH!XL*wZctk-X)CR!4z&X`Ps1$s{3Hpjt;v;aqQakD zs%OVO5cU<_rYwGp5df7N9Qq}!hoL@Vzl+5(4X9h;q;iA520N@2+^P~@!k^gzy{(K^ z?D@4~<(u=$Fkx(YGy1PRJNQ*F$Blm%iR<3B7QJ-@KVZ?xo}#YGjleXhbo&0-2L|wV zn9rd*&=@YRt@Y9!N_W#vXM87cDP49k;v4SWf+3+@YcCf(gBeG@O%?Y);}<7g-k*Q% zYuf1M_rtQ&&{O*3y;Mpkbgz1`*~{@Tme|aKf%th~^=(VkNOq`%JF# z*ej&9+Ivl=H$%UPLa<9{4l<VLxxwDIq9!pVLAL;-Dlai`YYt4`P&#$}L0b}_~ z#Wwk_!5*D()++#=RqcA?Z_|$lo!4Ll_(;HK4zXa+ojx4#zEr&yOzse5K zT+x&NVN{sgTqIK$7M^_f8@tR;%S`%Rfd<#ZuSyI*!RV&kUw3|#V4d<8CG}5{FkHJP zVj_~@JssA1@z`^*5vG2$xAK-a2g%Npe{%1eVBebAq02)~K+E|0jx=}%=-*PeS^Yit zIy|4QiM;xlG(Y_4gM`Mv``;&)UFNeoWS+0j+Tz*>!%G}>PRgp`aKWm!XD$S9ELnP; zt|NCavSQ!(zxm_@z5G9K_p3)jXLwk&LhVNE)%_rA?UF_869y^M-F1ZhUR~MoY@NwY zWHk8sZs6YqSqUEJW0QK3@>$bo_S-nJBFht|lahemzfN=FIlE2_+Nv(p`=%%0#O>f? zC-z!mc1`Ws1CP0(yy8Q&c7ZICBd0tb%PJy$k5Q^(Upu57bUJEG|3%`ut+!fkzC4a7 z+l(9KU4?4*w1JoWPa&tN)-``hjKt$%Fj`SD4~L!p9J?MG0K@YV_Srupa5>D==PKrA znPH8vfyT2gcVNoZZ?t}rgOeXen?+i;|MjnhEq6;dBK2{jok6=SPA8sPWpZx{TRh@6 zElYb0J+h)6+{61JD_qYgKCl((R|1r5L~kKert94*9xWW2u>RO_{UnBMSk(CD`(_x} z8udo(TO>Rl)?DwjjPU;%tM>KeCVoMN(bIc_aqYystU9yn!l6=R&~>%AFKouXya?|3 zSC7Hi>IB={-@oHP|9b0*diFYos2knpHP7;(LY9(kGjt4l)5O=aPHSUJ)W$ib2c}?{ z@6}nVnHm-bkMWLfT!pm-rH}ni-NDj%%kL?esA0pF*Xa*_I6&XzuEUwUp;$02dPZY4 zq1#9<4gIWjim*eeb;S=#WtYOU@Qp5G4k|dfy(M#P!3JbK6mnc7^Alr#1Z#CCbK#_4 za+pWcLXLf&ZqVmt;b{faVy{SkO0S^9YhM?Gk`CkkzqsjLozV3-URmP66n3p_ZfMfk ziL}ix7wF{>dW9L{qd~UbL|r4R^0b1@O%G@Y_#RxaPXQ={Nf&HDtJ~a`vAk8q^8z#<_PvymN#=Y}z74lAPn}$+F$>ncD0+9BaKTx4m1*tpC zpF2tHhW@n4e#%HV4!>>?31NjGnZHJ3a@kE1Pt%Ma<(3TAe&TsiqaFoCOYYtOu#Q9G@ef5)p!(ePPkJN8CvBs~jWe2^%7nd)~vq)9+yV#IJ)l zuk8J6-x`}Oe{~bnnX0-O*Nh=sXOZsMm;{V(F?;JSql$f$(*t#R_uzBJO+$@vF6@_M z%oShK3$(KvzI{(#h&=-Jj;C_wBkiiz=|RJTkQ_E;rS>0n;4HHz_3gs@VdnD}?V$Z@ zkn)IJ~>y&PEuHLn*lrD z1Qlz~K!@GA!bmp<(uJwol(!143H3qMc&z^>+q zAdT7qBo7|UDG5CR8La5X!Q4e8?!ZjRS$+Rn&Vhr4!Fq|AkD7}JKdYm* zm8rzxH*$BEZbK@LxM)Y-cWmJJzGJb|4Ac_!gLPFJZ62=1FmP~hD1+M;2j8@x zH&CmG_QqJJIa3}uZr3ieb=DN9s@t3cpXnjRaKnePF19`#w#TIHKO*f0<);E_OY#h{ zec-jzdcBL-D3fISFn=XXTlz8O#g;>gYqC>V%M&1! zS5{rc#$7{B)*5Qqqvxq16aO3NhV!>_B?UlxTJ(acT{h4#xcyTERRA;Byz;18NAMVx zZZ_Qb^u!oYhu)kI^UA^Yp)LF|UXP&PBlN27sXF-8+tC&(wHaQY5z*jaN);f)V>{uO#RH7d$=XJC~Zo z&4=O2n+g!&jAK3Z=2v>gpn`c%ZAg^hP$rLU->TFMZC`>F{K#{m)b%@3wb?qAK$FP1VXm+SnX>jLN%PtM3AVZ4*Rr^fK&+A#Oe-5alwDO}=_Fj|E}HW^#4 zHe5ow&cu7A4gE0sA?&AJpd?gAEU}@^OM>cZ>-=Z?nAje&-OHe7E!6$E@@XSe7(dBd zb@*_z<6n>*l|L4Apc&)i9t269CqdPVS2s@&v>-(?{Wy{_&t9cq9yCuHIF<%J z!zN*)N0GrxV6e&Lnmx4+hMi9}OK}tT`Df3Kuc|?WE*s+u*Mg+amw}OM_bOzVeGXzZ z_vV7S!(-4hSS0(?_&xMK__ux13A&J229A&Wv5G z0Mhnv3a0Q4!g$A%khv#iunKP|@RLQbXpUWj&NE`Z&@?C4CZFZRpAV2Kb)GyEH9Y>(>&9+94Y8J%ix29URq}h&jlIwITA-YgK+jrUU?v=v;kieT<$0M3!dkyna#{$Vx>`0dxl__tgzD zPZje_RNvM_e`yZx)BnpFC0KYY96e zQp`oSn%5KirX#JN2R6c}TSUgvs0Uc>d{MQ0oZu_yW9kK{?k>bmu@`M23@#XLZcy2j z=ZMVc6N~W|fp7itG3;nf=?fe?tI1t4y#=yU_%jXuu<>j#9$x=O#Lnk$J6Znf3&pWB zMe=AReFx^+2KWi(uEYu615@9I;-I!P`F5SfUwbCw(zBk!#QFBKae2I({}|GPf<1l| zFrjp2!TYm`1ix94isp0CXLUd^8Qt}a;*R7CbsMbv#IXHm$$GaP|DiLwAl^Z%i?AGDdg z2w3eGE9V#Oz}kxSrYWDZOpoOh|S^Rz(Oyyx1a(ijXHKD%GU8bO-y%fx2+Pgq+YC^2Vi zDEqn`35O*IpIim<@aJFZXMC}%UZo~j^)5cM7k<9?$ti5T>`89;y6gY?9n?yyv-V8zIjA_Mv^|D;1IL#y3^DU-!)(jjlY0xFvhS;R=>gZM>33L}Vsq!vjxwyQ ziFo0$RSkw`PrR-C<%*`IAB8S9S+0OW3QRqrO-h;m?0m1Ey8CD zA1XV`a^;`nz_Mlc`xCdp^fTGYz6ZOY)H&{^xbg{XTXpS~;}Js7kCJn%MC8_0OivY6 zJhD^`3-k|pe4Kj(3;QPL?BhCw=|}F||9DXtd-!q;O_vM&*L{H4XZ9ptYL0zQ;yc7{!UR+N#&CvFkvPenDe-y}1{56}eOdFaLYr z#kE$(cKfm5mL~7Q-Qz$OiyhQayNQFo2VU>Z*^iV!Y2MWz>ae$hu_Eu%EqrEGVeNUj z9LKz!zsOn~#JbNZLBH2}LQ5lGUY-9hs3Kp@O82wI)+-NPPU*dZirivB!DH8OLM!m3 zqOdOZ#-DN3rKS-5czSG4sZbM?C$8pQ%o7Zadi#v$r1l`QoJV<)WB^IXwD$b=B4eZ+ z*^;|PcPm`Gnk}{}#1^SG3#QBYJfNxh(&rVsg0Sb%sYOMe?T~M*$`XBJjD6dGyPcsG z0L$4VQtj4+sDv&I|jjM@GVX4+7B z6D8MUUqI%$yq9dh18KBJhP&67V%c4Zt8t2#fgX2BA?7T>8{7WC?_`ViE^u(l-{vFt z3TQJXTYANGaa3n@zJ_=lRAoN?eR#(GuU%qwZNX(WPa>okgor)fbqzbs9XyvEzl$BA zH>a}1oRFd^)GbEe50%j&8=Vi)k>;_^aDNszmcI{`PVBn`^)&I36K%d&*C8}V^fwio zq7)9!T@+^#_qE;U2fLp5SNd_UwN<byJ2k}i zn_;?A$R*{bDvY1BDZ8Bc9XmE$bDF=BeSVU--k9^BHJ(J)#LHmgM8dv%J4cuGrR@Wx zyxRMKpBVssc-d9A$QEPQimkD~RRW{OAB@Kx%0e37Tb+~F6tQ>ZeVe2wVn}V0!yT=y-n!_rJ(9Dlx3UZ@a<=DA3| zzh;jVyUyMBhtDE;)%xGcmkEBOfg5jP_%?<@ce(=^Z!`nFN~m3K{t*JJK?l@Db(nMe$$S61k-ma$5s`;L7Iv3 zn>`W)E0Zno0HZapZk#M_6186zN*4 z?QLm-xR1|H)fSlHRDjm-;akQyylTSqzUe$DoND}Pu13ZjvQZDHaU3k9=>d@#N*T#xf*zsX#39UVukU8{U=O^~~kNRoKO1mYV!THKe%2mTu=eSbbI6Nm%O2}>!-F_1`DclH*MQ;kck8Ft4y!6 z7+9>=TDej)0dfXEPj=O3K{n6i{5p+ksJH%pNkxAX(ssmENLv~a<9xN_`-5`?j-^bg zcFmy$U!cos)Ka%-12Uy9O2!RH+>)TLsP}(TwU;}h|xl6M8Zz8$pWmJBG8!%$;kNr}vLh?2{ zQ~Ta!Na?M0(^9egYxkX#R%Uo$A9MUcvde31JbrKcfS)~dCikoAX&l13vBJBBb0x8t z$Km+6&|b(Mklypy?IN*mPd}TnqbosGSnOKX@)yLuH?5=gk|v3LdsDMmzmGug_XBe; zo%x7;N}t6FjsN1N)taMI-6N=XC0n=jsRBAuUAfzDpy)-#zFWm0q~?3JxmWxml`C;MlypCmzSt zbJTMYcDCN5Y0VNfq~7!rm~(Lxmb+-3K5i(_{=Vc^^L?+2u7CZZGLeUCA>Hzxo#k zZVGiw4-+_18Z~TXsE_R5b*}8_CUPJ1l&>t1h?x94ulntq)>EO(Z)xj>mKVUp$CrMK zYaqj>Al9^%hVd(ObttQYkoNU(xwii~=x%)Q=l1ha9AfZA6sQsU1T_r#NmmaM9NN4~ z-uC`{{Qc_Ym8Qe#_{Z~>e|vT!j_?2MFW#sRaoZ)Ft51Hzc9Hj!A5&jI-iscIG~r7a z9;&qRc5g8H@CR?Q86JB7Y-0PrKn)W-Df@#TgB{k&vL?;2?fTND zFQ4S0i}&>7?qWe0I7#uM*OGwJw@_f7CH;C(hQ#}A zXKm)Ga_D;^;oo@44{DQV^;Fk{V#kAHyJfdMguMGNFIqoiV%R>X6)9i;j?c%7;cW?Q zzZ;5ATl+_8?uG$sw&7SEf%EFwpuWG(a~R8o61<9z68OV5YOLW;um7kAzM7E`@N5Y( zwEeDlh;_k7Sbp&Qmq6&A8uxd4E1^Tle*-?o9j zL$FGv7wpZl9GRvpDQafZe6`f_AB z%W0ACuE8d8CjwRK+ zE4ZdP>bGFD5O}Wseu%@?&(c3L&SBopd2KycOR;Hp8YOr?fg2ynAKJWfIY&SJkDsfH zrF2qKot^JS^60GP;(R_FT5l7Ve6|kj4m}hLeP)V-qQz@!M|}POO+d20*$(5&x>Ox; z63|V5k19L=vu+#nT+32M(&pn})~cQpT|C&T{A1C57J>gCUMhAl_A;UKKyEIu&Wabu zV#^RwZN3{=hdunivkxLUCu_BlkOELET#ek-*>P>a3hne$$y5LOf5@~F6F$`$51*C^ zsbp)vR{k6ve}}iuc^a~o^$mZQ&XUtZca~AD8TBxx&~fq1=44osynm% zpl#;6$9h9PBrjq-7=QTuuU#SE;Q6NsDdlvR4Z>nDrdTU1E5!b?stDjC- zeYwQZE^T4t=0EeyuwH)U@sq25VDp#Z!(~f@;PNufQ>SawvBzL6^6ifo*s+lDF(qn0 zQpLzSq#7rX&eNfD`QTb;5WJh`a2JrgYaoQjLjqb7JGyDLFOZtyFSsFNKDOEGs1E6e z0wqPyX4O_g$9{s8Kl0rq5ymW6$RA%ig*3_=SH46BQUadrcF>8!I?v`u2sC?btZdLX^JM8t>=mA9{`Wg{4#1k+vqkqPsPh<9pi4Q?IPdBp|Bz zu+>=i;lKN5;i3m>MBOUOpBixFp)9asZ7b4`Z6bab8(`W+kAs{i$1ceRzQLg&tJ2Gi zeMo!3&olDH4vTlwKD`hq#;WP+;tUNpq}*_HxRFQHU3O?(zyB@jB@R^ye_-XmhGE+N zf|r?tSXFy+@HF`m(mhDalu!La^8C5wX$FZHbAaFe8=a`L9h5Z@j`7sTfpS05UFWB< zci$HBGT~+%cCRV*`XYfNBCa;$&Z*GveK=KBSQCD&N!)I&_6yUW{8(yjcNP1ELg_mq z+M&y){nDO|WSC}dlV9cd8#_0OGR(40Bc)h>WAT?E9QUhlQZ|#rQNjK0*W#;Ts5j$S z)}}m|(*0^0Y8{P389!%&%$%Xm!T;QmclUwZBl+aZb75e73Y0mqf(6W_Dh=(02Z7b8 zY+~Vf9a)>_@FxkshDNK_^Z8+`fgi@zZS}5f1b5y+g(Y&QJ_fG zIC-zW2DLTjQ}ue;*u7mf>U)kGRvlP>|E9}csQe)PC_+4Y%gAU=Kc_} zu6~o=n*|UP&#X|6l!f;Fqe+(^&f>Vl1y|wnY3SsMk6d@)29B=2EGTJ3=+X={oa-x* zA#nJ#LBFKl?Nc~Jp6~UHm$2hw*6gv2+Ng?uR#+O0>9in=&sM@pFCF_=TAcC_C+cQL z{Czh}koIBE0kcO{G2%!L@2}*;1K77#@R$3#7HG-6^5(lAr*3t7A<4^$^#H?#FD!rN z?1EoY^)_29n?&YQ`gv;aG%yFYU$r*>g>==UCDF%2vCZIdH?3wl7F^hB|3l;oc6XTH z+n@3h>8x3EIi*7|efeU!`ukZJZWp#ceU}EbZE23WlH0H`W&g63KN6sn`|jC^GpB&! z_kObiRTcX!e==WI=^=Bt?V_dhN@Om4qVLGf4SgT(27Gz**N@v5+~}*%jU$?#uhf(( zu*6I$iLQMIo%ghyj=km$thQz!mu?kke|I{z{6Yxy7gG;P-SNd?((`)}+O5Ep6H^H5 zBJ7o3`EQ90Mp`&^_xrNz0(r#e+uw_wej^Xbr){W?m&36*>#CONHxpu=cpcVN8!>_M z>m4V~96E?JvRiMSZUU0sd>vl(XyV{TMFIM&|KNwavLaPx9-$M@g!PfVg%2P*w`}qA z!Wrn~dw%KVAtK*XCz=>!@q>cHX=klT#(S}Avldk}!KIsyCmd%q0(-dbf=*X#+d0Q` zNUjGvF4aDid!dgs?K!6u@h4RB>oG=(wIEeZt^MFeFC0H`#bt@4I1WtR=lbRJ2wACm zb_YK!gT6VX@&$Wsu=8i7IHj13L-H9)L7T;~o}szaXh|wmRZ{h}>tkW0DR_g@k^rQ1 z(Zf47njq!%#SQTbn9!E8TU9yw4-USK%??y?0smm-rX6#q{`!NBlh?K^hSDW{InNiZ zf{yb?g5?zQF>0$rd-R=;&|uYcK0HqgDK@XZhF(|$4F*)>?Tr<{@_BWJ-(M4EmaH8X7P5 z`metS%Da{dIe`^Gv$)*)CQ}>ARVA=p>?)?-e)ddh=`SD$CkmR){`Ip?eHc~=BkE^* z7AcfCtDZ;N?ZxzKZj~^p(pe}nwg6N4|Agko@!?qF(huVf4kVr%XFj|gUk+7MRl*im z*mx`$JY6v(yMxG6O|eqE6m2H47TwhXXLYcl_S4vQZbE-)qAbDX%g`1m+jQyF+c19k zDBO9_Rg%b84XX=VsC-$5gH1*uA*(1rj^FVk-bN04UmZ9x?)MPWcCAy3+_nQ^?`zPK zuV2K8yDc{lD~#eu)d9J^*+;NVJ=rwrt0VR~KCQ9(XpQ{^!RPp{MB>zf!qFV_cC7kx zUj1#z6m*?9CTrHO4y|*=KBm~;LaJcfArrgjr1>WbFL=fz;zZ7aOShMD)Xl^AqWQm6 ze~Uts)}vFd?qksXPGI46a{&_f36aiJ?_hk(btAgZf(C>7>r)^pf6H+&Uu*~b6c{y**Kj3ut*$c+n#^(bx}aD3AvBv%>fsGg%^aDQ9P&;vU-&)KR$VoDgbHlHB-ep$pr;xSi1Ks$i5EdykSxj7S zf%b)YZy#3lK>wF|3+pB2(7m8({<+J~fgbC5YpQhtbc*(MP43x_qDdFtLA!9Wf2|Zzr}W+x zR0|#U;6P)qUGq6dY%o75zH~1My4F|<>f>Azx25;XJ0Cr<$My8CGaJ(&_?Wkhs1ApH z3S{*aPIZo3fi{yMU^M&%+r1B#32#f{Shtkd2?i%arjY)dZ16c~C-#irk#h`U$3G$F zn!VVI-XciQNK9#oA#krR6@NTjFjfy^Z`wX8_7L&OwnQIuk#Ht)A6~T)jL0{be>2LY zYrO)g+HRwHG%g2lLBCsZ$O@|cu40=2Vtvg%idpd`ZhnQo>KwB-9- zOOAbvQ=RI9Db5>kqIg9c)=42%==ySQ#xd*(UD$T-nhr)~z1wWYeF8h@=Wy>W{fxA$ z`5G2U?7AH69C4bgZ;Hk2C8P%-*FSMQ2YGU(qJ_f@G8e?eh(4ade*J4#eW!;=LWk(? z?L667y?8U9WWz@M@=bc5;(`*azZ+1MpK}3*${R0tTz-$;7;Yl`@FI4-7j7~?5QELK z4+9D{PeH$)o%M{%ChXCAzPYrRg_Kjzy0x8u!a)A>;n?_Pz_|HcI&RYl&=fYCk94x@ z0H8al^ueRptI%pHGF~W8)S--?$~QZ3b`;5XZ|ExsbLv)gU3~n=XD=Hkg0-y&*B<)Q zgQbc07N-ww!4~Iw!Ns>^fc!`ELUG&}Oww*2mXBuR#-MlSl!?`_9a62zyX@EbA+58y z%H$9ON4gBv_lTT>-sTq16@g78Zm~g&g8kJ%w@e$mSaA==7tGRE^3jm2Th@Loas>>( zKQ>-g(hvPB%J1Ok*D%drWBvN~d6@J)q9rd71_PNVPcKtg{y*Kn*!NfbFs8vn^dn5w z$`yb3o(s$@ZBxx_(m=hi?zSmEM;#X14-xbl4HTNQ4|*he;=)&VaeU8pY&i6j<7Hr& z9u`=B<1tJMX@4%>Ysc~b%~P8lnNqu;*RXcsHN*8B{om=e%JeSdB(w)^-9mZt4U^OYmtLIk6FCX}M?E4t%Kl%1Z>VK;@2L1K{RgOMves7pGpK!3-st@BmCzx|iaOf%`RB#N6`F4H#W)R9t=gPpZrz41euTbCt2+cB=OmFmF>;-!alnJllb?W zk-GHA!L=dmJPT=FMfO_9!UAk>klNjIU;sPL%iok$^n|wa^RF)5XOBx=jZL$;xH#rH zkMU?mU-VO~`ktM%Nb&DhD%h1G>B`Orleqjvcb==9$G*QVMaxUO`I#{G?e}kmkuyN` zeDF$!*9D4|4NPJ-KZVNE1rPK05jr`e<8Sud`QnQMg#)WMSH?lxBO}l3o59I}R0$C&5O{|rjSH`db zWk1H-0id0s-&yw8zv|VDbX#Pk0G%sLVv9-pfNV%P3k$-5>bfm1YUgo|=blftH@@P| zq07oK9(6jj3rvNG|L(f~e|-hYqUuVQFE@d9tY>qxbSMmK!?Ag)DBKxK7|;3uAM% z)kvF>Z1Y<@#rB{4zl2@OdpfcI0*78ER4VBmTdYfl3b!!rQt1ewxTN+RB6M2++1{kz z{*CMNIQ%R2{n5`}QP2_R)O+u9oM=~CyAe|7J3smQSPUmjmwz=qx(VH4E3(QgiTL4! z?3{HrKl+g@l~r=KRqmhvGxGYHpZ#zX)VByeLp>*pnqHgLytK1tj zdP5!S81}0)9ptd+Lz+Vz*%ZbeJTxnJsj5!<$5IT zv3v@{uB*SdcYVj6N6V@t{Zg@~xpehzW+#cC#FSUwUW!dkJCy$D%|V8uC-0V;P#nwA zNiq_jgJXd^DSSV|Ve0yQQE$=~7@kYJ88X1ZPY#q7iW{zLi(s$N>&GHHQ-O8XnBr>5 z_Ls0%!BIJJl0%0IhTM3+zt?ZTzRN+j0Y;vfv9`i*;L4Z3`@pyF`W;^A6{HdOvVb4Q zd`dSPK4IhTu%>X^hA!VcWL)09Z{6k+3`#w3_oX-aznu>qCgWxzdK@@wV77&&zvlUC zcifgAar?Xn+C5gjOyp(r9l_x9WD%F_?Hs=wXmE4aaU^g{w9k*4B0}eL&NJI1Pm|CT zXyMlz>oyBQ^96T|OEedi5RynmWyHn}48V;-5Y zo+CdCtsfWf3+KB5Oxe!v=J-<5+{|0z+joY*&pd0N+b#rOC@Z0yUhXP|!%_49qzq2u zSBcfiyv0-4-=ek4AYhavpfh`n&ySy@ecD6U?b>DR{ei@L@#KvJeiGFGYA@emQ43VY zf_PO$j(i8_`E(drEBA#lEZXT;Oa6z)fc z>9Wi}hF5{{zI}!HHhvO!l$}y@s~}8!*afIRV&`N2?sxvXmw!pe4oQFID4#wU*qIad zQHRKDq#0!7%KUD^`uX-bhCKa9SGDbU)awBlwfKjVfzJJ<0Hl0gnpSwV*vO6!_u7F)ePmjG@Oz1h#hDd)b%Q@^&{KQr2x^R9K zws-&VExkL4(YF1;YazI#ZCrtT`r7$ie3TlaD4JOpI~(G5?$&LJyA$jdTA~ zKm32(4*yH($&OEqaUkm_WU7yFeXPmYc%<*Isz)c(iqcdHGG1nwdDRR&F0smI)X4%8UiuCU(o zFjUA>X%+Dwyh52H-;V~d_i=0pYal)CC*lm_Q+(bnBhr}Vd7spjlk?C29d|A&8gi+^ zvCA7amRm2tHp#85SMNG;WHZ0}=2@cNwe;=|pYu9wKaIq*@sh$?4?Aef2-b@kQiln3 zaIKOebl1rX1bk!-t^c?E3={2Amw&_R|N414*BWn&w&%q7P8WW#xsfps>GJ|potw@g zRj&T3O~F6=8z~Hfow|NYv2*pi>S6H&sClWAn?wtN@`mW=yn<}qdgx>9+qOBF!>__A zJstgb18jR0hso2YZL>nLXYR%!U5h`+G(Di2Ir9hmm86!{Eo&yO%NrZmzlkHx#PQsN zv67=HKNkN_`-#Lm-$wV>-hbxj+3ij(crzq;&V^WKE;}t(MR>!opJeFTT^#2ji91Vm zSAj??WE%gv5}c@xqwW6ZN!pXZkn7qS&--_Lvh5x9wf0~SHGB29)ypvJ!VX@yN5L?@ zQaqpW)Rd#&Xn89erRJGo#`2WY4oeAtdhFum-b>&5ky)%2dH4rEGJW26n+Whg)iLJB zhlj5LE70NvrJOx3$nq_i4lUu}iT?leD~|c@=y)8+?k`}$E;UtA#{b>+NIbfG>ZOa5 zuxY-mN4xr2Apc0@G5RnC)U0c=iR-+8+GjB8cX5iu^>`-n*9Q$8{$7`VB;zhrM=Cjz zq;+9p!?~uL%cUT<90ecr@&Z-m>m$FeAQE5kWwY8Pc_hKB_D{St1W0^Gik{qvx&su6 zjd$1N5_Qo7Ma!luWFB(Ni|>z=JVJ%vkhsPTV*;**lXyke2G}*Q=O2vuAFO#rY9(>K zCFeQ!iID{EoIMe$E=c0~b!vE6Z-m6FoGJOZW@ovt6HEeD4EP?WM&K1|;HE`M%X zlf*N#bLd*xP7=4c?QQ;!8!)E(J2g~mKak_?ZChsJpvTT0b%%0DTpriTROkUDo)GFc zv&GIF_u0G3J4ZiG>t$qts?P9!r~MLor9=dtW;UUQdYK9KFW6f|2WvGtwX$KQ2vQ2O!5?uNT5SYSEt!cMn!#JWi^d)9aC zub=lp;{4+uPq5&`+stCMvp`+{JU-vs4`?;n8$NCi2TE_+x`De-p>s@ePgDZ2U$VCB zSij1%j>P*y(B`)4K^V)?-gKLn2_xK8IXEN@1s(iZhW>w`jDJrv|H1#zLGW&DQ5y8Y z>blfVD^0tAK?yiIS{DvKdxhf+nv0+_>S$z3;8$RBA6c3~_#5o!ON%uMNsW|%6!XUh z{&DAEFwXbv?kqFRZZk5O!}{xgD(z&B9tnZ2w&1x!2kcrN&Opfs*AK%$& zJU*}D4m0lQahvR=fLga>-yaHQkfOobuzSrh=*Y6ooKs5!I^$+ia{dCy z(H>uu_AVUS`0wDHSV8FKzTiVUAOzD(vtEsgwF5QobjG;U08G}5llZPbgwf*zGh7KA zd~`5vvgq1jagO~Oat=8+&v|kL=)40weoOr!GyB;xm!$6`e(5+R8(s-u9Q_%5T<$hD zBplP5)c%M+EZq6KJQkC9ve#_gq1^zp6)j=U)1sg$>4oV{y>?_$baeFPm>6_nG(Dnp z1yqjvbb6(|!qTE&%_gi3Fv0bs*uJ(EI@{&c{IkkndfwaH^RI@( zboE`kGb04AGX47F==(+CKr4$|n-`)8T{PENm50SpeS&|%A;ZudO{%i!6AY=0R`Y#(>!xwzx7VTAV!Q;z?^jC$3MPIXsge3bGI-gh6<-}fq9FHeW| z-OVX4_j1G!fwtzz`>lL8u;r4w^Mkzv-(bf14e@jAm7(7BrstL?ZjkW)tp3es?(FX? z2d<|F&07b}PR1(Sw~74PyOqy3#BSw>`WNc6$3^5J+sj?1K=vmz^Oq+izv%+vzJ$+L)cpmt`=xnxU3d_X0Xij2>)vYz7uX((=l;GtdweuG?H{hj|Kj^%h$lgFm0` z$E^;(B-U{wc{~z%fYp5Bt#RjNqQ4wtFQuF-gz;tB;RE%~z)1Fzy1ZZ)Og->5Y1c}C z;Zu`+;xrbpycwEX@)}@NnRfVA1Ovvsthv_QBK+9Oa;3xN=SeWCb@a_97KaW2R6RQq z5^UQ5Ob5*-uHX9D;XPV5oS_ITS)=xnr~l;TBk}pzyFJ9CFnr=yU3g^|&f3b>gz>(G z!ElMA%v3cX>$}Nne&Nt3=J#0WMRmfWv@7)Sn z4MPE2Ul|`@uVYN#^6EQo&=jlK1bKqT9 znzipN=h%1H*UgCO@pj%E`DefKl)f#$zHTkl@xD63I!EYN)!kIFI1)tI*R->Gjd2;? zkhE~V@-UOLp0{PMkl5Cp1~9ly`P$REbHF;W_D#Yq4P?}Q?>{nU9EZaBj~Q2Hu&*Ca zZ<2bZAPUoQb(_P6Q?Ww>zI*$h1FHFhrh~j59P4vhNKSM^_YNq~8?GBq+64)7bwZAW z>;mTG&bay)m;UZAQ4d2r0GQhEW-U((VxyGxz6g`uB%Zm2AL`p@Nn9(Hl;@a_lDG~Z z@I2{m3FCgIVXMAP!OZLvG4qUn>yj|cS0C&0Y$Nf`d-F(Lj;QaUP$G`LTf^q5haRr` z2R^6WCC%ws`?ZFPsP`BYQNeE_NhGdm+pJT=w}Je+hcuW_1>}TQo@eiRalqp6`)wc1 zf!re-dHq%%QclGzKF3RezC~N_1=%M4Z|4nm>fT?O^qKP*(QTaZ)P`!lq-d^0 z=%p_zzU}P@^c0)j`qeI zivH}pvjmwA_g5+O_u%MOgMH+vI-r^QrBAdtW5TJ{xJW~`t}hHT6!HSExj_HVof^Fp z($H6TcgtNXMPTl}US(LZ3aIqU8w01^u!n9*>WO_y5|O-icTOOIhskf-NK9*;B+gHNi`0d*Vufi*lO@=`|@j{*sOP{SW2S-ql*i+mBikG z$&l^yxZhny+D>t5?4@F)S|%z7&CG|2^2;05_VfUAPTz!e2vOI`poyN++r10wuQ*+P zs4W6RJ#NFh(mgQe>pra^g<Q?MEy|zkt)$$X76y~)_{qP{Qui= z)_<KYGEn2Eh`I#2N%qxIcE?E9;mG2hv=Js*3XI=0f1iM-Pg&#wIL4uYSg z^72FG?To7=Uh8eL)spOd7S!G_XcsQ80M@U%nfZ7O%5S&tynLgW{r#AXT8PH^B z4a;WPI2jU;iO_1*t;SIJ>k)lDdH4VIc{>T;OK zo+o}OI0$;Jy49o1J|o4{MDyCy&5%~Nzrm+(J9ZAWQtKPo`+hq2@Uks5HyId`3$>CW zZNrNEyWeM@u;bv+EMv*LZ1W2&7rVWG^JqdkzwCMOkv&>4>!NjVMrs>W)fw(wbW;Gv zteSs+JQoKv^>qf?$!tCloZi55T!8Y5W85jpCqzEpvx3ol?=wkn9ii<@_FUbeNyvXC z{=UXL5BjYOfjLqK#pU}ApT_LPrgHt$jvEZ1*f=Y4;LKP$zcQ}MQ5J(Cor;IUI-h~b zT9+)M@)NuHTb|n8{RNbsj&19&aN-SEZns6U)N5h1PQz}nV;ztgq7&aq`ynA+;L7Ns zXqaVf&d%mtjRRfAmg@x#a6-`4yp-_~#tY`oEq&rj%nRSV;%_@8ptn(B&}85tw8ffc z{h^yehFFu;*qjWc$v2iq@oL~mu~&d}(i!Y``7&_rn+eoyc)ELa%OVor?)JuWB7{z3 zwL#_z-yau&$~U=ThrbhvKWApIDDxiDY6n9f6{iBl{K?o&+5?;*l}SEnss^gf_l+jE z1WCM-XLv)o1)xX&?L60{0cbXh)zMLyi;R0`t|Te`hVJdD+wCrJ#BH#nAS`6P$#tYX zGUdW*c3g(U>rgv;Sd*O}z`l{frTaS&b1#RPs+!!uNyf~%<19;#_UJqZskKj)kmV50 zFVOi7s-lgq8+@AthIH7g#{mRC4B21l&9QS2U|d69*IJt+-%jFr2X2;6He$Q%?*+1P zqEPqcdb7*%Fv#I?-zM^W4zd1{gm#-RAo%tMS1lNh-CT+t#kWU zV*hm<&)a9#;>QF z9NrP^(Up>Vtkpr-OV|0#w))P6*~VLzUoO^T*F2ezvPX&bW)0P49QL{VT@&wV;kP(= zNRa8bDmC>7N1h3$bf=`E4{S^q2$RVV7mvW&mvaiMPxxYg-_Tg6&Vj zS?T#%j7RJ`ZpeLL{h*9hfK1b)-gaFk*mv~xtFUq|oDP&PR5eS63b1u z#<8{El#HtHLT(tVspuBLFVY(&>(h0DZMVSK7J&(;=jV~)zGn5V!i~_G;kNzfl`f=q zuh8T1n;_;BOQ9_$;R^maH&t-u;SFfM5c}#ZO&>-NE!uwM#77+eHm=lpkjTT(65rQy z3oU`J>`YJ96|zW{die5A9XsBHLfmta%wto6S~Hnqdm+wAYDQiOtRCl(QoH~Oe8NbLMe$+ z!{?Fp^y{$K4z?eRxgy`~!}Z8mP}_NVUifYj*ZgBLC$7Y$bHxvY%h+@3vX6K!lyCdA z0O$pK!#7n%rO!F2?Pf(=5A+0=AND+cIG%H&_uSH__v3B8lYNj|(J&kHVOM+Y>2D zcKRHqc~<+yH!vabd8w{O8M3k!We)rx>ZL>1s`&fZocyPKG2K1lqP-~um@j?qohs@l z)~}AZ9r@EesIxz-_*~_`$1^={J!M)nbQ0;ERfD%aO2WWZ;o^BCoOliLHp4<`qX3S5 zjdy|dfR@fAX8qDK(iA=pJsY2+rnxmVzI}n(DH*wg)yr zyxqNx`^rb5wX3dBX6F=%r%~87rdOMzoiRIcXK{ZjbZr-EGEZRJ^*CpeV7f2X2;KSGQ=w#x5wLB zPF-;(BSrJ+WJA$E^UOdsRCDsr&;g%E4|8mPKfzz+F7b}ZPhi5<%<=hpc0Bj*JOrJy zd#5ntY4V(+bv*xSkHmY=XOn07Y3z_EJ8&hv`@0_=^OsAv0%q@qwBu>)@do;;eNl6E z5pmD<*hhZ#%Ey5E)u~F7>u@fzYJ&V=Xw2*?}Ek;g?nTbsT}v;Gi#eLd1N#8 z_^Aj=4-z_Cx$$A*2U8wEt8Q^l?btBTx0ofG?^;3P+W67pslE*ioMSC8cyJj;4C%v$ z9ovv5aJ+7_?e}z^0}US=w9GLxciTQgDPt^1AB|$R_Cbx@na0R4YizFa_3Cqt0V;2h z>Ebg^Fp{}$;Em80j{9O_pSb6w2+_}Lc|RCeh^dJSENo>n&0mBYy4DKb4G zh~xKU5B)D!EagkE`S*vYvEHp*k^wwWw#R{|$r zaePL%Qk&q@3CY{v5Esb><=T9QU7rU+rnCR_>0&eN+#(t+>f{8pL#B6LJ|^@*hVGt_ z`ys(;r;Q0RDSJL6$LF*8OOBsRKMGgxemj(SpYT^YM}GM(`0y&7Yu91z5GJQ?&Zyqr zTlZQ2#0hd-Uvco6uBH4Yj?WnpthHy$InF6ytm@ZtdcmI0FmOj#_Plg9(a%=lD&~f) zbl%^I8+LwHf}Wd|N{aoQI`VaUZdk5$Byd!Pl0I=_+N%FyM=r3?`PvKAdAUx3!IL-= z^xAHp9>H%NaGQt1T9)B-$)|(^FjU^j5?D*11)Rj;D zNEIq1U&CqvZU5EV30=-6|IUZy5D)ZDl|}VWICSfQNjAH6U8DwS@BW;!`7?x!W2u%U zC7koQ%;NFyUy`Sxr-*N@v=hOnRJyD2kD|5*VP9}(W$sQP_LY%!Bb%NC+GF>}C6n}y z8jk<3yuLqqjaxc0%sVol*xDg$_qnmnzJwi|vQ}E<(cRnW{1fZ$hxOUQt5sD!&C9Z( zqmsI6!CnDmygvJ)dQCjg+P~NZF65|}1Tu?d707)WQWi;iwEGaeYaPjsc1ngqSlM9p z`&r&Ln92~F$rlKK0Vn>XBRo4HpLOs1_XUQ;I7C@R@3onQ+|twsYrpo9cr}9gw=t`L zVK6p*_O~I9JyW{)y)zK{3NAgmv`Gj%wpeA4Xhm9M-@y+>*W62sl)D3MxjuSGNr+&!hCo5_xU;#$eomv3ZqLFj>lQsZjBaBR2hkvI+F zTz%4e+X9Y#V81V#KFSp+)P}5|^gW7s#5|0j5xzjXJquaWj*+iiiF0nMY~ibcP{OZx z(kt?T{+d5sV5h^=Z}qV_Q@dJZtk)D{15AUD?IZYxvJ_wEtrEJ3jfU|#rw$POqm9A2 z%XL>7V)d4$g@+BUVp)jBy8a(gI4(NhvhsNiw3plUWPdJ4X3^cekWYmE3qwd{xS*Vk z^G)ZM*y87Zk%))44&-L%Iov`5(f&DChh@a5S7|CkSJ|`#({&cPP{mXAbe|E9j0}^pRbLU#~ z&5XFhpVo_?c7JErX=3A;FfDno16m%%JLu}3fF9lFFJ<}M;Lo9=jEbG?xEoacqQ04R zv0}f^&*j?B?MqLeW0rXHzy}c$SDsW|f;>No|1xmTdwv$XD<15Tvf!+*sx&&4HTOzY zI*+FFtr_tUC?$u9r%!tzd1`FCqAWj@seHYDrJT?s97_&6BxhJa;tTlsi$7HvyRMC= zZP^q7^bLI#;!OmP*W{`#heu4ZfI<(sX|vCX#Mi&Mak1Jv9P?A`zI^iuRP5l2I~Y() z;+4{zyY6QNv5!yRdu4O147*I_zIoPZz|^Va$#qT2n3(C{Zz#S7h7R|a>F5nZarrr) zui31>>v1%Oe)bcI_vOn~zn>-{b&lqY=Y>5$)0JR8U(en*NIXR!!*nxG18woohKSiC zFm^j^Y_y1m6}fbmZw4J$z2;Zmill8Y?6k7~809%ke)X%}ntc>#Cp+YI-0A83D;ea) zbt{lH)HJ1F-A3XY5IPl?&*nRWQN^KI;~!Unn#%gIZhZ?<1PnbGQ4Tov@#!A^gMaNP zzmEBfD^H+LnD52~Kl`93aN*joscs~$qubtPKMMzz^;dnLFD=-$BjqQ&UWVG0bXHA%0cPh1 z*ZFIIg1mitsg`_cNL$6a@I!Y3$mxT+iet_2HPhs7+*J{vq~F;XeN7riW$98yR31!K z6S$~#HMQ>pB{b2`_|iH5G55iqkoTcLQGeZBxi zZtlC}KsK)c4DxZSv~eH%JI@<7mk9N8{NLn7u}4u(|HRc|-;m;Gg@r`CX`*txZujed z`eQh1@m(%Wos;Kf`a;HA^S3V0mmzp)g#@98NuK}S;!;WuP|eZWqJq7jV)13&*gWcL z_>j8Q)ZtPXO#S{;N`>_}D9rWEO)d}HMqfO=O47ojR?S|Y0m6^&bR=yY&xrvFS7ns= zds;eIa3OicN^zj( ztvqJ%WDMJkpAFHNZ;rB$Lc82{#C++w-q>D<0tV=Mm5hd!3Wfe+G(*y6SYfB(LWwI;T!EHn z+9a>w|(MUioO}p!p3@ zYZuIuF)o(Q%LOvurn)s|7DK;Vn#iCY#qai^P@6( z0?p==p!>~|x-=_Q?D080*$b3Jx0uzO7Kam13uUSAU;#yLE}cEQ^x((q7v^@1;FWFfVn+$*e$U9XWY zn0c{{Arp>)4vK^qS@`BaD_I=nfXP;rR08cuX!E#p(1!CO{FSp1aGG}4WCkTC( z@(`G_-$l#1i8w>Inx?sxf&`S2?c4)PO|UQ9Vp=Pb4+hi*??@AT-`Fpsc8BD(3%hSW z82ElI2zyQ6m%Hl{JfqZIb61mV&5&{7bHn`=m(#hLejU89{TU3>A6D2T6FRAsAkS5| zSnmMOI7+{DlgFX*g(j_mJlG<>Q1?b`6Hs!*mJ2Acc@ZI#@2AXxi0pJOZ*}_51!Guw zlyZK`5xshG-Q5>fu zxgPuTviYc?Uqmn5m&K0rV}AbX<1Zh4$4)!C+QB_1U?!%#F29`$l){3R6;^AoChgF% z^R>M&o_FDG{x5dDIg)u^GozkL0P}`y<&LX_yz^yx5?1x3ykO&1aW6Ii65tM7^7!>pvm7ufqv z`kbjJivq+jHDS zBSDFNgnmlkrHF;~eDau8HMW%B(*`KXLWft)U4jE&=pxRm6|qfXoBQRdP+%VW^kQ8R zCr+C|<*R#OxDsbXzXyKb$F4`fwuI}}ihDGHWnt?3d5uy!Z|dsd-M_6c%e!2D6?ufj z9aOmdWbzke3NQ9!j7>t%DyyJ?=5-_mgI0fU zA#iwUf@@!tc>F<2bIBv6eZ4dFh5j&~Q`szT{&y8Nu)QXc@ zrhY;t0gzsI=Z0Mq7gD1q=cnwv!oHqs*hkJ-T>xk`4aL_dR3Sl2bc_|t5!b@{5OxYRt3ej0i!J}9w;&}R(0W|9t%2wt$+FV(Ekt8Ornf~A*3 zJfODfq`csvK$sa@JO8w8GfYia^@h)Tfy_Is|s~7BZ8>3ey9Y24HJ9>7c=$SeL+x9zKjPZ{t--z|u2^-+ zPkZ9hxi0H|$l9&}y*nl}4{hqiwuQ>!wpVL0{``4igXZlx5p?2HfjJR(VYwYSnH}}l z&OQ-&4rj!m%cpH!P2p-1pY=$?r;!+B6^1LeMbUs+pf7R#X%!UiZF|_goz07Y-&%b? z&U0ngr=<%DYbm>*TSTnCvt_cT^R{C3$VjGHN&-|id~$Qx&c>CJct0L$R-fg8Hob4J zOzTf$zQmMBjwF$H8JfNwWU_&nhm0(B_m%IqA}(3exci_y&|ePE@9xXS*oSV3xv6e2 zz%u_BtK5&QwBUD7_1d5^smDQ&PMm*?@Vs4}1188iZ|5KPo!CE_k39CZ@34km$$2HA zi)E0xxKC3xe=j@~dfc~3nuw$HTsHXpL3k@rLi(B&;)gJrR#_0CrwHx4uBLxnzvy4{ z0~oeKYc^cm0=+hCgVMiq)J>;zB{fK#A}g}5Z!M#x$-m&h`9MWctmOk~LZ58LF()_h z#y%X^;5*C}Cy&u{#uL+MZ2oSjpV3!(pk@K&UK0<$Yi>}ZuhrY@smjqV#Y~RpZYKxs z>aQQ7?a;D=;K9u`Ja<2-+llDUHxE@FJ=#Xtae9+|-&$&b7V}+Jz}0}`f4V|+kNF=a z>W{k94?J2TjF7kc!0sYmTcoH4?BCnH8)}BBDaAL^Iey=?vU5;g@)5`9Z4y?oHc$Q2 z`ITMcyWXdvo6fD49U)~%%QCuBs8I#&R#oqxX|wT_*nQu3%GZ7?Qkp9CYwwKV0MqJi zdD)O`tZl*^|JTnkFfs|sj1$U?FWFTkj{x2hj0@2%JW9o+l+wu zVEx9-gd!m439ddzVd6bq)B9WY6aFWw>D2rFTuz=f+PU@Xm6BJ((Bzbz+%_j@6Z376 zJ^Tad(c$IhyGn54Qi#Ce4IDa$Koi%oR2p-^1X1-JUnaf+bNTVz`$Th)MxFmbKO_RC zmk5vSzBdKaDMw1?ttb5Oo&=u7l?z_LppFIC^+g9@RAVbuc{kyo&^tbTYHTL-X8VdF zZI)Tq!N^{#k8?{n@I&n1jel!%{rp~ycpr65&~n!RjNN*ikozJgS=*Ts;|DMh# ztFyReTTME*=7K^=jmuDdST{6|(DnGgzDI`TuPf*5AK_>`@1DzB3A=pU)iy!*JsYnAF6w>%-JqqMRtfx#k>SNzimxi;l$FcIx$j@9c zc3c%_!znGjSxgd7a-`$vHV4>$BY#M|=LN7Px;Oj7HW<6~lB{Md3e!smFZmub!5pnm zF%pkQi03Sh`>pZP5}8@zF+*EK@h2(MbKTA}(7#e*;ESy!&S*dO`sKd~uiBi=aW17| z_Z{I}zB5roKeUFaz~V~ia3j|bZ}tKj-}nZ9GcvSMhC;G=DzG8_ME6=2Q4iJfHFxq< z4=1j8bl$GaY1z9ZKFZc7d=&{uUvs2Xt%GgfL(MJvbFSc65?2AQfZv@%$fz_Lw$)~G z^w*dP+4Gvy6J#`7^w;hwf${0R=CnryUZ{KdEz>n4kKxI()9vil5%F(cY4ukVjXPm-vy24qYXWyO9dp(#Er^ZN!_G_l z&SJz8pff9np0n7xHR%GXU8_89hX6}Jw7+unYk#px^MwoQ&{UMN?tvwtFIgs&Qy!G^ z4d`ldmoz_d*nPx0cbYCbYy1Q#e0=5!$5e3oo_%O3#gMKHg?{KnLdWo#(g)YA(*t%0B-m8LBvqcl&|2pV=YcNMVF`d^fsQ*A= z0FY}oYqy)}W2k_b7grJcJVA=fPPq*|1nzX8xIFNTA=^*Dj(D#xbIO&{d8FVj#!u{&C9MYDL^#6(PN#`G0w|W279uiO4c=W!Cmyo$-(quyqp>xx? zuwPSx_5)~)jt@E~?;|x?StDQD2}Tpwjr@olgAVe(W(m15pr%Ntt89vZzVuDaFEaGm z|2H_N|Jk^wFFfD9YVv38J1DM@aX;2W@W77TXcW6sQwbA%Dwd0k5W2+Vpry&*V^P~?o5lM$=6$h) zij4Svm2~cJzE%GZ!u~rh=lA^|$J>KQg;W%ktR&eP$CXetk(3k-(omYRS`re{ppphH z+G%K>S9@yjz4y3wSNrGl$M<@?Ztrf-+wU(oSJ!wv&c|^c=P~a0`$0BYga3q4PIIU$ zG6g)(QR#*%PeeZQUac$74*P!X>>GDJ970sbO*t303<8nr9%6j_HhOuZ@U`fnbQpEp zU7_Z!if2EOY1A=01S8A?zc+mS_wS`=9c^xOl)fvGIFeU^qKGr$_ch+&SneG?Pv5&1 zM=@g&82f2QA)VmY&6U$fLn#DQ9>6Cte#-`&VT5p$aH?p7iGguQap{QYlSm z*YH#s-$#3HXeh$D6|>v-xwz3yGB~y}gT#XESU-;PGq_n~$6KJxF`rbREudJovG2w+ ztU#m$m8fK3esrKNLDR5}!Q;oVRt67`u&tS=jqUq^&<(q9Gxjg^?`r-u|2zeL4)ZGv z@&0$epgu%U(#$&?tyD9=fRqbpzK`r>`;vk4K;7Gp93zo>)_krm&O#=;Vl8iARMd37 zc8AS>uWM#o9+&{M>fEG*X9__pKJ6Q8P_?_ zBQyPWJ2tOpRzCja#v9S7M%tKTCrXA99wGS_cShWa?{ zz2!f!3JU?hRu6pi0P3J-G}mD%@G05%?YUeBq-``@AY5JtlU4#==4Oog1x$VRtWW&$ z6sWp$bq;Y@ovW3ZRoLVZ4_an=9I9y+3S)^!VpteYe%O zFKtX@Fx@HIldesJzZtxFSn$-uH~R1c^)vU;WpO?eIe3#%r&PSp6!#K_&qIf0 zY?!W@(7iSJ(A_1B2V53@>RWuLBLs1=nc+i(V0_Dt5yz+DXe|6Y-?sKD zAp0s@Iq*dp>UUVipH);s!&fi21n=7ql&WsaC2v(keHm=2lk^monlSCT>x_qyihbG+ zo0nlyfV3-Cea(JCD@T<=rFZ9}>EPri>io`7a&P5_j+G;#zSh2%qI>NBUp&rOmj)Th z_a$jvK(JnIjLf)!3JWh!C7s6nOt;tMh7)H`1Gy$HSCowX{z&g*{6?)bSa{I#O(HrC z+DnI-4`q+xSg(Cb_48Sq$CEprH&^^vMEM_G?l7w{`btnUSix_)4eRStk8kc(5nrueMk^I>t~?rxLAzW?&s zvzpg`%{D-G>UIe^mKHQG7$-Z?bOno_l1xj}q$A)9W$WSt^>84{WxiNUyntqpH>Mpv z%?VkVrQWYh@Gx+)v{=0@6xmbV=h6}w&j-gEjBYV++X(~o!}ld!1yFSQ>AVYmY0&@Z zyR-L^zd(t$$yjHO;T)(pY6&d`BZwF%H{70{faY$QI2RjZcHS&)&sXQ+3b|W#1nP3u z_N6*!Ze#~t$wX6P9Or6nIMMt}_8IFzM`*l%C_{TR94(}+>yN&li|`dcRZmM?K)*UW z1z$Yw1k#*~6Z_mp9FtU?QKOUtTJRIcJrOtvG+7O?EfxzvaM(=!zL5=4*U_fA5d-If zmRlbMt#`kVii_8S+#^OFh!&>LkakX*1GOyk$oW&@Fs%?XK7V%~>L=45Ub)zZiZgmL z_G{S!p;Wb@SAYuDf>H#xqML6aq^va$u8uBLyJiG9`v&1)pKghz<~R$ zeFH*~Kvr21+4EBnC{Khp9r~sM#`i!{PvdmxNT0{*zfkO zju!6>{+GXm_x8-$Fn9~7HfdYZo~n2tZt;-cFd+qFw@=&{Q9lZ_YT;(tun-v5`G+ zoUS}w@cSay7c12ps5{tT%j^aQ{~VgjTK^`z8v?>^N`A%tSQywIvEe}m3D34!?(Y}g zF6b6MurI-TGpemk8#(dRAIMVIoQ%(yBZ@-a$Qw~}Anv^zm@*p+tsV#Dcm%QdZRMH; z_jA*Ys6jRKrF9V->Uh3f+|g?d6rpgNBLrftzReb>o_ z{_<=KTY4Ui7V=s9Ogd7~$k8{wC%d~~uA-?_Nlg%%HKI))J-QDQzslM-H|qS~=ZNao z-;UgVP6-@U)SjeNCd2YKxlh-E;!qWJsa;u*!7oNlo@rU@Z6Bk$K~O4qTB4TTi9_p#QOA0sJ<1nv+#_^T6jA0skFw&}VD@En z)eu)S>Lu=HPTsX*RAJYZU<@?4CU@ufYJC+q5cYIpeki- zJ$2LT|Kpt^>Lzucv*j8veEm*>hk6`#e#-T-2Q8jOU9T=RD~W31nIazOu^kshy)ve| zSXs9tqSvDj)5Lf*79?{by%^K29{FvOcv>+A$8@yp!C^-=80j2MZLY!U{WBFd9w*t# zV0^>U$la}LbThHI*4HS0_M|zQ1U{!zJajZBq8#@)?gQ!`-6>Gz<^iqmIKqGUo`${? zF9Ji(QK9X}rV{;cjPn#es-ImjQyBwlPjXGv;S4l+V{x4LMgvwV&Rt2I8waX<;d`2Z z?!Wls*Ns`Om9F0HUbLQsi zGc*w;tK*s?0<)~zL!sJ?JO_yM#cNkhnPFCQli+PW%pPb@zxqEt%Y*8^Ra~}`zV+|< zsqJgbuY$OrUc<};OrMe_;qc14WpIrhk_~DHj>%x(H^sJbd_50_gI{))j~6HmhtX~C z!ZVf`xL|1SmT-G<^*)dk4|abyTw7-t?zPA(Ijn&O>_g~;R*dIHefWHVM{Lc!Fc>Ov z+CXd)f_Vc~idP^BnraX1txv53_v3LApM?G(GHcm^-J*}stm20(9j#I5cl-J`l_Y`r z1BG|pJr@XQqOp11VLFB%re*9h*p=>xX9^w?e&78H^~dYl`B_P$!b>bc;wqS*Y)V&M zFQt$Ti{bZF?0yX3Ig&RH(e1E$MR7ng|BHP`Vb$QJp!Jz8&^>-eTx3fZ%5)6cR;SiVe@9N4@d{OUK}-TLW6f}C%){E1TqAssr)TR6eAnbk-!;5 zW;2t-PV^7-Qi4z6+H+D&IR`8-)RH+hP#h! z|26nz9FKco6p7}(e7bn!vNGh7o$qHBYXkAax5Dp{Q!u@2L9iuF3C0%BKAOKo0dn8> zq}4mdh%~9nDylw@rj-b>VvjN0#0;JgxO2b%7@+O-CDblVl={f;s zWu9t&wX%ifw+>shx1U3l3avo-%I83K*r+^WR1N)E_ipUJc>*m7r0=YF;(>^7BBuaEcw6Vb~d-bw5*xM(rSurR$QDl0T!~0&^eE{YX$6h331% zb#wk+IF`wft2Zov1JPoyLbdm0JZp?q{bnAFr(4pp*>U4cIigo-+!X0!Ja<&O_%`eY zj-i_gt@=^##=;r(RUm9T?|v%R5eD{tY-T<=3A1WFhn;lR^oM{TaXYD_8^hl&Wh`Fp zKhuJiiuzR-wlx0BD@K<+BgKIbKGSo{BN1Z$*qy&_i|M2e`5h|2fBcs}41Inn_p1+j z@vP|)Em|-)Co+9KJ`{VObaEcLR}DmOyK>p@*y|LjXZb7hJ775Ae4OH|W2p1Q=2Elv z`#9#B9DjL549{1$Pho?9%GAGo|3pnGFcVE=ifwZd~e z{_GQZvKI}y%Ki#@T7~D7OO?VYn1EALyZU2`F7KY9&W~n>UAUE%?h<{@u zBKDW&yvW1qF|^w|nlzL*!*tfE?sY7=XhQfgKi@b7=B@Icx$GmLIm@EQjVTPiHZpyv zIIm`_rkYJ7r7Y`X)T~{A)DxZ^n^WgwIqnzEHSjdDZ#%8two@uD$k&4EF zS_V!6(M8v7%CC$?ka$44r9COj28UMqE*G_4UCU!D?;G%u-TL2iWe8POoeeWBsMGkR zbU7;x&1&2r?&WMl6w5!`kT#CRXUs!txI<|1&VHJB>#tQnBrPV z;rp;gUkpfN$5mxth{4!hU+=r#JJ39f#U>pk>^xuPr{D6hD+1CP*03*?2y$FFM~9U& zpicGchI3=LQ6HrG{W*6LsuZnw_X%M9qju&N$IumjKp(3gB6G|?j|6*q@j3bc{X>Cq z`#7U74s%T|@n4%tQP+KY=kIVFd;+~&E5zQaah`mUAMP>-#cYy_{ksiiYI6YA}-H4*)O#~ z)4+v&SMx&<@ypOFFgL<8UC}JvLA#H>aG1Z_ZGg>b&cE_36qAb5^xjEm<1S^wU` zyfKEO|HF}x-4HK^DnGUvTwB5HpSsT|FX2lFbl-GKEss2pV|B9Yyt^=s+V{SqZ+ICE zgo4DMtWJ)2mO~PE8&xGxi|UlX&^s@b@onR;>(GcsN51K|q+f=%`v)7In*RW5dHr5P zH%46+bxZF17Afl2%)~z0WpsB2jc;W=5^)Zz&vmPgvzq2mpoF@f z)vD?ip5^E=WEj9r2=MGc1^~W+qA5;+Ti;{frBZLqo?#UgbulqQ(<#J5RBu0U=*Q zmjBHW2uOSoYO(GlTFUM?%Z{srRA)A~j>Ds9RP82VRVp1JNBdiE3$Xp4#;g6BOR^nJ zm}DLZydMkga!-$Yed&STzZvXz`*g7VJ}>jQD#r}kl8#Mm$vT1v79y_&eq-nA@&}0) zuRroIo*zn1DYpgs^R%rqPHXB)IM&JA=T1z`uUVfNKe3Iw>y9Gwqn4Gx2fpY>Jx=1j ztriSyTc6x7b`dCVxdmCixk0xJKa=b73=DleuYc)Zez@fQ?37nJ8|rI+LGv4b1q<7Q zBa*qo(exaX#K`r-(08Qr$R6FJ&?xuCy&&ZZ5ZX4`mO1c2q1uz)iTQXOlff|4FP1u( zIy|{Iw(%V*rQ~XN-%LZ)RAz-&ZPqZoW#6mg$LVO2>Wk0aDuQxEuHCMjz~%(ElSZv# z{|7fY7DPU?7~_Crk<+Vh*|Da61*GiNicfF^G;RyQqZmhb- zU583r-g|c+oPZu&+C-L@1)iDI&i~eN4$Y0}7diKM08z-FY0n6T7jB(x5vz7wK()QU z!h<%h<;_sivTyiV>8-skZtj{*DWx#BkmMszX7IKV$zb`S5Pk#|yxMYRp{y6j`h5E{ zenTt{E&5RZg5`28koS{sJpA+nNEgbaUh1s_;(Y~gVu~UfR?RtNxC!HFj@$oM@I5Js zrcZ^5#T)6N{ymLELv0c4?-#c-KlQx@4Of{)IsdNd&obgIc17vHx2L;d=3EN3eV~rkY}rgndu8O5NTS zTQsmT8Q>k}f=Ye_1>;_0eEz__(ti!+)c@-`;E$Be_#3`8Jbf5bmCx%AX+tv}PSyK9 z8snK*?)LG4Fc3%y+0`-(94MZd@-zZ>KN%HC{;Y3b$-=Xh&g)w4WWYLOl(N z_O!+_C3~PTqs>~yl&yeR7UnlBc|pJ91M5{GUKr5S8hUR?1(K~mYDh#j#O%vFm3MFr z{*&>&&5ch=FI*}`?nxo+{VPtWnQ75+I04h$?{*7r)*FySGyRLV--|dv``fliZaa*3 zI2!tS{jQiAG$49%x_s&}8j`Fzek6Pz%}JJ-eL2R!i6Hl}rR0lT7;kz+F0*GeI0mR} zX^%-=Yx#V67Y-K{D%hgww3+WDe-9Y3RN{3uU(0h}jFEr&q}&olL&}_y&?nT)hTkn- z%i#CI;QhKG(n%_wDv2k1Q3EN-=UblcvfT4ye9h}qJgSpN-03Y>bH$q1+1hWXg6%=ql1J@q z#WCZL`q3DXU*7$_I*!%iESfvQz-J=jX;VobyZEeq zW6*N5<=N%+XQAS2;P+zg5kv~qIlqB(-M{#*XPf02Yt$CCt?i~k0UEMwkG?7T0gW4+ z<$Z2D4$V;}b$s=TK!`WADL5(%b3>wetLt4+w~(SfYvMZ8Af^AxTaE+IWZ^3PaA*x~ zf6eoan*T@|m2gG0B68Mj#sZ$@jo^BvbO!zfsvW*DMLkAlqwakeFJioQC)>`${HV_L+wblhTVS{; z$S*E(tv&}~&xaT5lbM0K{cO@@uAM-R4gOPW=kt*pL(fhGuEDnhVeZNIh;3`$kNO__o}8|j{=cu| zSi=YO3U~p+6n^cjW5)W{G~o}C%)6t3oMUsb_i7pt+89+4D5BSMtC+C) zOG^_;+P~WwbyyTJld-?Z><7>`r^LG^vBId}tgqjx^FVSn-F7WHAE?qQOEQ-6h`?f# zT8~pgMZ4zrxLO6_nSXvu6M1wML04q9~(aw3)T0$uz}*7mo# zsP0T{a+&@eSQ=fgAP{v6h;REZnKthlgy~$YjAU#k58ec) z#-BWobMK;5_Z{8qd|sl;DZx!mi`Y61n6&Q6Jl%q4IpM5p`3s9KZ5TsQgLAqQ38a)c*2WM%KeFm_OeC-Q+?xB&1H4WZcUK z!uid?Y9{J{|77;urqB}48D5zAHlq{`8!X(Z9%ty3fQNn2k?bT7n3_ran`dwth?{nF zZMVTg;Yo+8qJ7O!;F{-N%(rIV9-5W@rDeTG49{UpHVk)XhH(5xV$Qe~6mg61dVil$ z??;m_x*z}kjK!IAmX-ne`?SLGu|Y-R3#}i%_47j;A1Bm+-s-L z`o0vXBR3MB?y1D~TR&}%-oXhI79EdAG&bRw_6DsWQ-&VWKfkr@FliGJ0;1o(w==~0 zbz}~e>qdt@0|{7#I78Ri0}L7WZWdPbMiUxG2qVpH(DJbs+N6Zmyzlj1&ZgY9B^=Yy z)a~DX_abUX#}KPJ3B{kBve0Sh0z%xrv|>vPFEAz*Ys4<}4)x4pixF- z6RhWjVfm+ZAptqa_Jvu=lMzf0tM9%CXOcIDw^%%LnkAa;1N7&N%JiN*33+b%FXEDh z|K&%*Ck{Pd1mcak4@IT~MDxkNX*25#r1>vP4Liz#M1C18*2xFtxAf56CODY#E-uK% zeFaKP>G^`!i9m9D{p>pLT_BO%I0lOKP~Fwxdux zP;hbU%{{tIu*BQ0DSe|GM&7TN`fMNoM4pW^>YPNNkcGVzt_DMe&18X6Am$%?$rn$t z+`#bnv_}_o@kT5#!SdmrFg9`yB-e0!GUFdq>kh^t(%z~kW&;?` z>F<88l3-pFAUVe<7U>7ylGdnc$qz7fl@VW&XJNogaZD@obN(WHw}>#?7nf z13#xu7NO=Lw8=-@jtnVD^0<5L4>xM6R7gr0ww+tEHRJ@HFe#aUvO&hjqDeH(>Kf zsRFuUQK5|d4dKuH;=7znhaRg#&ReO;5K+3$j%NbH9T8(pb!+fAlq_YhNvu1KR(Cgr zDZR?Zu^Q^=-jPD6kJP^K<5VX6sQYzV^aKZ@-hXG(l*@*ex5k@O@~h zCzY8_u9+W+V>5H*U@j0tRYzMC_F6oExlQyY31b&D)F$&syf6|-lpbLRvo;)Cce>_p zYkgQ%G}XDrWQ*p$rd2jPV(d>m_eQBBgUJJUuDL#yTS9C3l#PF52QF|#z@W@(?x}sN zhzOr=EZQ^h!!Wc*VlsDvaX)CHsmoJ|dF}Z^4c?hQ^cK_Y8x!R_hvpe^FPgfMp<=H` zMWiL}hhpu;Xt?io0(TNS5ULf`Qx$2@CAgj{hgX4Q_iU4u@i+|dP0aLv z!|+Q};)m9OU_Vg*X7;PCbL zj=+2uZLw=`6}4%n^)g2Q%8}Z(Iy72~?dLI{=NlC<96s@SeB+mCMxO;TXxShzDJ4GIy>fVQ%%M5eR>H8q^yYLRV-;8(}3bz^P zsXd6nvw!gA-kxRvnO~X;6(39ktx_}g)FUpSIpJ~xgq_gO*I9n^kDO8O@{;ZA4a+$8 zGrOrFLP3xf$nq}fbUF}eJMX2pNx*2w?#&Yd%c#|{;G^M2>|E^(nPG`IEV z=hTjXq;>Q5KG*X=7O!uiJdg%rl$jv$0Y5D8$+y*ywGgM@tadH38OCvC-*L5bCKrx^?&mEo&=i4>Hh7h zd(o01vRHC9X8gTsxA5d_Cf48SJuj%WSGNgf)_DiMdQ}GWmWZ}+Ym65)rj?@j$Hfwj zN{*O++`)-s@yyy>nH&ox_x;j5O~cXP>Fx9SLKL)o+Mk?OTnCls4<3@R-HnJQN9D|3 zubJD4XFIWRq{$meVq%de=QqUsNsg645(u70*U?w_ekgJtI<=-Y{DQlv#Xjk) z!4bJN&v|(Mbv9u{1+&+*zI7M)EKnk>HUy0&0KVAa&AI)lh$?Fs5?r zGXJgk(JErp{UM>3Y1<(Otd2gLq~+||qmPoRJy(9e#Q4B-r}m~gMy%jjo@{znYp@f3 z-t#tO7Wsqdchl6OhVQ`Y#K~uk&P{k`mAIpJ6|d3X=Y+a^n~f;XslwUiO)d)GJ7sWH z<^T*R`ZTW>utu{Asx!F{82bgsx+l;bR(|v|CJAk0qmL{>X8;DwaIhxurTqrT<;6>N>b!aU1j=xnE7SD{m zZeh`%cn^f~n;AxZSl^QLRB`40J}+oJT2Q4tu^(swMI&=I7#@p|`XKDRTLKVnrvJTE z+X3@Sns3()ZihiD@=m_yN@&af8e``>iN=qfx!$t4>;K*dj&<(t_b`24l;d7t_`^a5 zb$+_Rt^05r8jCX+Jga^imah*8C57Q{^YVShO6pbS|}hu8)`U^8S*uE^Z&DB z^Mm}?VcaaWe$Nzx?*fC1jkpN36^J*LzIQij04c{g;@dJ-2b=(bCA>M-cU|#%cG=)y zER=DNI++xXq4~mr2ZAk_-tMSu4ab))ENk{1>-dK99kc~hL5*rlo_LKY_Gz*fPxF8x znsBP?+anzF-lnl*mM%b+jn4`mi3VDIM&qTZR_(*eG`oR6lNebHzw=Ofly z#{6Y8UUGnaV|*w~-Ahr}#KQ%F?l(2UDGCTx&L%9UjiR~8x1>u!bwFVr^n8Cy1`QpZ z=e)`N9=f%Pu9jX`N0aN)&NvZw0(Cw0{tl*W=-*-|S#d-YTDi-$=pyX)rAGVd(lm>9+IAVgnao z=CNU}wbyeP5!APL_Pq|I%zKS)KTiQQeO9(v4Ab@R-;lwhw($$f?~^~@X~c|Z4_<2= zmy1G~lP7-^ga)GoPs4XPo*$v0j_EANSt&%GIjUGuVuISvR$J>{69USJ#~qok>kx73 z-j?~Vkw8?K)YwmcgGODgu1b=^%1~v%u%y=CHW^=<$n%?l@*Jxo0N4hF=Mk0VE?ha0$)+c%aqtKRBG3 zACK4$rW(T_Pkq?*Y74gRPP>JUpFfZ0)N6;Pe=b1Ny#r|rO1h|l!!47U^#r`KQM3BZ ziQ#y&vb?jzL?TfN_qE#l*BSbycxL|FRIcP~NYTtNcjw*;rK$AuRdozKJ{*&;n&9De zYw~aC*$Ho1s@Kdx#Nv)uQ~j@#8M@kN&IfHM>heSNzNhx&Z(=+@9GmB=PC_pu&qnou zoOVoSusL5dQtG;Y3fIgBVf=mM*baR`2P}W4ZMyY-zx-1u)+HX_{T$Oh9M=qdY_z%y z(Ovw?*~b5Kt}2Rpr)$s=$>66Vl6;1BZqo-ev%}!288F+WO=^B?#!#)KA}#aWA#&0#=u-f{JoEt-DsG=H~f1D>UE>PEvE zhTTBxjaJU5LgHxXlT5VwLjyd!tkBJOym>(T>SWK!$&W@emEuo_>e6F$F=G4MLbMbuIzjeeGB!6 zJi~ccY(g0mR|I@YaERpd(?|;M2zmRr*GoLu2ow)?Q=67a7~3EwW_d^-$eRoh*Y*h* z=5M?DxI7pr+BPy#6Km!S;#v1?{=l}q2St#^np?TGQAp&*XJ=W-K+Bv@G7A(&%do0v zbyppcs9v+*H3FgUAYt2!26I$pR5h4h(|`t3s4Ei%IWU_V6+}711BF~MqK6eQ{==kE z&$E;p416?>DQc$kcFO>w?Eb!RQiW0HLiDo9yGJtJQAeKNQ-6vP3<~XO;}dkkbJlr; z=|qI1iK(K%>n@nS=N}G()N3ANh_rjl28p+t5PrG|smWk{{EptBf?EkJsPK*Udrl2^ zRMgkE)i^H}5vnVG_pEfQ`P5 zGYX>Djav8*_W*I~mdQ6Q=E*H#;!j*_2y{Vqkr*mdq~OIn@~Bh zqt%qw1NdD>J6Hm?0^Qt1khaka#;?%*$0o4&vN!2nKz;ddAeZI@aDV1SGly3OUTLlZ zWuIZ=nU(E8KKMmFf!Pc78~K%O9b!R~%BLi+{9(+Uz;j+~e_F!z8zt9>O$g->{(X;d zz{R~-opv}sRc~uKqaK86*ok3j{TR=EKE2qB0!Bd2P%oD7?g7Ke+=1=`da&@>Yxkz{ zXy_~EkxCQ|NB$cdO#YZN@_#hlck$)q9bP<(bK8NYYb9u?^`Ya2^q(;Ar}KvL+#hJu zhb%eT&SU#KYEK~7eJ5xb4=r+GGRJd#^vj?7!iH{F3}`rY8Oe(D>{ z{O;G&vu%SSD}x*tbQw{JLE(0OhG_o6w`bxB1hg=JZqpMFf0zSK*~5AJQNg->T~a>N zFd!hdWAV}d%-5zV*x3sap#QU(7$HOyszuXNvQ?t->`o`0XPNFmi-cp2&PxR}@U?p6 zK+XVk+G)Hzq5J^5zN=SRWJtbf#Qz3;i~B}IH+K^V5&0Ka-!a}M>!^-u%{QA~^tli9 z=kAqT9meW*RU#||UK=|!wh)_|S+xc~jbriYp4=(=5E|_F3MQ?dgZ`qKodqgQ&_!0_ zb63RpNeKk;*)0W(JPHvHYg6p~E+L9ZRZK_?;~asgf~#-F;`SmcLDM(sF;*Y@t2%c3 z0q+3xQrMFh-xfkYfiDJCZbCECQ+2<5>Ji!VbBMHcaD)8u7POKvv?DNxp(6t&El#%ndcvqyIce!s zbq1pCC_R3I4Xa~M3!VCLEu9!*`d+ zs+`C5kT^>u8k=$vNPgRd+P*)%6fyl4EeH0TmA1HuXVWaX`%b$HVtgj=CG{Re%bUBt zSA6L}?H^iqjcr$hp-tz@M=nmGxjVUOIc^Woz@r^MU4K7;p(IIeQ6p^r)M|WMVnQqh zh}pk3W;(q=Wn2w*hpiZK9*(v5Z19Z>Sx|Cf>&ZO&HPk?{#K}FhhPm4Zcyi1}V1j%- z@oTvk%C9e6_e>YlQ|WhX=T6-p1@ng=q(yjagfXrC#Oqo8Sp1M1h|^J6LKWvp^aVmV znoJJqAN649z2RB+;F!}+q(b=ynXx`RR?i@_D0vwi>jP?Ex8fC5tPVAn$${RJ2M`t3 z9b?KcM~g!4x)pns0RQ!#e1Y5tAnhgo;xgWbDtwmTw4bj8LX4-{k%VhNwcj@FE=2&! zgO}HhI>%tH`iY*}xxJvbdwO&1eWw4}*L+80!_uH*=$G~fjqNa}vweX*qXQZ?4QZ^8 z!1ReJ`g@FA!a`8kk$c?BZAYLdUQt2G0K-cRkD}pTM+RR2&2w;Pc)!AQA%-J=h4R0P z#WTsSZa&N2Z(6cIrGGJ9OGO= zV~OTz+zf?8VfLpo!fKngVJ)t#;bs?hR-1hwWh@#XTN4}ZVhm4K2Vp5L4D z8GTShCaHW(@n~dRSL2Y=A#)iQpiJ+WJR_9x?>X1*Q`@-)M~dCY?Fp?9$1(kV>dCXc zBlc_Np2Om|WdHZ^2&N*9Zka4KAex`e0pk}#h~ll6+OB}ju^WiRWxtYhTyy=C{*_j$ zSGTQseKa~(%D<1HlY+{wYSmp{Z~>}?cW>n4A4EO;xY_U3RWujxvVT*TK8n`gD*s|- zKMXc)BNXdSz+kJ&g*%5qm2J7=057RLn$(4tG&Z8PF5TlVl z`HRaP^-VJ`yW3wx1@Sh2JD=eRpUCH7&}4EJdqTf~>Ay@d>&AvD9Gi;Ea9 zK+SvBTb0Ri0J&kY;&vayf1%h?4PI+r9Ypd^(%R5cj7GVnzg}B^8_ny#(ek|(4P#Ta zI&lq~QEz(W9L25!hRu|{jvv?ubFm*p2KT>#5vS%kBr=0Os$^2XZdgVwA$C)RZHIuS zU8!;Ay$*El7tDE()d`au7LVdPNI-wf*4XmY38n0hB3R3gtXbFjGUJzQqO534=;pJ;XXqkyQxX&k+o41bDZYR=RbI52pBsHiu=W-!MV+JsXMt7M8m zfAki5_5DWFvqIW3wNeM2al|KD@oVPRFg`zcQls&*#eN{&jrw}iiVnfJ+Z%V%;}Laz z3wPYJ|KThY7y4BP{r#r0%K=UF}&tJf76{Fs6$Tz9{G81ulMCx7VN zcef}YtM$hbDD$cJ2Snz@Y$*ms9h} zZd)L&+w>K=HvoB~RBHYBB2YO6T%_4#P|cX`{WIr6QKf>xpxOK5c#fejG6P&B=(1-y z^~F31i1|^EOE~&q`KQO9-CuVicS1mxq1GB*P&BRqO#wd`btyE}@%rEc?xS{^J3d1w?)!ZGC-6OLmzu;83y(B7@2SnB6uNeZ2{2$M;FH|(WZXFs)Nvb>Z z0OQ{eWv4uL@>zwlVq%2HbR+i0HRC zm?DMYe5lu=MJ01XfZD1XrFS9{l{-gr7hm>;m7ZGT9E8=os+Kt|XA9rpSxRrK47RRW z2Sn(u|KyI0t2@$RO^!ehvx`&?3%(tg_~9f6uP-%HB>BuZY#z!5>Y{Au}xmOm)AThafki7jR~d*AEW?C(SE+nl8; zxES+0(41JNZ`JT67$C&&P(c zmf5dUTQQv{()(4<9lNH0NU|D5Won2}m$J!o*BO-j;jB_S6{|-Qahv$fN-^B%Siv7T zvj>#5cDHa@XCU?ys{I6~=oy%v`3fbZwBi0q=rC;P`dWqIND9X)Jqm|+0oneP+!yYD z&)uqAPVlV}Ob71r*1!BO-YjvOPP_3C#yr_|`(&~6rF&riYyGlc(6vAKRr6zvC$X^g zm}l0TJXCM%N7Y#}MN=)$!N-BjlAJkSbBS8Z?Jfx%~LKmFHzgTD`a?>rqp z4-KzS#B-cKB8N6yf4xBlM%S}>tNdc%IblRwrkjQs^P8cbWMuiCa|a9ypAK8rWb{RF ztmgdrHFvSOkUbpAvdUMEqXPV+8Fy0!pgiPbdqLZYdQ~0YXzQCG%Ct#9L^7j}0u%eC zLaGL}|9x+H{c30l4Bp;UsJL|i$ISEe@x7)(MC+|wA9~5=pI<16HWbD1r2f^{-tj&< z1LG&&9~}$G!n1aV-}&Yohv?y!tG7HEcw1QhrZO8TfXyEu7w`IWI`0~054sy>r6U=- zg{XH}O!?@E`~UJd%f%gh3_X1m*;x13<_)GNO%aRaIVRAJXTQ`u{|%;rSmr+ZTVDd^ zJ*1kZ@9)GjL2>X)Eljt$Zpw`7`VB_Bfny(-P+_zWLBpYThc8Je zKfsj#j|4BQj%Kj!%YFrmH##J??2AKvZX%~L0^cC&7NHCN53#->;cZP8v*IU2ZB?kMZ!BP!rM$bT3OH4n0&6Y9{r z_38YzxV7`+NBRxjvb_ypPISF`Tr~+bT}g|>WnxGAz3$7 zA%ovglB20H<@WE~FXaeI>@>l1?b6igK7Db`_Yd#hbLW}pE*u-{^{X}>)I0)?rax>N4!<-)tq&DNI{J(MRe}yC; zrmevXBf8+8V@q%l=uGz2=O4Jh=v`S(_iKC5(4G7JMy?TzbCFFr@cQANod0)T7M(lz z1OPQa%D>ON-2H9Wpe~MiXW`k0n{`oNrIoUy3j^1JDh*`~gs#oNkiyFYwLz+|IK2M$ zwzzAEDtwH6jH?@__E-9k{xCs(NwSad!kF%7r`uosoo6uJm=zTl!V>cZ=;dFUSWgH- z-L*O-4aGE68NAWqw^b8bw8<8F_6n=(6~_?9zTJ+8;tNxjS2*9GWUc#$q2Lp=;;8r7 z6tTI$6d#g2FA<=;?UtTUOe5;<@agB3Xo97OIr~mdTR_HjEhee2SiNO={Q`R_c{5Ne zj12OqJAv{N&s7%l2hnR}H%RPa%-Kaxw>>aoi^1X8q(^rsNW}rM;nm}u>}Viacm%+qc)yM2h*vXjP2=LHgt(^-dXwnjJ+& z)jqt3xhs-#uRgEgbF6v(iV+-}rQtQXH&#F*UkKXi!ii_e2oX|xaUIAJT<6iv!Zp`j z=V@QTq-}@Ur@v|8)HJ3G-+t|b{EqEuh*~nfdgDhYp7rJGOyfWQR{d32>G++mXwD|@ z(jGG&D6+VyBzn6A{0HscF7je{p;+xPH5+Xyn7Elk)>^^pPKC*~&20}qq5=)x!+)hP zop}W_jV^Iyik zgN6x#QzoG94D)fiNX!R&Z8T5Di3KJ4Mfmz;m;eKBOG~2}PGqk>^RW4MF>D-RDUGsif{T21RJMr}GpcT;i zFDC!(!|Fd>vW;A_oKJwnWTdML6KG-glgHI8r&0Hg>*D*WM1cAMzxXw+0MU8YW#CKG z&|Cub4?5WjvqJ4GI@K8Ndj602vOHHU8dK@nd52yO{YwsJscx9wQ{wTM{t+`{G_)>J z^lo+u8hWjEATgF<7tnT2wDWLzButK*ODU*+gnPR4u*)IYmx-fN)-$QwK^tMNTVQw6-nXG8*EVwi27E&2mAS$LdHx)lzS zVPC%l-v5nbwT|r_Xvlz>&XM7j+5gpViXYo^^ADbbw(Ibq_Fc&Hj~D-b?mSFtDW0x8 zB8n&#)s;Q6Pf&T6AMc+F4Eux0%O(%cDE~!6yKSG(-c!c2w!g9ycHNFzwmqvqS7eK4 zH#ZjEYKF}fUsYWVT+Dw2B!$qo1AJGZ%bZwdSF{H;a-F7fk`3_8@y85|C4GPz^7P^R zgFk`Fha)_%;D=i7yi8l;^*HtmBM?>`2172TRaX>$1I?4Uc8Q${Nbd^nnVn#V0dEyv z-Y|^UIJBwjh}G+V`Ou+u4u6|jMA-92@5wO=Oo+T`@77X5bGJ8(&M7mV8>%_xxDL;D z4i!H&z^Tlv)y;}FtNXQ$j@1jh+QoOSn0<%lklNP`DFW2;ae6a4ymn7BydH`CH)8;q>&9~H0<3N>!>z_WC; z$p`tnK^cd%dJxeV&(tIO_j?%+P(qZG#8$e|Jin2G@@+0?j=f{vK7-W<={FKYSZ+$f zL|573Oys&X&x=lZw;zpRb#>AT>!Vu&sd$#)^{yGGQcz3Z70t9#o;7}r-jguW(R>SN zZ_H{8AC$wq(KYoy@g8WznRCqHHa0JpR#e^N6#JiiSTs9%@QcpbdoXS=_&xeY4#eN7 zs#ontz{MlR>oSCns~R z^FgAcS5N8r%Mec`kWG9s{8vl-q51dkHehkYW){n~4f1HRIf>V9QwaL$-(2;Y_YNvK ze5D{S*a_7!J?pR;ISuo-(%w$@c*C4&!PR)dQ0O543|R8Q^d%R$t((morSZ(K<@aUq z2;w=mT+FhF@#jO+?nc#b#Lomq-bWO*-v{_&9sunSE%okIHX!m9rXP=Bhk|)m zt)NXBh}J1l{{F8hX5ZRAtaU}v(5?DzL1VWmq76nhf1K`zp{^5~XZ`xoRLTJp$1N|> z;??zf~bKg&H#~H4f0|6bXH(tc8 zKZ8DJXY5g5(L}Qj^6@ItE-W*lsLN>NO-;F z=iyyF(EBsmI&b|9nsX5UlpMg&kA;cU;A3TCGpJ0^*orIWJ+y3Yi^`2JLv2U?Sc`Ns z^e16#W5;6wCk4oQg|GOoBL&SY51#ydCI=LeZ4G;sIDmZ3@pQITJ(jmRs49rZZ9p}X zNuyWN=>Pn>taR681T=Zsu$^7<0g75>(N70wARM=t=p{^Z#G2FjJ&@8E= zEQ5CisAuT+)fMYu=(rlogGZidO!zsA#^%4UWES($ZQXN3oWy^PvFrx2)=j2yxh5FU zJHAU>*c;FKm+Ha#+!RR8G7BbE`w(@aSU_O*7wT}4ou4(cgQ&bLTH>`jpvIioyxsVJ zbViqt(sdqVdI02uVN2x30W`ZscxRqM6^|>9#CQE-1WoB@{;AUaNj?!RYU! zOry|iHDb?Uc!ZnWo;JF$okikS_+sxR@X_@5eLs4n0WPUQ+EZ{mm+4J$Z)urT6B8_(=TBE_9U0G*bO=FOn5oxhS8wgp&eqqpMluC z;o2}sA4VjKt;%jQ>OPRIS@YT`p#o)GeyO;1)fyE&qxIGNKi)lhOBGOBMwCg;buqze zs3U(sMfdqzRJYfdJf>xe8j>Sj_O!jgv-Ra`Z~1fXKhMEVv!YeMl9aXQqoH?F+0wuG zHQ=?nwIDXfqUEZo`L}&>h|ZyY#J_`~FM~Qf@fz8}!mt`>QRMiLfiu7}$qrR2hp?al z)?4q(cSQX2w^LIzVT=bcBx>(kd&~t!n1_e@A7gk~T3N&+ebpwIvg=BS-Cvg&FJqb3#Yp)jC=ym+OHCtaAg)+Uj<&!?q_@+$4ZOkj8nntv3-ZuzorfR^IJ0W zbrK!_6F)H9#%#QR)i1|n^FFHEYH1#u*3Ro8H@&bf{oSs~Zy zdtKay2dnGWdCIb69_Yohf{)}M)i4;56Hw~)!T7h07%Zee}t>`t1>t0OI_hWp19 zSIPhI6dtOa`}Ct24fu;d_SFYaBK_;);S>MtbbSq%cvmti^K+}$ddiLHn|mY+-8>P6 zl7Ook7zE;D_3Dl}Csek|v=V9iFF(BFYCB#3^j{ovsf8mv3`lq1as3)!2gFU+-^H-6 zpw8CzbIaN)(8JB|ZL0SHNXA>GZH6b%;(VvWLp_A1(R{>e2?bf5pD6NAzyXmjwtew!KshKIv^j^NcY+2uA9hduJJ)Kn zqnB$(iV?YczjbZvAflERF|G1rb!kEcVSApyQ5dr~KM`e#=}lDEnJhfs{N~>|@RvF8 zejbpGpDZ2D8$+q&`Kp&~6ENXaz4bD_91T76FgQMK1Vf?fmukkIVf$y-OhmDtH;ksp zWcB>c0os)^>$~i%C}Kma>z4C^s3D24JKuvJm2ddHFn%BATsrb9YB_*f;-GMB$ORRU zf6E*25n--G|Ekqjtllx=!FuL#E)Nul+y|g9wvT$SX#j7>#N_-!9SD>A< z+%GGjUx`ZJBlQ_N8#9O|qKb&*wL6zDcoG%uPD;Lu$Ld%Gi3K7-=Hlp2r8KK3)d0=$ zvphU$Vva`EEkp@4eFln9;VU`6sef@xZ()>35X@~+@>U43fW*;4O%b9O$~ZIh`N_Lj zMB2exQ8XcrXJNHF^I5kP%_OrKK9|jbk<$~AJ9)AC?QDZQ|BmB~d740Mk-UHlL5OTw z;=xW~;ED0k9b7D6o z*-)CBE_pAjC{P2+Kcsn>{qs{Pg7O<1fzt9Nfy5OGg|+?>dn`1BQfp*U*r~PrN0y1}k4`6Ib3ns3ED|K% zSz-1rL@v9S?Sy(?sw;JMlTpK&47zk5pk5x~6N>SLD45qVkKt=if&2_7uUn&O{helfU)zFf-7?8s_8V zE79zNMV@$kI?U^iZ*-x{{mX0PUN}9TM&$gF3-(3#fzC1OTR<0pG5yw#rvuphipq#b zCG+;B=!YY+WloO) zZI{K=Zf{F8lOvbcro0hNDA6CWY;i@+%W(I81cu|R7j6>MYn?#UzQjUd{WUt2ke8}g zo<4)^w|ObWKhWoGbNgXNio2qzY}Y?~9a?z9_TPF@Er%1j z+gt2WUoU&F@CR&8N00u#TT6UC@a=HtEVra6qMlgs^=o6S3nF=Ha{qXl4YbpiXT`qM zA+qmfA*<3X9Gh+9(Oq2mi16Ie=PgeK8gN?F@Uxi*+KahxRJJBQLfuj!=gx8c%k$s) zDO){f=nEmjW{Ez7*;$A@Ns!Ab&45LovjjEHOQ_d8jvo2P3H}(1JKlORkCq-#*|>dU z5i#9cXL@oF3ad_JoMz63a``Kl?#2AOuFTG_IUlgRoe+FC?ait#8kF7qTj*F68ZoE~ zI=SM)SjTYhWU=pO*60(VnPVSs^;%UdX$1FAQmzua2et)JEb8Tk(Ltko|aO-&Gcglx=8|RDYkZoEr@uDwMp!`!64JcTJf)sSK3Z_fFAv(Hbt;S8_^l^} z-tHl^_b=um%7xqW2QK`En!+W{to^2Vj`TDf`@=PT0UXm%kCNZ+E{IJ3{Nz!qEShRK zq)T*EfYk3_zP}+oL=>>6#yx>(?cG#OMjFZvUa{|%H_HN z{n#Kxt#2OA3EKmd6u$Uu&ok(!xnCIJ(E%Luz>zmgM=GJ{);i6eaXZM|Z=asL;SG%K zKQ|bf=!hn_ls6k4&_^wa8ZJeBN3rJ}@(39wt(lXC=PdULwb$NGlMF>o+ zB%fY7RS&~fq+FI`EI6h}PF_^m55ryxp{V^NjIU=BR2LaUO&)Ar!Zi$h5^B`uN?VDRs)T(*O5wQF}!vXrA^BxjM49cw!POc)b(FP zy++A*-4^kvLHRUE@LWF1{QNtQeYp*hhI}k^<_%EtqLA#%ml)rwUyLvCX6qJcn)SE3 z&>RbkxW+dtdl)<>7?t>a@V-hJ>OgFLlHJ&xN3v4=tv$1aIJT&i@!yrlbOG{}J zDi!co6K%xi40AekMNZ3qhVms2wO;qDXz{SNQSPw7KR@tWZ}S;EOe{v*@jGFC_4*#f z_3bu;CjOX|EDe-5;_U$)jPW#L%IzTY{hk%Dq`C0v{iYoOVeLQ$7iUZQ+R zjxH5@fvS3Re=bHyA`)-J+sCiZz(iWJLE1RROQk(FEsOYt@eNvVV)y>6$L8qH7Cm)v z?ZfH|g_aG;q4RZ&|M$9{6R9H7bNeR%QQ1{p^2_Ca@#dX|z-bcdDck#vgV`BKbG7F)Uu;2tS*@}! zy=_8$%O3he3WG4#xu3RmVGhrHeOq$u&w3zOBinq2=1Uvwn`ZS;Y3j%H%H|AY%FJhGcF`54^Sy(!^s?~m zM;`Yr29Qwhe#PSic7D`$Q&=c)zbl^U#y<4GZXX)|I%Vc_Lmtm@$;0`*bpjBcW$nM9 zbPDE{KX$x4uo*4%_1Ye7N@ds!i+EH4ZeRhHdX%GQ-Iftyy^X~V2u4*0hRA%scOpVp zqz?UEE{vVyD}GzT46UD~gqXkjK&yO8#%BpHApfmRJo4=zqJ59NdqRr@&6`++ejj1r z6Y#9mhCdu6gk#q-8>;Fc!K!}94twPplw)sVq1WR9oyKq698M}g`%!K3dt`}Y9lnvR zxuYF2>*rv$7Jx1stAAk35|OXY8BGLW_$N}gVV#Vn1X?f;;&2<$h8aJ{*}HB8H2uOi z^n(U=jt?BH?tCa|fQSj}wnX09hPwGT;T`#sVYH~%*HT&^(On0~T+dhV%yEr3qc=OF zK#r7mGtEa)cA>mg=V}w4hi}2;W*JyeeP|g!k9k z`Ehb+NkZLZVcQzKH1bk;$;?{!2Z*oVFBEMkz_S>A{*_bw4Us>!>0R?}K>f{K#@d%P zfc#uJMSAu)wAu*?XcQcW(Ta%ryJ~-6{?3o^6T0uwOmx%OV$ePq-K6V%aNC-C9#DsV zBpjnL<`SaegVu!wB};3rXJO@Yz^;!Ds7a$T^PGGolrNI_O$`T7e@=4hjL#2jA3c6| z_^O#Sj6Nvm`E*qfhWHhDxFiKoQ^xdt!!j;JUS8U$RnM54g=5-U^-14&7*_l9>~!_8 zxnXW|d}pTvxNt1ojqhfZo2b+X}RNp|-VAsp+9 z^SVn`$5F){C-K)UpI~DC=Fi}3-2e7l<-G@?Y|t;+GHvi{2aIPNjP3sU7!3}kwxxOI zp|Q@H8J<__fG1@=ncCP66LuJih=Y6YZ(WI;R-0k1~|9&sJa>zLs(OIqrSR`Zd>hihkrt-J; zK(qY2&Z?sWFdp!h_vGqb)Sy!H6P~ZaxFE?YHcSf7E+_l*sIw5vyf$UlvA6}DtQ?=y z@^LWw%2eQK#5dIOqdmR#MJJx+{m|t2bU9kkz%L0Y9a!_6XF5<{*yUT$AL_ETMi+Zt z(`O~+gx?OUWnB%rsF7eC+h)b9lp4hzc0pfe!zWkNiKo@hhzKaFCsh_O++!{ratO9vU zoF3|r<E1q&AfgPA&=)WOm;R5{fy`Vnnz;_E7ACbwW+-pqmPG3 z7n`ph(%=WGz%vJZRTG%SYl(-=a{w`#pG4~kh0fHfV`I0UqmDmn&#tW&K%QmaL&_ND z{|PDX#XK`N&??9F`50*&47}IcCRno_rL_4+T(ONsltSsz6Mvmy`Prt-cfNv9^IURW zN}vvuhR?Nxx#|G%8GBf6$!i#FchMf_pT^clZj;qCArksGaew#YJB#NKWmfxHDgeWI zY9~LL@Br!0k9%Wo3|ua%-W;s_TnNLVuuCrr!uEp*t-PPV zebVVC)Lhi$z5M{g{|yGGj_dUtf=@37&kjsG;n^u1)9O-YFk?M)X7;Tmn!855)G+oQ z%5FY(;d+9_-vL#>z$OO=jm2&x5y@ZwLamawxqQkn;+U@I6TNYbopo6#g^O`3;hB}r zc4RkWx(vm$?{ARP-GTDVQsTrGDi+@-o@V2717E99h$YU=g2B=p72y*5D`9xI5rj+ zLmi7<2KVVYSlAewCKcO&DvLSj@;f-7as4TJtX(?hXLh<>Xx1dwE&J)IhQ1v~gZ;%LXhR;M+7ln%NkL;-(UIX{XRiz!m3jTpg&2s9&6-CW&YA`W-Yk)PY)mh_$>`@Q3>A zTvW-0HFJKU&-esqmlFxdXA6RSzI$W$)7m~re;UqMFPi@H-8b4xFk;yKUC^1qLqt@; z^@9b{qA(TvE7@gs9U=vseJ6gk4-qaIxLviiLY2DR(}_EyQLpIf3X@3+ij%I^PSY-f zf;;|ZN95hn0N3O5!IQx-dgo!oXG+-n;@8n|AnW=Ue$)7$ys?q|g!JnY8ca;P9b#UGioG^hG95pJLbINSah_w~ zq+vni+C<`ZL(E=}6r9ajUPKeSn72r3Vf<-YNmM{=AvRwfUt;~zw(vC!-L913{*KL; zY!8)T(OsfL&z`EhI6R~7jOTd5-tAaf1daYN<<0^7&_t&2_fs+>c-8?PX00Q}Xjtvj zy&ax8sQ>v!iykn-Grc+V*i^}S&3zA5E9h#9Mxc7TB^x(!H5k-1;r^P0)fsBNUjL|0 z!}=QeY<>~;pJ)Fc-dMX*er?^yW1U&VuVy?@&J;$q%cKLP;N<}w?KS)@tgf?%g^5k? z8`Qn&ZAd$!0;Ab|Kli~NJoD$H{a!7Yu6ydr;cs3fmL`gR6)r4s_pYi#xy#L}5+5FsWY+i}i^$)I>Yq0su{gqpJ zlqVSZ01T#OoO)}=*!NKXcJd%k!5&x+C+m1sZbWVI13!O6M!@LTX0JROOlN%9=8#O& z$;Qg$-x6{h@f@o^RlfRkb1c+ z^2V@p1n=~9IJGqyvu|58#}4SM9WtK0A zirFR%`FE-^AnrVlG7d#;8%w-0nU^ckb{~W13N!)G4QGAuUHcEH-Tc9@X6swIueG+ zI~xZzlh44)71No|W|fH6(r9l!H5PGyE_bDX~{PxeV2v8b5JVBpGs;^v)i-ZSvpG zmm@InA>I>O4YpjY;l$=hk7z$y`uYLW>FC~lY~h_SqfUkzgw~bRN(BNrI1-m67qMoY zDJ=#WD#o^`j4<|2-j)Rll$B>x21a0Ye{|NLi~Q(!o#($;Q4VNh2Jw9r$M8%^rV{C+ zpMa)W_4oU443|`7eeV~i7$Yx$nm>+q*TpnZgjQOz&|EeU8%6r?9$Zj8cIDK^F&E74 z6z<{+PE6pL%&#rFJoz787gKTevyT$cz?t)tpfzD>fDgyOfkrkZP7%qR{ZpaMH>=z6=M!zEZJjAo_2vIHC6b!Rc(#3{v*Z2!q zkbarpD&U7>yI_4$e>Xd7Se58>_@#nI`Q`kL?^~lbXXaJKdTSVGomh~4(v0Ws{bG7B zMH!KGe9bc2t{|$k>$@ILP8dEqsTRzvf=C%51(`o|P@hbWpoSR@<|>EU4WtWkY#*Xe zrW2)btX+F~cPnM$ndiH4&gdvw#lKeB;=m8H^6lT2u029ceokbUw_-T9ltHPjtLtIu z?BhV1E;gTWInY<^v5p`jraaZX^hp~nSNHJ|197NyRVZ$-tP~Z#_upkRejm~3b{89X zOrdV{f_)W!wbgHdZF-G-}N2h}I-aCRMY zLvt%LV%wu5fJiRw-IzHBG)oy;di+N;;GYeFV#}z3!(4#dWe~cvqivS-9H|pt;Z4pS6t7Am5nR z3kKQch_q)oy(Ibto@v~ zDWbJS%3F0SLRYTE_us-4L_C>cr5YOw3lDj!?~BoadU)o+ar$Ge&bE~lX^@m0~+DT;_bJtN$BIU7;Du9s~;(gOW9K@no!+=%YR|29cv7D&yP zt~eYyhDJ_U`$%nlfXX9qqOBw^sO@!6*fo0_ChB}b#Q8QO;(Lu_7sna>13at76p!MU z$1rupvf=V6U(AnaX6P2}F-NIwhpT&WSRR_jv-qJXQ5IU8L={+s$tY{Q&Vk}vjYv$& zhhnlAdUvP;a+4Pc1!yT|T`%8jVKkSJ`=^(94OVimExZaiizqXZ2{p?D|Kb_5mACDY zkRHobJd?`sgLviz++)|FFi6+Vj|@}1g~mT!eZM=3(a%KWt#YXGZyT!BoLA*3!}=k_ zq_iq}aW0zMVU?avZ-F=csiRIg-M@nAbv*Bo6rTD{KbCDci2&pgyctG z=w@4?69~xW z$axWg4pc3~^E;~o(_8BMwmWI%dJoKBe=C(HasnDFO3Sw1J`BWjb9onkV13__Kk*uQ z1q}QYkjIO@y3rKSup5)Iy}K@s<;t)*Zq*06yWf2A%P#{$M&6eGxCt~id2dPW1QC|R zmy5&|7&;*EJ3U6*azYTOYG#H)S1>(^3ICO=HIdQKGBKNoL@}Nt{Zin`>!C#K_n(g@ zwlce{*)OO_H*uk~7S9%u?VAxUf$*p3=JDYhVS2thZO2nA{-=BnP)Sk!3G)`}*(W`C z5iKa`*_475nhaa$eg9<@CGV}$u=L;nvdifsocqEMZNM*tPmBt=Z|g5U8N&FutFET! zTGTLJVv=Hz%nv+P$5?i_8$`Nw0rft(`SWkSJStOdU^vD`ySyj7-C?LRPb|)PoC(pX>A#iu z3ZP$jiEw12EDWn0b-sGJ81?ny?rY|qfVS6wk)2L=4$W{C95c{k$_d6^6w^w&fKo z;n}W-Uvz)u2un&mmuAOB5LL2}NgWlSy3O{VJ|8&?lsleAgH}$6Fj~&M<&-H5gcR(~ z-1i&Lk>^sQ_G|2&C{%mv$k5qx3?DXq zyrJ_>H`b3LrXDleS=I?9HV#|UOEA4IV%4YHrXPl2Ua<3ZY#CM`>F=G9>n^r{FM>Qa zKU2BZTz|QM>xw`vHotP7_0#t)e9A!OyIGWOi1oo~qkEUkA87+YSNKatf;gI+KM{8I z3_B2CA9na6cL&JltdCnPQGxDb;N4LXgJadpIumxm21wp)mK^H$pq?BkA12<9=0+tH z3q`_l%oX}SUHjZ^$s~lYDz)(NRt4%UK0en~4=9K} zbMM6#Zy3+e?)nvt@rX%d(mEVIyWj^mL2}?^G9sL&tLn4?Djl-kpLIAKwHqagOkQ$> znSsoMr*j$oejMArJY@ZlF}Dp($BN~%f69b@(xI6nrEx&`!527f@&<-=-e%L6u=#xR zxb1zB`_BMbdW%=!VFsS)pWQu;`DV(EXA^CClrgvqUNP;r3l)0+gkRR@f4%*&X5YVJ zzIG>d;0xdn)l~`0-a^BY7mH*R#4vlY^I&;XD2Hb{FY%p*B7hnid67+(5f`H6$R|7E zTfEVfq|#Ys*HR$fI(^N%1go0hW3pbwPae@nZcP@cZb#og zPG_3zzK@1t?8O{Y-Jvu^M!i`s5>b42h)r8i&^+;A7th#57-&44{SbWxT%D6KK*^9o5#+W!(3~hK`n% zsb4^z7VBTk(MM6Q@BjD|WP^y6F2|U*W9MSOzdyxz`)TMk3z_N3aDWNht*l051L*&( z|9Rx7E>NE2ect*9(@UQpvwkl(M?tg{sbdPR-_fl-4n1dQn9y|UJHpKeOo%f4BDPI5 z81*e@_&IioLbmou;)Y8_Ksmq1R^u1OS6sDga&z9i3FsDI9u})=qD<}ky}Y~Lp_l$_ zH#WC!$TN@1GKtlfC~urHN{-1xD!ki1#fRzhPpR*9Dw$M5BSv}()1x2IfGu-CNb4O$ zivMstDYy_?<>sDCOk?MFQNT?@<-M3L_4I%d?V&9;M}+$5^kv>d9}q3$h|tsO?~t;k zsra_T4IHayLZQk2OQ@}(n8MmYL6ph3*@eFSK;)|NxYIiTb+6c6L?pjLeU9}zB{5%! zJH1fysqkN%FE?YTc-|Bhrp(Tp-x)xIje1laquU} zZ8J^V{+6P_`?}&uyLf<-?6WRot)*&^^TaD>96c{S&-_W~4fy(;qQ;Lyuh$!pj z?e&p?ONZRB>W1>nAw;RkJx4Uhcqp{rPq*c%JL8#Zr>-48S%@<5Vj^1n4Ez-gUQ*m@ zRzL%)gr!GGdIwrD^(9W9zJx}4FBl&`*DA6nFU zWP-oO17Yg)ke|sJMC~k4l+IzrF{?Z6)jhTc2H4EDikR1;nXc2@xVpkozxD+)zKz>a z>j|HLE6QFl-AMTuWjKkF3@TrIY&Z_ol{jwWQLGL?ra-cN*abXGwxV3j)^Hf~eSc&t z`4watIy>)d-Gmm!8_#{Y@-L6Pn!Y}4Ya$vx6F<)O>0g{y&>J~<@)A_WZfl#9!uTn3 z(|5UCy*?rGeGY1JP&Epz{g8u{lrjHnH<`+&eI0dtiWnVyeHRVyOU*YcScm>jM4#;D z{e^;0sV-&g76GF4Wb#qGBxHxz>K68dq6u_Sa#Fbn&%P*Ea5mNvO`N>+{iX6>)RSh= z{X>LJHoiD3QAzOx&IJTqGmy~X71=%a(EM@HkY7W;Kn ze2__NSi~GC{ry+JTPy*gmcxi+B^r$+IfgheeZ@0>tA9Q7Y6|s_;a|yer=U{Y#>YjU zJ7C89lKJt#bVQ!G%{8f|ik9~2@=2sO=idJBDX@ zZZo#Q`V~xY@meKsVc@f&w_aS&V7C*{DZRz}MN83`Y`*b$!4M)w*T@P@VEy=XhqAmM zLMo7{tmm1k`9HdC(pewfw@3lajPLyL9R`jK&lL6{UYVr=h=&_aJ-f=N*CNu{XeG6r zGVJ;m;*;H8Sfk+on>NQh% zw!RQLZGxo!T)T?eH<(@$I2dqei=l|>jx9|BkIC_MYOB|So-JW4$iXzeP(=zrlkdiT2s z%E=(gUZ-tKyMMt_t;o{lx?Px`$R$;#wq*cOYlEBaAuJDa^$K8WS$c#>w_fb}`|CB( zoY`InC|8y7@W5h#&wJBv z6AmYn#>$Yw^|wIYFO-NI<5;H+ zoN6yKcxFIk@zl4muz~(3Gmcl%F+K>b;)0;=$!s8%dUB{cVt$$|M%Qt8{og#VlFb*M z3;pf_{0f%=3lG03T%SuQ!cs| zKMvwpH<(Q=uweQyq#^F!kti7;bBEBZXSmSpvu`VZpSD8XmdSUQpI~|sqo?=q${BdT zaOlUeNv~gM)YmWEX2uoI9NE8mKBjNYeZH?KyYMRQE^2#`)qBtg)9dXb?uxlNh4t$O z#!H**^S44ltfg_wA?$t0zmvq2-48(7^^)?FDHu+5>e#97uh}uMc>NyNhf}@KpCSL7 zqj`?;_sWqG{)6vKA*VP&!egB|P>;vyZ`gt9-A(US4IRHTf(GUexjI*3b4kY86`Wl4 z+hNd+6vo%33;)0S_g!4MupQBY z?R*Sp7<0m*B6w%2M;{($MFM`A8TA34`-gRgP$io4!3pExBaZdWCtl6y79c$^eUNqM z-#&VB*xubX5T&0B68J*>0>uK)Td67RKw5s9^OPWu{AN9FCNA~h+3acpq)xm={d?ZO z-{Y}>rn1=YiBnTiw0`OGH+wE1v=6E#P;>Acg|~UmZ@mH|zSSQ$s{5f4L7~JDk%v(H zZdtPIvMgE=zo%IIjtM5~TIc7rjsw}dz#-jj0WDT29Jl+y$P=J0Op1fYnSpCU^;H#a zuj$r^kn)+wU>gsL&8pBSUFAWeit44W<}n@zrA1(4i#}Elq#ewqI8_m$|BCRf+QkMa zTtu%t<)5QLht<$`a399!<7IO)nhL^NVCtJ`}4%^eFN)w|52ex{D8 zi!+W0I!?EH)Im`bA_g^br?dBFImK7+=<2^^D-HP30!-Du&Qw!Ka& z1^$YJh8@`P_5bbv(1b z{^9iD=RopuoO#GM@c;P@)bC>6eoUeRR-_ytq_+=E52XoPe`n~FpgDK0JNmaj!f@rv zXZy`>vHc|Sjc;_4fJRU5nsoZ?3)2^LD|II>K%ThOxZ&k1K;Pw{c37kYaw|BSE0nM~ zuOkBo9kWNHfMRo|#m$|8cY^f1cUE^?OYki7QV%!(V1=%zm1&L`YoJc6ZGXBAJNKi9 zu3V|M#`LcS1oE4zs~CMEJeO<)&r1IuG(YeEO~ynRR;T0tbT>|*{@xew6nHV+;Qlv@ zo`;OEITGWN4WirSFkJT1&7_x)_n$!Rp4(;gRvGii@GLkBPTRzPc6`{vA?DdEo?Y>Y z#rIr{r!m@aVLy_Wbbm^|F4l+K0SHzg`dP-@!; znDyMgXC_()QYVfU8s6N77F-^y^?oZylWk+FpPtJguS=8UJs0#4`O-iMkU$DOpA8}O|3xC_prt!RkbN1s-a2UM@sEB0#G zbyGu00yX#JU?{Ftu&M7l3`c9WXv&N|=F8I$<1G6& zILmmV;rXMM!e@@-SgJg4ez(E;F4QwpyoM7WVS2gVxc$igylyY%Sz$!QaqKy3XnYUT zO(o3gZyvg@3)Ih+sXoWf7*)%>{AS--@HLn6Hin0Jo;eH+hT#d za{=fZh50QdrQzph?F{B_Y`*!VSa@pt4vY^w>3;%7I&m<*>$grq2~)zcZZ%GwZ-Q8(ZAOh?@QM~ z)H=~p*-}r)UsMimy4($&IPrhKAJ1GWzuEo;hBKtJ z?Y}u;I|TW%5l?F`-a|uZ!>RP!ztQrf*!o*aOh9h*My)|i?`rTc{;gLwGeW=!${qrYRm0!7Ec0hU4b%Ey{Zb2jMfz>7|84=@j zx7Nrbpz?HIL6%FXa`DK3NTVo>5{Ms}RhWS!uD?EK`#waf*S)-ZbOjOA;>@Q%7o$ao zBb9iXI27Geh!u6N0dn1~_;WKBFkEEv^QPlI$esO2X^SsJ*=0LN%jg$iy!K^S%Or-2 zSokE!Yo)=!QzCzbnasTAT%de8(nGG0#Oy#;-*3}66;-k@uYY{g9_ak>!=!JQfXM8P zKlVR5pab;znn}r3NS4i&Q zJ|!&o0xfOvYz)o-F+k40?m)y}{;~Am19(>Ao?Dzxv3U6x?n&wPU0sJCAvpMQn}jUSG`s}wAR2=>m&^vtz&1KNcH!i3i7HLsIrmG7AT zl7cB|Zsm{*O^7VxUL(H;!@pBsm~?D7%-GLBJ8raF)qNawZMQQrSpJu9@L65T6q^QW z!GXidOjtj*=|#cNekrW4*RC~N{@}qsdwYIFeB?VDqI0PQRuWq=yBYXml7AQDLp4Tj zxa1wb0uy&P5{QX4h|)=%c!Fjj?~CvGm`03W*W-ON^2H-xnDV!)a1CVO1yS1eW6|&2 zG5z)voK~zh%2+?M#f|&J;{`rM&095Cw&y_f(T}U=bAAKK<&=_N{L24bCpZA74x7s6^oD$A{cN%+r^rz~=$CNl6t5G7MRm2;0mKp77KFi2+ zq3V@YcJJ}qI8I6Zh(o~#QB(et9jTWBp+w-2$zLx9Ukip}PXvp2)&a#X`^BI2deEsI z#;c{b8y1gpf6W;OmU| zTvnd{%jvHeCLdG9%+$-wfGd!$`BmTCzL$Xti;+o#6&%I9M1 zL_d7X{l`Zf+xeT$7M6@U1|nz=;Uce|MD5$sj;l>NLchy9>hKw?FVOtS?^=HF14P|< z$58&@nt7gx`f5ToJ3Srcz1Pa0kY+`}s-E(J`@^c=(~&>(pod@f`Cvo(r;m(nhF+!$dQ8 z&knSj_3QP}78aPJuj?(rV|8EByz)irHXj(}ZLx~Dn}oVuVkA0^pP-2wViXJMHazS6 z-1%JhFF+M}7ZGF>1e7FJ+^@DwG@j`uz|8N9V~@!cKDWajs->h&b>r>OR9^MYZT*Ho zPA_pHibcS%R;8rI+ddfT|K;qsqY@}2cV*-4%4_|E{`SHN+L}57_V?AtjV(3xKu)^5 zrNar+>u-Ou-SvlxIM8-EZ-`pb!?7k0mBUFJ)b0QMxbe3lnC_PrZ82xmneZ&@e3$RO zXY@^Qtb_hrS7SCI`fa6a0__zr-l)ab9&sPM?B>6i%tydLTH!>1vk(v>&g~x`&p<;F zGj(LCEkNJubvmp%1l8>eIrk?Sn9Ebmy}wYN$d3}n_Ml(qDd9>=pb_+|I@VEm%#h!grr+P**x+O*_D%0a|>ZoB>2?C34iTYOXOJ4lh0*%Yl+ zi|3H07rC2=!Km&}xe#k?t`PlfRnV4gF$f*;zxX&U6^O${xLYstP?Mrqal;T+2g)%i z+%D3?$eVC%K@QhD=V^%Y8f}Yt5Csj5?5F)c<1qhlY-t~jdlqK6>;7my8HK)3vi&XD ze|FiYA97gh4@~3FQNG20z_IT9zQ?_H0hLkSl9!KQIuiYX>--l79T4fp35#3iSbuqn zK0o?KaT(@1+1}=Dev8PhBJ5q4Fy2e^L1JM;T?C%>r~;dR^8y-{mHgpx4&w(dD)eZ7 zx#|yNZzS22I@J*2q+CgxUK>hOS`xLFV(>=s9QTJcc}>bulDIHy)XYD7_oa|*b!^ed zL)8_XBTexCW9`qwsS3Y80MwiYLZ#9~R7gsrO09!53Js*9K^m0MNF~uAAxV)`B4d&{ zbJj6tp6BtHXU=TqcJK4tv%k;p+@A06cklhnI?;1Yqv*^o=o-C|eiga0@ z^KOaG(BQfD^ydL_GEdXts=U^FIN-7RopBvkz8|Nb6Rg92YcC%_s@>X6)YB2bT+gb^ zZ@Yvelcbb=BP5vMOY867luzbs(aSYzV?mc#5HQ-l0Lv#>1p-rWi?i} zFR0Jke4WhyzPNPKiNo)SJws(fj$u_~zDlXM=LRV_c(jVwbut&*?)xbU)ZBr(#qWZ? zE^xsNbB&|xMy#+~e+kd4<<8h~c*nJmF^)htos+X*cO&$ClH9lZeKoefU`0qjL}bs} zvis4aY5cux#TWjLc`&|seX5~0!HXZ$ay@h7yK?M1-*C(3z8O%qXx_Zjx)?i8esk%! z~M;ppJQU2V-o-G{O@o*v@nvv9y!Q0Ut#j=n+W3EsN? zn9DqDeR6ZR{TG7wzOQPPoOa+NsGWOW`TXNXXo))b;iE(Z%zTOvq5Sf|v8Zp->Is)I z>vS%yVzmu4+`M&bR@P}~G<$sd)^T3wk)Zk=QNIUWs!L6tI*386wBOC9iswj9PhR@? z{T(ppCW(60O}~Vn(wjMPW3r&3zfp{-Et&Vubi&g^ggsib>#>IX zObCqmMkpys8~|!Cb7a?-G)S5$nX@^Nii3i~%oR^~An2#>jop@qk(IDx`{Sj1fYEA{ z@&0(&-}CRehP=o)_P$e?E4<|^bW~Yv9{zn3|17<&h7NW13i#vgtaX+f@XIzx?I z@-bq2!k#ytfA_DzzWerf+_A=gdyM#OAD|2u2*h5t#h#zN>Nlf4VAlngfL6C`q-JTn zsNX7$?2XA@?~8N(>TjiwRB59)A~MWZy+jcQC69D0DB2E_9dwcRi5^I=dbGsCkOZ`W zCn>Tm+<3kw-7-OKJuB?cUQm5v(N$o6krfN;Ao9*?E>>k75w?RYNi8~`2~iJ=bu3#g z>+Eb`UcJ~Ir*i=qowk+EeN=4dst(`g;D+tWBIASk=I|~|HD5_d8`&>z_@cWll=r?r zCBjF{i_oht)487apkgf7Gk6z~mt5N=m@oI{5&X$XcAXY8z~M{nF;| zYY2^1KQ>>eTaT?m%|(q@1b`|3eEW0lO*nk;;99>cnec92+Y^=7rjRbwcUaAr7!P*8G#h^b4G*`*`B&iQ~{WQhdQIpNMl8T|aH(`Rp4u`aXX3Q8*M>^f-2siYOHR z82Mo=u1*&2xF4lG8V$|v35T?#7&v{I{^d9TqTXXo5|FK78}LXVi?_vfuV(}x>@^--3qI&i1kK33X z^keVlyTm@B&s?pHderxKo&VW7v2QO z#WaxiYj;fLMZ`OHu{+%3?1AMLv&!rA7I-@OY`gsrVx4r0&F|SBkqD!d%m(@gTQW~i z=;8Sfw!nBr_0!Mv!(;(-{qW~o6M()~oKMhtGY{ToLD5TEx6R2#J@$$GgO?wsZB1w*3-K}1G*0_u=@UxNuIr{$8Bb@ z1@`X{Qvm~mEiyB*9-MlRo$qlq1_ z^=}T16Y{aM_ec3_kGsH_{p(q}NHdPC7m2wlvjFRhSOPZog!~h(uZt+YBMy{=qeg2B zx%`GudT2j{pUi_%B~1s#axE+?>`hafQx5%4CTb2}B$gIi-_mDS=8>y$+Ln+$;EA71R0RY&P`mJbnUNQBYT|B^8~VlvorO^rGeR_mn~>1 zjcv7a&shZ%JT~phD#Ha?9K93=ioRXCkk$`f%k%}^bcy++v@8rEA5?}Bff9qgqvBlO z6SH$)-MNeAocp`KR|HEa68@QPQRTaad4JE9>v{^`$_PD*Wi2&V@xf9u&-(P|ib@2| zgBjhpI4h0`Jv%No^eYoQ^lZMmZ->?pewQfgLx(C7z$mEo5gww$ zbd6;<#_We)-mgusB{+I9G+wwSxzE6i*$1#-m@Km;JR%Sq)bA_TM}OP12HK zNlXhE>`RnNeIWr}Z+_6E1nfxsw!wq0Dt$m#O8HXXa0`cJye_U@z{x8i@qIh^M5q4+ z_ABhG&mZ(43B+yr)Dhx^Hp}nBq82hqki5`R*h2>flNRDR6cHW!Qo{d-3J zox_Tlc{;%Tt(O-LP~ufqS13Z;7W(hu>kOPQXe9{+Pvf`n)???ol(6K(VW))4ZGY|2 z`-&Ho9$`=2#{RRli?KgXkRP8#0J|}H%Z`;>fco7%3v(0%X6l@X@+8oHimtfCf{gdH7zhG*%=0a+n6!F}d$rXz$UjzGG z=2Ag>Iq27etIG2^{4gW|6%Q-d)(zMuJbrEwTNT-Qek)eHZ^E`E+mgQOya3iVi5qVx z??S($*S7t?hlu;Qj9WhGLTgb(ERM&k7Hg3L5|!q3SX7fyGT zKZ?D*`UlTs68_rS`N{sf-hAiW*Xc5>q;s?pMl)h(D=+ztLw;tbGVX{XMMQ&nNnsMY zvM(Pt%~bY=~ zWaOn6p8O_67L5Jk^gLD_Xw{=5+l{A@Zg?em8{;d`*PV;4fs16J2M5pg9x=nJ`$k8^ z8XL&7ss~w0H%DP)%hMA>)~UdZ&GSTEhO?X zyM3MW61P8qt{;9uGDRFdaT34eqcE%03>=+vcc)_P3=YRlTPuI7#UJ(R#^T-n5U00O zR*_8tM!0flIBOHKXf+E1e8zz~dSTj@Pl`*fu_d2LluYRY<;+L(H=0@`UZZEVX<}=k zZ~XRXzp4?kE=s4=U8^Mt$!yPhUG{>+>muiP(}%$EjV*1MZ23g^KiZ?~JJn77vBA^) zY%7`InPiXjcJ#dG_$`n!>3&}-yA#uHq#YbL|B0h5=^dt!h153}EF2(|JWJEf0hwz_ zLayg1DktV)wcW{%MK8m_KW!^KN;ARse3Qw7_M?#IJ*JuQhlP&)^?|BpdeCfytJh~f zhry>`KJFGH_)mE5E%~%tVLLF39mF=}oI;9U1$F;S9nd{DImgrxdFUmZOyBeloP!bg zMz@Nw$APs=qFz{ZVZU3zlFxlRNqp-?yh={+;)L{9K5rLmq}yHIJags(mf77MHYe2p z`<~7DoD?_gI{aNJ`zle-iRob-A+v%X+Vg`yyL1uxPR){Mjmc6Tkn*^0S4{24|2aQM z74&O~rwZxvAgLOYY zc=-IL;8=^$QvITS98Rn_5w<%4iXTMqKlPWw771xuPzJ#ZmUFqVHSya*$nf~0JrfxU zt?ns1j*2!T`@xD6%wi>&7*}7_Bw)sQ|Ip}~%NNKSuuegF^ni*b(5~K#FW6s>-B-#5 z6zhq+h}I~5~~6*OGi}u$j80d-Sbvui*p6C_0z34gr6k~IEm+}Fqc8wWa0VD9}|v3{jG^e(6?Nl6+b31(1vv_$kKj1 z@5lT_z<9cDzqecu#$4NUS#p&gw&ol-xJ|wgTgUq19_Mw!@Qqy?w#M5L{hxK9^H6Fo zPfzQ^Mk`{0F0|{* zF$ohK+nZ4nC3F(ldynxwODYBSO4<7K7h=S|Pc5}N|LYJo6)dX^4VsUAiLV8%<1WL5 z|8wJO4Z39h$9!_+WHDgwIjAp^X^*XQ3;x)7`yj2iJ^Zwx-e3KDac%3LH8@hcGS#o> zIWSDLSGGUl@JwN2H|6p6I)eX$|E^waz@y{XZkwHT?O>u5cE_+J0>Yi3etkKD6 zGC%OnNP2VZOc;D5>x_?k$-?s*W7lk$hl2*6XLdx|Kx?o^kA@nBG^4A zFMNha`40}4AGVyF`9v1-m~9Yj8Ut)Q>e05UV5oN4db+Sc0B9vT8~O|5;oFnBTlYkk z6LLu7yq;zo!N-)Rgde0<5&CAImHigC3J%W%SzytmuKBYw|g5v2hY!cw6UFt(@o8-_`a@zh!1COS5@Dn zeG!Hg%Rb)LT1n=utGY2Yn}eT2w$gl~jGRNzE@o*@i6L;;gBnL)*nfD7{QJy3G_g?)E5Wj3t{Ch^XH!zcM4_DILR z$-XuQVXw8({A_(KcS6Urx7(=Ew>iI0w~iSbu37-22OGC6+j||ye2bf@e@t=Q^n{%I z72>>V8=Y*jG2+BQNCNhYGBP`Cv0I@Ydyj7<^Vq`K=>T5rk)L=g)=b39dS-@09}zee zvhc3EuJpqa&}-E&v98E9Oxp1f*4fuAbXvnqXUaGuPQjGz3raqv1= zrmocNc9Uao!NAZ&@f;+?n(I zFZr@2V+VC;-M{nD!yXPkjm)>FKyR%Y!J|4eapK)o|BqyzONZSqzubfs+mai_(g>b4 z2HR?Pf8I-|)ha2tdWqm+9%~A2pPhK6fLW-3gKIM+uq zig;u@iTrG;%8REL`S)N|_9^A#!5x@t8L77}m8-53j_0j)p5NpOE#9om80k(NPd3Uu z(HsrEx>A^7N?WKd!BnUg1s z!?vY$y?)!V^-A#dcPZldQhezaotTV&!RumCopCEx4vR@+95qPvVDVcq`HzKii*N>~1%Gqy7ZxUQ)y($C8FS}PAz4HO-qutG7 zxuM9MvtWn7`h7S}X$e!Ymqf|pG%`AD^+c&*3J*sFJVi)}sw`?}-g&RimRgJ?^oxpr}^)HswM^XtG~Hwl@N6Vx)yGjds0FKM>{-Cw?&H}UFqh^fVXq7@5l3QRv34E z^F-yxD-3j3Bb%jQ`0Bz7>^Stk=i}YiNM4DHu848Ik1TBecy46KS{&Ga{^5CnTb$?9 zu6~;TIcAV7=-NGK{_zm7Vm>BQAJMVrOvF5gJg)kqWF8l(FUw+YV3~AsmZ)?sb{)F! zJ#Z`$%jR6A=rOf$Y<`F!kN6N&&CE`EwwaqxtX@Xy&ecSM?|rHw^6qo3xzM<=w$qoF zh%>i;4YFLa^Z`jgLHc4%aS_sIRXq3^_aFI$1$6C_qhrWgd2ry>20x^@9GsoL@C=t6 z$X3w|$nX&3>IbXkl|9WAAl=}p@6?_|*kXLyeR?GkADyU<&wf|5oGfro$+%FRdi#lKIw;QNUXE!_)TDcj&J=|0{VN!Q;?c6}xug1RZON zHg(?pU5&j{eqXQbPlQ5h?|9+oY8<({Gtc*Y7j%-=Zg8CBT4(>?{v`=q{$y&9)I#R} z)^X%*2*H0nR`}xP+U)uNdJZdo$o0sUwf?vE$vodp2h)R&LNDpf^?)0aP;Sup-7M}X zao(<7J#uMMhU+)<|qxzQYLn0r&ap1;kp&uN1N#>IqXT{ti^i1BQOU}81oV-08lCBDW zNaCD-B;GFvR!A(4A@1LPJyeZ#;(zv^%)8DnwXGuNpW|8~GyMMZ*(|Q#C;R5cGuL=w zndZs+HV12P=#cr`=EEL1@S(WlPY{>9APX(Uhu=TFCC^eGsO;M$jnoQ;KL64nGH?6f zqN_n%e8e!kFuwKM{?$PLNI$u#@+mP7kFL7^ne&rdu8ifcrthWTsO*9@9^#5Xrr#0{ zy}^~o4}+agxAHtA;u<6I6-K9DaB%+E=qj|H{L&mFMAhowJzEWwhVdMQfsq&>Z9+>*;kgra& z5Y|g;3}u?@VY^&`jqYJ*7#DHbrT9x7BZ~DLb_bY4`|C95A$h^H2wYH^N`NVWnuE@x zgV61N#FssEA37e<3!bm-#0n?B)rqy~(EZGFweG28ve3r6!37^Pan#wMBthZ`u<~#A zKd%eLy4UUcamDr!D{7zCR(KRTMuR^EDlF!jC;E*yPA`mwvGT%q(TiGvB;NSP^PZRI z0P|9lY1N1!cBb4nR=vCpMr!0&JfTQ%p6k1AnET`AHTYw*K=MNQICQ67ujOOpLPLN| zl@e%Thq}_vl@B>_TNs-1Z@ZPm15>2zjf@XWm^e03Gol)X6zh+!o$HADaH%EhqwH8L z9PObfKMPL;*6-i%7Toto_Dy-6`+2$~o`W|>&c&X>RxhgNiRZd7HE>qu3$+}(bnmqE z`FUaf+?3+ok#ZQkL;p&lr6u@(fA{)iTpEl7e7pQ&GY7W-tWPbeuk!0K_ng|(?u>_+ zoFeV?<)b!CTn|#qE+O(!W*pzXPN5Td#Pt3zMUS&8k$Ufn=NAP+Z|HS-OSRwAhFP~r zJauXV*mi&SR$i-n(0zSs#e5wue=*Q?J3Jo=jzXc_9BR)iG5n=*`a3U~jDdcZej*n8 zVC-v2SMp;Y=+Ip9O5$Py{&{%i?Vc?*P`UgATj$VD3~K&#FwvnJI`U5V*X0Xh6Er$M z;#C5-#u@|jmRy*br}i@A?m281vk$kr&WTgu^yY4>s8~Uyulf*PwlEH<79TF`zD?l2 zdKTk&aJ$Du0A8_1avEmAAjCi2gU5DJt0d8Tz7wgh0Cvd1Wy9B zSFS$lv>4F74GDf(CJ%M*FEj4wyFpuYk>&@@F6j2qAbUO%!S=G2MINynyBsRIY(unO ztKg(r=)=n*fmpkC&3(N;A<)4a5^gV+jSX{FKDLQH3N4lw_B6^;a-y}t3rI?n&!U-)@z$W}vF5=|=KcpsqV@0CVazj`J8&+83G^kwfX6^7`t!WNef4!<#E%w8)K9Ye&i zTRy3LlIuPLb;s-$v@Yj-KQ!7mEXcS(|8M>H=ldsz;=X=*#)W6+`fhq~@S4)!T=;6B zRnPsIYQj~w3RBM*cn(}W&GmbEMgKdaUv|)ZIoj7RjmW2n&yQ->NazLD&)F)z^XbDh{HV}1g zx@kUk;;%PD%geOjI?qaYc7H|4Pf@@BI&NJm$I8iBc6C8m zYSQ!Z+LH4?d;j56)NOstx@x&X^~M>ZKX&(?us7!Ozp-MqR~fLEYHaB+Hv$^TA;qf60Meb(&ve#rhaMk#=}UDYuc;$v zU2*Iu2L2Ja&p-dBJ=V_lbko(ez>nKLovOIy1Z_ieB&oLl)Q_Rl$SiKXeh_qr>;JxV zkf=*MDkMCpzFq{n&n!)SrbzqO^CX_aUuXKZdXV@+gF7axj{NgIgR7rf&erbyyUsQ^ zY*=6X&w2kXJ|UINzYhAtDNj`=7jU&REcI5|OWY9X9`f(HeF)sk?Q*?~H^)fa?=2jk zcS>&=&_uE`S+AP^>Ya6mubRyT#*e@?gG*mR#eUy&TCClWAviYm_O~j~jnrRmJ}ioT zlpg|_ldrhiW1e3Aiqf(S+KtXA`@Q(59RzH-fY4yG{g4+cC7?^$0_A7<6Gn}xKz02X zHFW4SvOAv3M%P?`A?t?owvl^r_$7(g;Mr>!aXeM==XovC9Z#>At;7R$`yPvV=DZ`m z=X_?M+3Dw81z!K>Xpgx{BaKZs$Vo=2P$ z``<}!zB>&JMcEzY<%&o>yk2^hJi_qk**t%fNgO@s_xQ}o2S{7}$;{T?2zqBPa8b8a zgpP}^4cwG}V{2mj_A6GBK$e{)u-v}`Xc7afRL*`zT7cWax1v56L^sXr`MCj`)h(lw z_E=+vRG#?f1)O*})uH zN@6xz%snDT;`cUGxNx59oJQK{k9SE#92W)-?`f;O$%Ey4=I{-P6M6l=a}Ok0rcxpP z-R@~d| zAZ4eYD9`Ehz*ZjbWw-p-^9+kzzZ}RCA^h6IwE4!Zx~hNm!t#rZMh+hcGT74r+RrqB z6}N(ApXClCFK0f#qg9dwX2l+zExHp}tMr4aHWGEW$Fp41CfHo#3ZX9}ZEFq*AoW6H zO$m3t5Lf?#oBbfc?ytUbQo-uTi)I`T2sO#r`5vOTgkSuimx`1F**oV__@To?lA&ol ziuAqLOVziT0(-pd#_;_CuJ`vFzgL+&odY8!-j)3sT>ST3@Ar^15MXv)BMENUrMR}Q z3Qw%xb!@_&$jk0h+aMZHcMn)|GbLA^`GJX{2J@$lw*0F-5`XB5@uC=Gp!cjz6m}%|e21eK ziyo>~hAv~#n@;O*{FMh60xPVtplZuSr{!utfU3Lg(d^JkNM{xqM!q-#;FFht)0j%;M4bzjKk?E+Q=57%I)Pa3B>SQ0_K-bvNw|)7LTjV2NGx`G3Z0_t{ zx8fGiOJat-+#(@8$HjU6`sFyFxc0}zr#pbQ#fp`XI|}Tujia*i86?5IcM^ZEw*b~M zdH$*?F8v&XUfzhdr2kVN|L6QO&KVv(vhFkt?+DTnE^dN$*>m9nUFk6NC`u+%%9rc@ zGP#ttA&pEJ;qTdbMRGN9z7}~7znJ9m?-17$4j;@$xDVlCEGH;buVGK)Fb|q*8^x;GAw{@z;Wmbre}bA8S=I zDhti5gsGFxYhu{@AU}+f8Fu48c(u0+H|p_4KwNmC&yS*L40ZB6-}Ur9Fdv*dkdfmH zET<)+^mPHytx&3V<9RByg$caMclHA262ZA!f9Yab@qlxzD!~^PxU_e?qmi?&fmv=H zx?!aew$-F8{%&&zs3vI*hwlS6EeoIR_=T{m`n8;x@p_g()mnYe$%(+h^l9WMdoSXO zb8@aXdT+2OeMRuA($+cm>@3K_KMk3xG;=rbE?t~vJit}AiqMn7c^B=YECj0RBB47k zF9Ln!k?8Xk+i|c?y3F-NJ20MJbl81{AI4An(ZYTV0rRT41TRe<828T#w!d&k`m=)% zH|SBIJ%9NMowwUad_Aih3*`u2E$YV7FZhy0Or2c~NA9B1FV|NI`dxaMtv zkajX}Sn||qyX9QhE%f{_EqqFmbnvViEmyu*}^0X^1c-NwUP-PC!-^Mja_iG$*QI7qCA%JzOz(1aGF@3 zx$R$-7#Dy+9!SWpZSbLt}j+sX8?ge!sL>xeUIF}&^oecN-M zl?@ZTPgO15kMo0$z{si%w%MX#T<@pff1{?NT>7u^<+^`F*<zbh(X`r z+2?yb9|OI2G(>;(_|x;;HMf9+yEP*4eM{Jg~UX2tzo7KF}cf(k#!A-Um!F!hV;aZWMz7$@1K8bAf5)Uq z80gEw>$PnCfIYOb%>`O5iJ^J>E zTo=KsOaIXB9lAsWsMWI+S13G$+!iP8a<4^5^ZRpA55dQVv$R>k2+a8QL+x4 zeG4?cX!mNi3=Eb^#-1D2!Vz~3$v!6nHy?J?Xo^M3fwq&PZt1xZu{7<>0;A6yy$2i>9#D4f{Ez%P`Ae=b#X9&z5LOHJ)PCI!8U zH|LmSa`rRK+zk}_zGETeUcX=MZ`=iwE45eg*sKJ`N2|ks_Hg+9AxU?s?3#lunDcH8 zRp0%e=LgWL`p@T_B=mriK-(dA**GXZz}J7Mz8Tr`-yZq4<~PQQ-w-OjGZz?lcBKgN z5Po3#h)sd~iE$Vl3AKn}5WGUk1NYsFXAY8h&N=VAY$uKknB#bEeCq#woy4)jLZ0u6<7IuX2Hzkd{}g|JVJj z-LjDW#DH^7V_ojIH@7$1|6lt}eD_nOODe7Wp!Y%Wb4L^jGaUSiwYUBF=lAx^?a}mo z_}6}E6dkxIHH~Srf6vT1tOyK&`#$#4av0J-`XsEth$PULVO;r}gWCX_?ECO7;yuu* z*N}UKhmPeg&-bVA<$4dYLydWFB||4z1z4AMoxAW*4+e|otXn7L18gz3-y!547+g2f z_P&4aKgacsa>Ht=FI1oi$tZmkIR96#HRP$TBk)a(`sxg`I=TPazg|9x;Ew_VFm>AY7%f~KUA4Mb8lK82{V1>t3O=~gU*ktbpOLl$g4B-GkQgW ziF;P-#Xb}G-A;i**|W+eSguZqnDv6rdC&Oq9f7Owaq(zDyvl<0x!oL`E3mh)f?U7b z0J9->ZP0EhZ0$BLKlgby*Y{=?Z8jS9)xn(M{fjS&-G&a2aV?FIDqu7g)eCPV>b3L- z&EDPmu^1?hmUq@|miXs*_uc+BRb9RZ8|HgGE}h>2?I*s=ut+J^+|Tnwcr%RG^aen{~gQKz3yJ#RXI@ zI~dz4vz(XFofvh`49pn>^Csd?jQx&VGWAHzSq4*WX#u8(7cAhXc%r`$oej z#Jl#M`fuxl^S*(RakJanxBoc~%qv210gEfKz4hgp*)JqXd_FtGQaa}VqjLvUyeW^X zJ=(54kCybG0mhZD#yCk(y~D5HMWWDWnbzMf%4fK=IO6rgN)UAb63|B{F41g7f?0c>p+U~s>$N8jWDpW z)HD0k-*xs*Z`9?~Zfv}&kty6y@U7RL63FcvmPCeU!rt2_I5-}damciK+AV{%wPqj0 z?r`KXNkB4u$kC3_JBJ$sc0S`@36%I)kfZjE;0*M}w+pV>{(;m; zW#Q-SBY*8nzpA5qxc%iZ_*y|X@gxk&|NOm*;)cU|Kb(c{zrb30>Y3;#g!~?P|91Vy zm-mp>tfZe8x&-RieU@1Fir}AX;)_e?t(iyW(G-6%+gBJee%mfINan@~&769wJAF3` zS@w&qOlZHLm-2dZ&++HLsP;@eu_Onm<_7fmuD|=+kpLU)mlZn9njNn1LhknxZdCKs<5qOUI4aB z`?XW{#o~~I)81eEeque4pS_h~16wO%hGCixEy5pHt#S^5pf#@qBk`wRN{+F74}Y$lyROQofeq~FOAa!; zB!0>A*~J2cJyTP=L`}UX0J?cjsqg=F7nxfUm-IGJNW8&+{w)3C2(6~NJsCYjTytvm zwwG_kw?K2@%kqcDqd@E4aP|C70TRFO!cTibpF&Nb+1)w_fa%xOvi`|q&{k$Kz4o37 zw(nfJMyF^gOxfHYnwI*Cll)&gQ%${b)O1lzNCii~1X{zNv1GjyFuK~}-q#&dWd23M zjUgr+y_CeurXPC!fO9^P`1|ym_=eYDeCa)=?ekQiJZuJY@vnaH$?7MKBUd3y)g zA8+99oxUBD1V16I@z*Rz_Xx6(m~3`&%U+f zN7^s8^>VdaG3#{3-SY%qZBX-Lk7|h+_C~a<>6IqrE3MnAYgeQwNnlPA)#lF+EH%kp zi5s|Zu0R!ep|o9r$lDw|d-=!(!%AfG<(7K-68@D=cAVb(-njq$`IDGA*Y>qGrn(3` za4EPBW%nyzZIdh}3q)+SlX^lS@!c9a>88bij!&{{QbwYHu0JB-M5=amlj=b*H=LA`(7YYZMed9SOk9cqnaMqYd(>Pretn9*H+ z67|D#ulZ+-%>)8pm$@9I~) z2G;c-<6~lHpvKeFnJPxqlPjei(*B|y32m223Socu`)`-p&#!Kn0ppC5jK$Auz`AC$ z-O`NX7Y1g6p}zas5n#OZN?LYA(&{-cG^IXUc+DNC%@2=>9 zz8k#y?}Ipe%SgNXW+paW9=i?%ti04s%v-6^Dv1NX#c^D5@s7b2C$V*TKIJ;g1G{ZY zcf|@5b~n{BccpFwk-xC1ou_ zAEX9UMl9a&80aM}_If%*+@3k#cDsPN3W>+>?&u?*5Sh6V{YaW>W8IjNH4i4?f4^ytDm&Xfr}WE+;tdYzEHP)Q{&)-X!;_tXFl~UeC!Jd-9@S8#e z?{$S;@nxH~ZU1$idYxZyHoxQt6?gmfcaG73x*oz5wq5|5itJIN6Yj-{vnJA zXU*7}h5>!;YP)uof7=Vpi-s=hyj(aZV5;V1_at_3%@501XWg08E3oy_+oS8Pim^I* zlldjDa7--DJzjVI5Apo%1zJh7@*!X8(82Zk+%S;Q^(Ydq-BLw*f1cPsdgdPytB^D<#!0#4Gfv{{{Dq4u(tPNOx?~B=zcR_nao@URWHtc;XS=pdwv3I<4o-T-K0mPa+I9W5Z3fuBc~5ZdSSQflkS9yUs&TYv>*CKdZou;UCJ|r1 z2bqP6WTJ0R~~BP4!C?Qi~O}S=R2ILJIU2Q#&?-Sa`ChO zHh-M^DMMn1Li-Q@ukRoCbQ0?wzXLzD9IjNj(MkNJZd*is3BPiCbJt!GaRj=cpJ!ax zGhhZi=?ah{o^Rc*r88*4m0!Xc_t@dr4Bm%xVIunBcJ`b)Ec>{sRrwN;FH5)X{d_RQ z9!Klu33j{~@@c?!SJ=P*c{2H>yPa0QRr3)vGDxK&SArZZkLtRgPl#Z0P?(eWXKj+Akbju(#c?$4847di*a$`Q~>(oAU)- zLpvg_<^92szQ4Z%I-_2#eb-0igN_IpksHqN0OOp7Ua#@R_pemM0{XcOm0VX~ zyx#A1dZ|06-yK}hQhyM7J@)I^O=JO!w)%@(i3W_EczE#Eq8JRG4p8kYe~LYOvRe+= z@LwX^kVbB=kBuPPZwu#I@bG74>y+=0Zxzdupk?O{O;?GGu zyZt>(7({*wqo^fd$~Fs|tz&e2c1uE+fza}Y!cXx{zPqN=CXQW-)ab7}-g|KQX)$){ zv(kwiPF^Fh(hs=_z08Fi(_m-u#Zn|b(|N)+;{<-ZZO_by3)Tye2Vu zM@8wEWbCw;c)Eag8ynW>8`$LT<{EdV;yvjHIdSNIt?5sYH{m~Sm1z-*O1Vw6yYIZ^ zk3tj9|2^5uf5dM60l9DIeS1#&w|>(ocsF#P(y{-xK4APet+eim2iJU1xSbTBYe}->)j$`>{nC8I?IKY}hMMg)lP2H@6>Y1wxd{__tR0zt-yPpSgu0BS zZ+Z8BgJDZbv&Wh=9AMH8t`+C5FTlQ&?Cd#E2y~BSA)dHkZ=EqhIY-b8b#|U!S z{lxn7J@w%F6Aq3XyB^MsXl@;XLCM2QA8lI;H1ivl54;{>&DbFy{`bdW#5s$`zmKba zAyQSVzn7ijk~cUa61bp8)(85;3ObdAIQtHNt{1kyC&`85MDJYVfQJLy(DP7Ga?6K# zNc;6={6a2K-(kRNmE0RKqTYCd^Lov%)x(w{FML&PD+oPxy9|Ky54aP_-- z*Oo}>jQbGjs#>LX;}6&Mu02bZH=HH#?<4XFSMN*Ep*b#RMq`Z*&>eR)l{(8|tw+!S z+mveX7g=k1LX@kXHguUCjBWi7e)mSY)RfMrFYxaD-q3@V?hshw@8NNa%iaR&ug}}8 zZvO_az-`Gp1*L)JaC=VWQlbuN{m}7m?}7;4%~9r?>a8J?SSdW+{$Q34W}UmtYPva% zt^WJ@huPfqA~JkDz5c9Pf<4cd$A)KkVc$8i_D4T2VMIi!jhmYz49>NW9ymMD1$7;~_&=lXJ;uy0s3b5KaqQxZB; zf3(>0{QwII{U!;uXILJ0JL00of9P3TH2P+**hAtExT!u&lfbAL_D-3Z^DuJ1>LI;a z4(QtM^tEfa@P3@y!kybx3PnKER2Z1F=WQ3D?JGQlA+~|d>e0Z=fg5{ zRUR5_)`OfQy&-g7379AnjTE8p#tPmCIZ3HT*!gzj=A$K^&=a2#lFSn>_L3?+aa?OrT7=OEBQvIkHc9z8q@_*yP zPjbeyd;1mnO`_%4E?B(d*&|NgHi>`hiszQw_W(F;ggMs|E7cYxN$|!@zzjqER3f3;o~f*YhnBMdn-eFEgvmV5(H%3Hv%ne#7si z6;tmOIOB`)D+|vowB>=mc}92M@8a4|#Cb0tZSl@T4ZBml5A)2?CGlTgbK6Qtl<0TT ztb!l*+;XlnT;5*0=p30}*D0%e!C(2*AnCqvHem-dqRn1iP$PIXBXk-*Q%85e;8?<+ z&MZ!T57eBbF8?~|faIT@LsE84$Q)U{)riLiS<3@1|CHB(-fKlWz`Yz8Q;vHUTyG=v z#`i~N6sSb~a(5eDcbjQW-VqEozp{8DmkrT{&2yc*{{wfpPBoctq8xh8ZjU+PMBqZ& z!UG@I9k;~#f>KsMCFdN*NR3NwJNA8ooG{rvJ_3E%@{AR?cAVf79CF(8)wZ9o_qx&* zs~D&Cp}*>;{Z!K-D05UbZ`Jx|{v5O^-`sX>YXdYqdD4Ap7n{WQ!rM4R|G(PtoaZOC zUB%QI9zgHS0WWJOPTnR`toC1gCe0ZyObk8rRNs)m5f4`Nh`jnb3|)mkFCRI?{8#%V z!N<84@|J4Aei!;dB`+M<&ZUdgFKyuJ2XpOb8*}Y0U>`X%fBT~u=-MW~dTOUMuomr} zegDsMNG&(_+%QD&Fwc-wA4+v{aQ-BoIo>1t;DBBPsRIZY*|( zX6E+Dh4n^%?Y;R%=5j>7b@&D8(Ga1hFmog3Y~wW&Fa|PcSxynqN?sfqC3*qb_d0H_ z7TEzTy8B!f&o1b^HdtbPx*IE9r}oeAIl#b4QIQ=^$2tAzNos$%SttwJr3|;en#Bf2 z#yFTaJnwy;3Tk%n|keU6Z$jToBI2twFzYG(=#!4euII074prwt~lg(cX8Pn zB7bNow)gFtoqU}Cr|+%MyO_$ApTad?v}hOQ#IVmmeK*iruvQ*wqe~wh*qx7A*`(dO z4#Z$*lwH8RcMqYf+9RB{R2eJo9h5bG1~4igMxpw3VEb%==Z13Xz>?F{by2)Z?AsR+ zE?zw!u}RZEHG9c_)Hz)5vg%jMH)LF`*jj&C99SPh<$_*s!R%8HhvrETymxG`wX9Nq zg2%7(-oGDCtWDdZ zk3gSUj($jl7{_J%b3e zlfR;uK7C8b4`$F7w?#y~+>lxIA`PpyBmJ_I{cFv6*s*NgbMmd**oJD}`rWs&BeYqm zAdfijT5T5WogZM#^_`QZl*{#QL|%Spoclpdr+(;GvD_N4oeC^*IfJjM9KRDdd($nJ zt>onOkOX&5-wN}5%GK@|e58EM(!dsW4^^dzOyWO6Jrfzo!Np)3yEs~{N)2nJXP;k7 z34lM&_vh{X(+3$cPb^*(g+cB+^|Z1$4`6th=RWt1fq2s%sV$p0_yX+7bzU^KiKweN zn6+tUxsp1tU7w3NUn1jx`tB!>i$b~j_3du?wNA5t%5$#!>CV<&@-BaX9hUcZGC zPlI~;`+NHomAS|HHE--x2_g3ewx{`9#1V48qHojJmDh%Wb+LgxB_#l?z~Ya#_y5CA z8-4#G`@&PCe=w^$5)q175f^MOo+tEB<~QM<3n@C#)7c(iB0UJ~qr8tO8{>ds;p=ZN zmk3>|uP?0_I*P5qFNM~zZvwsN?S`V`QNTQ~*ilU_{{P($nYXsgefbq}U?z;yq~>Zs zN9y73)fF+=HtGLFc$^ED3(T(t`lAll;D^)R%3F6;Va)!(`>2nHVC3P)WzX{zU@%fz zEJ=sU{{yVWu2aR|7s7aq_kikljy%Mv1DiW-{0gytnZt)QwuJr1G!QA;INuk`jtW-{ z4=6#){;8FuvH9TDy{+A!q zS$PiE0;3_!`-e>XvT>w*TIaCdj4Q6j)nCfv29|GN4F(S!ypu4OP3F63GgqsA7yenK zZE{p{5r(c`@*_fC7-p&u6?a!@5YH71)9kkOVW<1~lg>h1{4m$ zP$6&-6?#{%24~7Zd!~T+mT&>c==S@wA^Un0Ytqc4VIEQZ4{v3XX z=BqB6PRCnwZ)DvabXc-%h+rwD*zS?JQZde!KcgSNI!AsXlQm5Y! zyp5$-{)qb+yn~^Jvmei$)xzNCnMb3ZzlH7yla7zi2;2!vqF3SbNe;dPskZw|kJ=LR zo#wQ_F=4SkQWo`Ys&k!!NwRAyqNuF#n~G!2jKfv(F+Ymh&pe~`q6WyFIm7?+9lL@w+MREmXnU?cw^>@zSRQs zFrWnH#_qHv+Tt~e~Ie@UiT4XLY{JMP=m0sUZC&)Al&B;K9s#fwJe;5S_> zOo(?ju=s5Ta&6aR^Sw=LZyLpL>znaS{Au&zx#lIa=Zt~2&LybPJ*~4sLJzag*-yCC zAA~L~z4tWF5oo=CeLeJW{GZr>n!ZEqUM!jZwkTIG16i*OzT7)ehBYW({xzVSkmHN& z7u5g!YY)`)UYqZzz||jC&LfdBldoL;scSJxZWXWt#u}ev2l|Ym-D%nLIrqfL!t*5^ z9amZ6RATBWvT`c+U0U8Is-2HR!`~K`%M$h5+SHE>NCK{^7OF zOL>4^GJkJmGv~avW{p!85H;rzV@C39OgeU>8$tf>4{UegYlOV2gDTH6h6A}0KgMVv4; zmr<#GiNN2D3_XkA<4M$g$u9p8Th&p)^**ZJ{E_Sp`oOGNEha0!nZy^_vgfR04GbLp zwLy5a5R!Y39~_;(0mkX2cGgY*fh(80rasS34rr6DttS&W?E>w@q^NX1!RyvKbEU(- zObSL_j_;*}9f6^Jkq%3nRA3}3t3>*?Hn0-;jX-lYasQfPJLB-de~!VXn@OCnFY zd3U#T!RU7U6c;m@uOSR=rB274-*N0s>@2Ymx|YSU%P~b&Mz!Gk7Z~w5R-1o;F- zxRkgq=@PAHaU|YN68`$b^1zH)71f^0<>$s2ic(6M_$F+=YP;R6kMPS5w}!v@_;4Qd zs;;z<8UkS3&$nUt68oFVr^Pr`{&#(pp4u3-P8xF8WqdGLaT41mzVAD%L*P)$pEBMC zD>lKn%?9>XuSO`fRl8cdauCxKjDDQiejnR67X4uNa`NUNX0p+2bVDsr(teSH60QSV zZnW@{?m}#JSs)S!ok*Jq_R0VJANE-1?G3NbdgIWM%`u5Pf`Fm)^6KMb1g{&9^%ZB= zvI-cz{Ul`iQWeG;&he%%&v4gL7tP&(sL_-mXNx!=M>@JA9=UbEZB;Z zj{r5Yx6<@CbM!ozDo_aB!Cww-4H~kmPEGx{c|zvEb86Yj%h=8G5-vR9 ziAm16d-MzMLTh%-m0t>$*rnyM&u-HU4hQXPOyZrw36TV=`7ut|k+>m6%ySn2e3ZkHc4*aV;Oi{Q$K2KEQu0h!3@ zzxy(s>1W9C^Fw{zveW8AdHBbz*496YoA;zj%Smw2f*LG+e`UFdfhe+PvOZKMq#||c z)T4L(GdNznqFL_eZ?Mc-d45iaANJcElUch?5hDsn)^VS{VE)-`RZ=hEkM5E*PW(0g z9pkR#cWqpK5x>`N3t9Xaq50IFuQIV|*l}W6`iqDEYkB*>o7QhgJ0L{`oxBea7Eno)XT4g(p$gFF^LenBlxXZAh<4d2L|(5=vD!PM^_N z#MT>+g4AQPkfyiy^wnz@kUaP?cca{0EPA$1zb1x)Ow;zkxnA9vWz1HgA0%*MtdXrP zJC|O8j=7A&owviFwWYJg*M2T^xLna@-WB>~G@mN)!2}MkdhS%}c7flZChB zZ>7HLDMfm1yXp;-0i0Yk5?Vl=k3HLkv2Y&|&l$hze4zZ10rpFNj<~Ts4ZcY&vRSBW zg&oW*^NJFjffcE0q!e=u$9ELI)mU;Jy9Hh`Oao6NS-#f!_sc6-SK%+4qA?3uzBdjh zRr=s?YD}t!JR3SoG`-$lIs`NFm1iWcaN-;!0loUq+xO4Li8s#aPBL`tb}|1JC^QQ@ zGXFfxxHAspu}icqsxM&o&7yAs2E$~&3yTwv)f_}-m-&&AZ+1`xKYt&e&4XR1s-E~+ z-zV`p8J!Tjxei%#e!VM{dkn)ohfnRapd!QLh^A6OG{$drUj8m<5zu~0++VZD4eB4v zU3j{<0T}C!S~|$f!%$~La;LY`U-^B@^4CYMybH|CDR2xiiGhr}k7frb*aB@&hWtTq zf=~X>=$36xKM4CPBij27uR$I5H@LfR5!nkZ-!>deNzg!s{ZGxUf9Iur$JNW}{R9sb zW8HbBRig}?vaNbD)xHuL+kY+ns^S3*8LuNZ+lc(`R_$;S&&ek2J1g)wn!Ip0lwOnC`nPmr{~}>9LqsD!3Tm-^hFUq&9|FuZ;Wpr*wUZ{a-{sC$RMz zrf*s%cJ;spr1AD_l2nVpYNZnj5kGHZe~+2{u|+`;q-&?8B%6%gCbH=(E~jJfg&(UQ z>ANCpx}d+b%-Z4u1Re;$%qa2%QC=%sKtfsn6km%Gz~x5zx>i`RJnE`@rE z@G&nX0)N#}TN(D<{S30BJM zvTA5sp=WD-#2Y8x^Raq7tuSNxvT0X44IJ)y?tS}38nVRr%fnBGWB=y$i37y~*gJ1C zMKtO*w0})885sDCOjT8zmWPH|*=V!kwn8{0-^Lu%2u@xjPTc-NIgtEU?`fmNhKrOV zW6iDDM=~$4U+D4s$2D6qd-j1m>*2!?ba+eB#%BahI%f-gN1Tukj`}9Be3_a5l3ygA zC)PQie!s>^)1zZs7I5MwBwooeZRw|tFtLiEVlvFVFJc~O9)2-T^MC#h_34MNzF+X3 z>-V6=$~S&q@%jI|{2uAq(s1Mj!Jpe!^D!aPgdcN0^zLtax(1tq3nNn(>0$4+*W$Y} zpF-A!Q#Z!~O_04TE5LQ01~I>V^QP~L4H!X9H8zP%pDiG5yvKBC`(;O8G%d{(qVlS`k0mT3dg+6XH8 z`aY^1Tvd%DJgT1u=Mnx;HjA=!SJzw|`#mrc_4PmU)jr%k7=By(pWjjKZ;t`1s|*`CR@eWK~3-W{R8P;G3=LtG3J_+j`oi&|@<; zC+s{<_aX3cGmC=P2)x(9F>&FB=96^n-SX~LedRa=xL-dJxS54fcO*3=Uz%XMbiKgh z%rn4FFKesk> zL*^G&-dOcoaBuu>tn}0ndp~Y`Jx`3lpZ4W3OhZkBq3ZerSI&9FL?W?HNc&UW54sYda(# zl92JOS;)c18<~DPL_Rw=kpx9V*Jd~KVgFV6u=P2S*m8N_VwvYdINH@LW%P9&)@9rO93qQV31AAd9Zrn$Uxn`t5`uw zl2|9#h}EBu*LzYIAUW*Ggw&O_*mIu0b+tmemwAFF*;1|q&}fxcCeq8@>zIC=g3 z@ptz_addfSqa0Zpi^V$%i$e0T@7u=AO+7KlIQr`R%Z3M-{ygdPQZph?npSk@)@SJh z_{n?doST9xQjWx?UWu;4rzCg7E0=r70z;}_WDh*TzM~7P^j>_&Y_XH~-P(4+^_ocS zg@YsA|%PBVF z6a_NR?RCd9EcX1hYg*C3*#4;ovf{CqFDwK~xi{6z!cg z0?Jr+uBy8B4kv#UM;jF7EylDlEN{u8_c~nod@PvI_(h4y0#Pp}rQg>)l3{;^h#8kcV=y9&+9@QV^rz37+)ABK2rMW!FK*Q2KrwowsZb8QN&0OC_=CO#%N?r3$ zd_Inr+{xm5nM@Ya%`KHFY=wl&2U7%#i2AIRuTQ0Y(hWy8)$F}{U=ZJF- z>f3{vPh<6%D{jAEBa4E?X|Y~6gngiU$uM2wN)}nbK!TJ!=OPShM{8ur&S0}N$zpvp zKQf=|Sh4qsVvm>UmBtMXWFeQvFO(kyzfAk`yK!=oKcKVScGJc5vt(iEw~zEh!^i@U zRF0)7FT<>c=li9l6XB2Im4JrdhcSbcIV8WQ7MZfF$Tka&f6jN!^Sd({^8&FV(Rh#H z0)ApWn+2Lw)xJgMVOzC5rd)aqnMb!i+BuR29RkH6rVIHY*6i3a!}(W$7XS5%T$&b1 z;O&Uw)1CXE!$;}p^3H$qE_3yVmN9$Z%gvc2q4_V$K0Wz_Ba#RGqPkZT_Uw}*$8%U5 zJOi0Gu{KLaKAE|E4Buj9I`b^~^ba|SZ z-`T7Oef^J=SC(-6_sH7P`R1wA7od6Vpli+L*o$NVP3bMxp9RQ5+V^`Mb`$)^!*WV= zw*e`x{!WPm_&$jt^y8t5xxYmn<&gd1tiZL?-B7oiyh>t!6n0e7bZ=f;hNFUtX;(i) zAu}@PPJ{(aKO_rT>!tO#LvSV;O6es#0`eqo6lvTb0XD<d8kID$I z#HtG~ck59(`Kcs9ji0}LOgMR@z`i|?Vb6MuT?*{CcBR;m~2OjI@>yGA)imIPhxYHx|PR z=yDUhOwWHD=SSk% zZ)YF3ryXK)+}E`ndxRa`hLj$XJwzKzebUZp)oE5`Waq=}V?=WZ|+nQ);0H@trPj zX1|)yA=d4~hzP-LA|ye-@4wBmq@dxa_-L}_9$;F%7(SXdjYILa(;tr(AT8;qn4TKJ z|K4cL7Y_S?%d}twd z*Iu}>RbBu`<(Cck914S}f`hp&n>!)HGMiMejt&%+^76hGPCSFeXYUZbbnFfE%}bIQ zzorI*GK&M_Yxg02_pOq>M~sO56TZ@9;VsVjkJDvdKEZFxv75(Xz~>1SJ5t_UObMwZ z^X$n`R9107Ms(m&_TpJMem-wmt%MuW*DgIezN;4V)~?gK9eT#4JUs7w0I6v=|j!+2UC%Dd-s}Kln@+M7}t0E`V1L7>O053VIPzx+Sb1Z%)pK1%Ll2;_8Vod^)NX3Lc@P!bka=zkD~54~~u;uMN)VosBd-vw|48ibYNWCw?!)T^dPAS*$=nncz5iFj`weBj>-~sRp|PU9_DWOC;pl1y>vMFGTgwJ#OJhi zq4ZHMUKdRHeC%|kkU7?CS4-{sV1sSXr1Wfx*W~oYAZE7>OT3pYRdknS<+*IdN-m zgU38*bp66EakK~e#*3u$631bL*C*O&$_W|kDXzikC73@Px^L)i6Hsj%o;`A|g&w(O zA5W;yK;YUBNYfd?0%>*0$G!g*_YMONEt;+?8ae;Bty%Oq?Fzv+JZ?2yIQ+GmYy3Mz zCW0eZp9I!ce$m@s{@UGt##YQ@P3Zvw(q5oMfOjn$I0RFB>#<3c?R~mvCZdA zNR#=5DkkDfPGa}2-Pid~6e3HXWO&Jth#U1PM2GHp{SZ3()lOZ@2|`BdlFJ|U`(gT| zpxlDwyU@Nc+nlzSOWqLYXVKO7@(Y!?e&-c)lfLZ4#nZ+$UcH&xzT38QfRRm47AHd!HXU@S}Mn zPL{1;Z{6mEk0ZN}d@b$T2N@PMTWntv^>Ig5 zv4%%@qM!`g6;8YjVo6oczsRiP>i3vnp%3YCDE9H~?3S2)4j4PT zG$wbBWA&V;${V`Xv2lrIz2>cC9NAGUUon{owAI3t&i+$KO&1Dk;*}@y$8QSxRL-&2 zanR-G>j(E1;`H{;i6!g(uz#?3F8_;Y9KB}S_o>JLM-J(izhOt=v4t^D^@>oc&tz$aQAcQ z5#%fSX|@rERo@7Y_ie*TTmJ61m0mDfsI@io@>BRO*+=r&Tn~&XoAU69r`T(20@_O+ z;3ynB^=MngrM!}cnJ&i1!+0Qz=Pr~G%@9T80 ztn@$Wgv`C(Kfj7_@n3Q7YtLW!Dkz79jGbR&K2;sZ_Oc}R>V*Vuu9=J&hlc_Sz1GMd zo8jPpktSzo(-ivv`yLveDXSL6uH(5P@vAs_14v(LzT^GEjZk)d*IAx~7U)h^JCSkj z-{aeAKc`{h%m|LzPQJ9zWI+CheO6uW`8ZX7`T3gEJS=l_TW>q-pXV`lkPqzN@mKJ{!nmC0_hk z_5o;1PF@XkD}TbEs1{&S`ltEI2F2%dX@oe|b!5+U_uo++o$&WA-nq3o1eN~tCD z__Ui{mYR=Q55g2=j8@|GZ}*VclnfkN?(^L2Y&}v}j!djDw!_LR-MKHNZo`Pi)}Ez* zzkk}_`<{5WBeE^`KGEBgi}k^S%i^sbW2ISr`Rc_yWI;+~zlHT;=)Sh+=*`9+==*4S z?=kMe9`zmbyvQ8hE1)x;zkBWU1x9`wox@})9H7+lAK71lm6I;j5_F3xp0WfhI=Sz-F-?=-o~$JM@^lj$H4%yr(<{IZ(gqywL7|E0B`% zaMArmH?a=BA9=!%!RbFK$1aP0gKX~&d;chV;mCWnp-s+&zS*C-V(P274p#1XbLGJ@ z3vBi6Q)_aagI!9;>t-i&_)##{-E{hmCI|lkENS~^g%1c`yP;W859X?_MtZWTTjw=0 zjvoIY{&`m+L^#9*d^n~E%y?Io>9o^-<$K5*hZt4xz2o3j*zyHOw|Pu{J!!?guJ#@E zAL?`jww8QB)Twr4f6ly7T&E1Z;;vZ&SHA+IpuoScSQ9DNc3fU@ClT2li;ow+afH^x z$DX~}e;8|i%(dt>^n{Azmox+3T0ynMb3UV4M7))Go0hXxFb)6AN|@uCya~Gt4UFH; zx4^FQj@L&#oT`qLO9x*cWCyXViNya!N&_;T9JiBpT=obgDH7(DKrg=G_x|n!%=+}|aq4e6G7iTsxVrcX_8mB;qoOT` zGn(fm1tOw=sdcQf`@(H(9>1FXM(G1~?4u2&yDE_dCu<}g8f0TTg}UE*LK<7mR91GF zc0%5*rA;AS`ashP(tqqS3e@{5mtj)v2@_!9pu zXPm-{9WN9@>$l@E!)DdkZb>g(eMb_4t}FFFa&?4>jANcLO6kx()^obEm7C`y_;!`u z{47F$W`8%of3KRr!I5YEk=jts#nT9liHmlKpDcl1ir9HG2?C#yq1QWL>@tq%9OHt=UNO(csv;q@*FEX^y_@LMB(oz zTh=>2YsO^ljEfexGjLSMZOhp=&X{MNGt2M`6WT<6RP!qx$3VNyDe9y+EFW(DaOz7I z(%kZnR=jeDKJUTp-lm(eYrT=9%&;d^3gunAAKQv;oy^@Oxm7q;8aWjpA_%=cDxped zAy|LIAf>Hs4Rl*NDoBX>W0!#^?5JD_6#?s(&zKY;)27SV*fkzT_`GFO%;HEqaW6Jf zT=wJGR+p~BK2F%YZr-0c!6DE}Dyew=TN(_k{)jUDyI>^T-E_kNHZb1MR_@v$g!Jk7 z!?A+#K)>LxJCD5N@43i4|7(gXKLM%=lFzgl*@4@0+3{c6p}^80DOxJ&bKOT5=O3_p zN$@2`Q-pldd~R{w*EVWt`^)1Z{DM%sP#ccF9!7IyE{=-~;iw7YP1P_NX70RI{HeSe z%FY`qT|797c~>6ake77?#y7o?Ps#R3o7if9WFD0)nCmkC*n@LW<1J_NYi}8{Z_i%+ zB+MTZNSS&~)EJ;qj}>gaE`y~_Jeq>@%OTBi;jGxWxmc=|&Irnihv|iy!EVdmMVJ{=mzgqZg6*Eq>lzpsxuH7sQipQ8|7&pyu7(VV%z5<-}&|qpB?h6lCfI zN491BfKN?rddi6$y%LMXFRQiu;ncgp&bJ@uM$V5!>gJD`TId0+-5E#hSffbEn+aWV z+5$VjncAj^df>Ht-;0c`|7(2`Y`!0q^Yd6Vmdbp;x2W%*`6gsOi?B^TrJOh((n=RD zP>GjEn}=u3XU*Zt3q^Kz-YX?VBL8f3>&sXkOD;Q)EOh4kw~vRf5bgS`@8ZQy=vrNJ zRh^xJj0IM&h8v54EqOkqw}Vq}3;J6>9~rpHv9ln{S8!L;D<`Bpk6@IR9)uD&5|QMw z2}i$tdz9Ll0rZ=ijaOpVAiFdwMcFo$#PhmzX#36ulW0sF(2NOp z6NX9!Cn(Kxg>jf|u=W3mb3GBM;8t}G`nQ>4|zeX@D%zl|T)caMZ7 z7OWH`>e=>04Ojjit-;axPmHo{h`MVNYwp5bj|8sgJEd45DOY1qnABS z`jKBi{?k0$NAIPvl(M{MDsd^M40!hVtR#4d#-ax$5>q+*0qWP8mU@M3#?O@CJ!}2q zuu9Z0_pFiu4(jJkM+gLC`=7YrMx$cPy>+N$d6e5D#%_1Yyu-A90`J7@l_ z3$c)9XXSarj-Nd1ptr=`d702PNFH6dA_>dp*uNQCPQ>3RS0%hmjwd7I$p@-q_+LGe zF7VZ@JQgQYJ65|C5jdjv??Z}BUJ7GN8=b0oKNo3BuBfVfzKYBxwx-W|rlA~M%@~pk zVR(&&lh$1u9A41QezkQij;bXZkE@=<=}?Pz(?v0maQyCOx&l$xk~ZLP^Ly7QNpPhk ze^X{Nw23!t`?PO1X50^3Ypit>3xA$-ENGTPru|ZpMImCyTK$P{RH7akTdgnt+@TLF zZCQ)vkK2IycFPU^>M-oMp+4yDaU1)2cWSKo*@`{hJzH9$^^o=`y6=hDZX6qrvtEAW z9FFBLHCfH;j_;8Yw&=$Z2Odd4>JhJuNKJ(cyb!yBuFRJXo60H4IF#EVhfb z1CB0pTYkKgJC9?O|6cmHWBakzDz2k^@phcDq^EyT;p8vk@4g4gSI!3@!%Jg(<9bWT z(cJjyRmXMg{b`pIv_u#ynOeV-PALKHOuTWN%57x$TD(x)O7H+Cp2^kb)mly#_^@ZY zne!oF?2=|HldKlj?|||1HeIC(Rbo}R z>$}B;$DmfG?V;~1#{cX;G?#CkHsFguT9RM21FsWwrdU&Y=(f<>F=ssL$6lle{#fvF z7nk3OtDO@5-$(dIYPo)oFTI-m+hGcrw68h)WX@r_)ZUBDZ%%WKW1DMi^&;=TcH~L> zxVi;R*su1!KOy-#49iRXTzSo{7!%M zMBODb?}X7c-||D=0oGr;@I#8-xi%OrFP}W?C5vtHYgf%gaN;e%jNLM2d^CwLMly(xd(J`+5s%)Ge!-D|f2gWi?0 z)tcb@AF(y^NE_ad6%S`@5_5?9;?z#2_bG=Vu#e7oqbzL%%;i6$b|u_}eg(niO~xhY z-yNu@o3)*o*BfH5dQ~i7z;E-Bx&i`+)9=4ZO-*P6GOB)ky=_gzuUik+9&wm#gZykc zuaEn9pohhOpD+3f*Se}d_IU4yEpMSM@jUr-F$cGW&By)bUYjcctWVn~^sjQ+`_R1e z*}BcOm zB699?^T*l1n&-vhn=A*KRK<*4_IDUur>pePkOqu=Q0WU4_1#2b0f+5p`X5wwVjR6D)`$0K~m=vRANpho1!jI8Q8L4RollLy(AYtM7li2!6-1y1iPgLv!a zpk)slu*>ed!2Z4;n0~9tzW*+VcL0Y*x68|4UI>#HW4=wiZu{3d!n`AoZy&Z2hE8FX z&SBR)puVXJwOmTXi(Ahx+{Ki92c60Kk9N+8V~T|L?9^r&u#Pnk>qrv&k*@RJ=gLP8 zUj>O@`@#N(x0Tp*@mBGM@GD6Eq&C=H>hWLKHBa_{Ipr#+T`cr#y5!=@gC6xOUa5t6 zllbn&y<0QTfcYw(-Y+ikL8+NT^P27Y*z+KS8vVWxD6+GEeWu<4#(RganeFc|B#IVk zbd1aIjkW84ho5w?1qMXR23K;$U6IuoYybOO4)!!hi<-Y7csOXQuY4J6Tmi#MAvpyO zo1t?#rT+6Sg7>mRd%4nO-v2e9J+TgH2lkzNiP+xgl)c{1o0vxnvCS20U;MQ{B5gz) z+>sVH>bcwE2o8C$5Aenv1)7uP;hJ-td@yLv*VD+{G6Zz#P4z$4i2{8c!(;YJA3zCl z&j&i({2(3Q)|W)@{f0rAe5P+oIe7#y=%w*t#~g3$*zX)>GN%nD$BuT2lrMzV&w2x| zO(UVqobSc!+#z7iwrIL8P53o?SNLVV=;io@fEub>9N{yH&0!&dc|IAKTKA%U(>)#T z?>~`W9vyTXlcIQX^+MP%lkjPM$Lvqo_UK;DgZFQs{=3&w)8LE1x;$;PHhw#FZ8dN) zUhhNV+j_u9LGdw?jTB1#h%)6S+n3IwbqOJNN2&J7lamzCzREGxQpNpxs^W35;&*?b*Ra(D>5&&sfzYd~&HWobFM7@4t%bHmMJw}t+i+;x(JudM1Royz-OfSb zP4|Fx$MMpe8cFQ>D1YPKQCnzZ;qI80hcFW0`q{LEgSWsAlcN{tRoc+TNK0OGxEiKQ zMK3?=ZHJ+mLmw9Z=t1Vt$tz;E99$vv>=fF)Z~b9tRpJ@{d~ZJ#)a`9ZE++g5^j9;* zZ@+Tk--z|@;6~}*sQ}~I+c(XcdmHF>v@Of8M{>_!Lw>riO&~BD$i7j&WTd*dJi;DA z-(YV(b+dncAPk#V)PL0Q!p_0G6_p#VK`Z&*37rCE40k&azwh!WrdUo2ijZZnmtQH3 z?x>86udYYmRG30XL8#`V9>O0zyTCC|AH*suTmavE;N*5u#YXSQnrb41qP zBvt@g580~eKZu9Xz=peZpJTbkfl;~a;`fuL*zvY0`C>bvXAjQ~*8ij)NaCgSO#Zv z>qwdSwLfsEl=!_mUUIM74J=SlewNy(0d)0xspp@xxSp>u+b40M#|?S{my{S#!m&|) zzQAD(J@AW@DqC@%tBwM;x|rQzy4672n)+w!+im{mbu! z&zfApITvtbNBZZLx~mEM<>L8G?_y&xQ+N6CcVQ=yqI^u?gCv1-puELL7`@xG@rFPZuq~=rKbg&g zIlq-VP8m_4eX=EFww(bmXQ}8nK99nZ>LuAFSN|=)TZ_K5?)pUJKQKC`{fjLwK>UVP z*^f$>VB1z-qt+fGe%t!?$Ey^w0njwulldT?#bQr`PmXSZKS z_%%^)Wcp~aM8%(c=+;7~F=bdl_vh4eP>TTTm7o)n*-DVR} zPoz7rfd1N;JFkM76zV_AmB1IIJ#Za||n2nPoxa`Mlv*OF8}%65p!v!Xu(p zSd;xz`RCp;XniFy?_whrMi%HAbT|_0ulM(haUs|H7^zKJEBZD7ShCd{_st{nl(Qlj z1J--$fqs2|?!;6fiQigI`gszEj{wS?|IAuI;>s^0^Ygn|6@UMVjr(~MAMYn{QWdR! zk~RvDf$rSzde!bR)SWNR9N8uS`L}%?_f;(?`j!7camxq=#%k3Ue$Xd)(le@@C8^`r zvF&8zwSrFspJ+T;BmGqnlf-M4e&+PC7#Mz`IoM~W%e6kJXWk~6^Aq*a^Uvs>jg*&z z&Xp%qFRvE+EB`grp6(>-Z**&slnQ2_!_GtB`tNMws?+k%br`}{kmI>`0{WW_*OuR) zbNWeRC5b*6Bjiywe}#6JCKcLbB&2rFR|00z^{BqZs!$m=b7`jkEA)1eu0TTxlv|Ga zQ`3Bax-a|sfgmXy(n_!pH{+aNP_y=+sk#miutv3C<;~xK^os|ZPBQ1ibh^d8ob%lI zgG~KNpWi}zp|$do&nfZ;9B@8StoXYeCaV&X_w7iAg0|Ck;?D>k(;*(%Huo06)0I0G z;bEX#3GFH-4~HG8P)S`C`*E*4w2oMutgV`b9l0ZN_hUc9MA?JwrABg4Va(p|Q_%{| zr_0m^dJM3%ZtP~2wh6GhPo>40RAM)OcZ=Pl45*WQ;?wW*9)})#Yz`!NL!pko{Nb#H zM7|TvpeXO5GEt9?CSCT+hwmc}8Ir?F=Q@xC7UwQn`FEZ;PPdB^FL=0_FP$#f>8}WffHu!I$y*$ zo<{WRNpTy0hZaneeSH^hoezCa?0b%>ltE`8-{<_^Yd|%$+NYdKgCuYw{5xR zt^BF_y$L4bsLrjiIT9RPJG9ld?r21*qMOP7Vmb+e;c_K8UfVD_VG4%|=|3o{}+jNl1;d@7_fosK6 zCn7E|vMX=CL)0%|hqX>g&EeE72S!_1#I|Gd$T-ped4etnY=i5+5BpjpRamV?@z69> z^$S$(U$6@QAME`HRFu!sH;R&j0TdNQR0I_y7*H^4jfx72m=P5NAO-|PQB*)tM8!Z* z%mGjkP=bJfD4~X&bIv)Z$z-^^-+!K&z27;m{eI`%d(S#+9oKR-9jCjV>gr#0sQNU> zTE2ZMuOF{2f{^-IrIK0)}wpHVeLJ^k0-@)_?mu zeq0x4_CvcEhO8wWf<3#TXX&f2I~Iyyd%4lrO?R%{*qrEAwC{7|u3N$zgx44O18rM! zafWFJ{z}{!SZ&mbQ4akq3!!L|5ZwJN>%@^y_yNpkOfkKQ4yWl;%_~{ZZ(zsgT>6ZM z9-ZsSB@gO({NZrr+g!ziE3b3y`qt^{)*qMRd1r|AKTOiA)C3@IlhCqxNBKDALp%3b zdxX?ObM`NL&CRATSZExyz=`AM0XufifCI|jz@#J!y62um#{7W(3-0rwzJi{o@$eAP zoo}aY$m{|7jn-E>l6k<~JwLN(&vIanJpa?GlmnGvnNuVLe37xHRf3_*^)uXR{-CF8 zhB`^q=s~vV8jhV>ES7!4WF>@6ECts>9y~0^fhDVtkNwocv>oeHOk0$p<*!gOD$X?Oz3At6PW_QRQTGn~E2QyeO@G!K} zz2+3Oj7x>pfBFQaSB3S58m&m8+oyJXv5iHq_&{?@|0*thpDtdeuuu#blO(ozf9Br9 zYTYZZ7RS_uLX+sIRV{z#=NJ>oqYq=bdKrDjaYY|%VJHck6>ZbvjfFRq16@~D!?^KU z*5$Px3fD1<=d^?B3CTq_T%xBL4hhi?3}5T9FnDpg&`Ll#cEQy{AbzKzT{vK z6u!=yrE;SZ=n*rYFdt0@R?FD8TqE9h*6km?*L*YB1GTevPO}T<+O4GDNN#pkKf}X6 zgNMfQ)0WJHf!V4%Rz`F7KS@Gn+gy{%e?u!dvdMh*cWBOg_*S;{84OU>yaOA!c%XK( zXUa^~#n84>@!IBVe3)|M#a>+f z@(^@rK7I1YmwWEoS!XTx;>1UwOFp!k{pu3*TmSmL%x%Npz5_agP(}V7UcaOCaSP{B zT?NdXks->;`3;m+wo0@|y!WE0m5nvWUT%ZT*tGA)YWsmQkkol^u0FV0`a>(3zSQ+HcYiyt zWRcL$Z=82!*d=*6jTX+mpW5~6Z0xQ58@Ub^%o?|DY&B;6IXOc8AM3-q9$iw9a% zw4do$&bcD?s@yv+5xoWe+&dGR7SFj4092`IUq>rHLWjYn1F0MCWBVmbjrmhABklfP zAE6aY{FOE-;M-ztXb#*L8x_(I;oF6K2SVRqIQ_b3vmX^Z9L44rwU}UA`}zBU%d>#0 zQ}C7a#}1;sW>0(mP!v0VrtD^38{*>C=eF`zTRD0s(l^8>1uIQ_-|(5vH|@FiC0#!3 z@>87A4)i!^ZVx8>6+ROJ_x$TtS;NQE*1j!3b+g;C7VQl>-eU&1v+=-4bS%>jF zb|N|a%%?=!^d8pYA z>FX5D4{A`zcXN1i9M?|$xTF4(gp;1opdj#jSd#_a`Y8ejb!(w9f0?_?9Ba(knE0z` z>^ro2=e;kmqUz`9&dy`=3 zeFpbC%_**Pova1vSCquQJX#9PW<73tB8Q;S{l&RJAs&0)Ft%yxb9LE+Tz)wt^;>(B z#{`aCX_~Q{W0xKiE`=>rPvyRA>M>4Sy8Pq^U|5;--IGg$^3q*hPemIsb1yS1KF1pd z%om86%!Mgaoeg92p?Gh*#kL1~>$0Vb!r^U?Q80zKF^ygzlLVZOm}s; z0Br%=%tb4?=Lz{GgWh8W2awkH#CTf!Rqz=-_1*m36>ROw4xyfTkJXBEQd+CH^RxSk zP34T~!a%!}-Zp3Q4XnE{<7J*4SAU?-sHr+q&gma^9CUY_C+CDT$ABZjc3gcSI+#&~ zf1-eK)Ope8-z4ZsOP3xzu1ex>D}R#3gvYExTf5)7> zWBBPM8)$ts@5jS~Ja*Q2=tG&S6riJZkR%kkCR1x3_kABT_~;Ql$Fn|j>is!CQZ3=m zeb>NP88XQ<;TPmmsjHsMU5g!Y_YRzoiMwRPL%r|r;9TG z^C87{%mc@+TOvzIBC-lU%iXfcf?sbu?S7Jm{Vj`+xstf{eR|{0`LLE>g|Eu9gAYa; z15?^VeAk4&IP}AGPW-j+FziB```h6m*6bf*>KRW(hD?7EQ|t=1pI=BmHLnw+)Sq5_ zar+nMpDp_)tos!MA76HeOZtN~^^5j=vgyKh%{TE6Wp)CsaqY+?;P}5K3x9n-Cs^K; zi`T&C(!K3uU|FlBJUt#p7A|>qGQx^umzT`H#&sEeru={Wo`u@Q31WvAL-V{Dec9sw z(XKhv8Bq6~r*Q0UL(2i*7vq0;{riQk-Q1x*_cb!JSD1^m`eLhdjQOGypRsn@`;RiF z>PU&#mv>?D;BZdAKFriid3%%R9VqnJ!kI(!BipcJr}8B(JiO&SR$)JN!-sxeM|5Z^Jg(tsBZV zarY6&_Kd@j0QdeCOCu!VYp^$$51cX_>>U^Jv_pN|wr92!&pQ}w_clCv@A*fhMU2k) zIXDOCsshJ$4s&oCS!ig0)pYzf#$0c?{GdJ>TBZud25LniGbS}k{}Ru>5Tx#P4HvvU zfVm)i|B%8`jJhXbnJsaL%zs>Zk?Wr2$O;e7tyj4~68O_7HDi-BG9L-WOjZrYk6T`r z8c1;8DGl^qR%ZIV#;;=?_oQrDKs#GwTC}GRI-1;EMC7@43|Oz;9&S|QoO_T2W>#ip zDT@E?dy~aVBeu;kc1Gs)){2AA4X}M@QD80CKOm;`sgYNB|JCmv5B*vg9uI8awR8KU z?b7;`ubM2dA#vjrvMgEj=sTInb>Fcg!S`a)BU!S@(SrfsB-+RVk9SW$tauBD8}{+} ztmWVdvS_i;hR;drIPBN2`^wZB=yQX0+)Wpe`BU$QrPfxHg-z?JvL7|cA}UK`Y4=!U zk(12$j@n@yw$FSh8LLI+bG;)pZvT@kxMO(rROLt5W9{y}O@&G3KfSwtVQ(^|Lzt?HWxWacRH>qw!SORo7SxXk9chvy z3r`K4ks6{8^-*J#%_=6)xNphSkg{xSF`XW{as%f+2==#CY)iY?i6ytbEgUhHM*5vg zfsN0(b`YpLOWU@6i$jK;S6f9l24o4Ts&*+`6LKgYsG!i%2 zhAiSP;WyI7{myaB--;S5dE_6qRLu`QjZsLkc~8-=mqXeQ+T&~bJbsw|$`3WyUi&5P zki;0ipwpkGm*e0h!a7d16=?gUZu#1l`{nhU_S;Q&#fxUkeng7EMm(wI499}V4nqR{4 z(}_)Gak-UM9Q*>w)0OMH=Wy;p0z=TH_U*0!9N!i@yG*qfnd_cT4OlP<=vrsiRE}Op zmTUV(sjuGHulvWsICm{~9$b5v4}-2ayf1T))`@0JADm4-dyjMOgC+O7_J?mxISR&N;Y8gLxrUZrbloze^70~lk-6T;Q32) zI@e!g|KWyCTGUle{qjDN=fg)@WZ@liM0cFv><5tq7S9j~zWa?+?^jVkQnfVK{ztx3 zluY7k4j<#_{>^5EbvvNBi(m4^$@$?{fV;Ys(6xEMG#|Ji3M~aK?UeTgpi!=O@;W$PXdI@%A^f ziCzunt@n`s#OqIaWVUtpgR7sg^XK)KOWtoK3*TL8e?i_A8$PUie~?Q#0Y- zhyA+LBJlb&Yosrs80zGhVM{@IE$j6kWSv;}`r^%Y>{?24ixt^Cp(iZa_tS`Dw-#te z(oWu#7ywFO?s_>*YTj+_u&^*kIH?O&-zBU4!3l8uVmc?S#l+B)w+C2Ir z4rGtF4_Rp7kWKZS4c9ENQLsd0?tE!7|BNZdM;8xZ_ZN!_qo8lttM_>Ao8dJ`dH8Xo zAz0vuU%$u#uR-V)bh$sa+LkO5{L|nrX%JdA3!7?!d zO1jlz?DQINq}}KE_kilBY0B!R2$z!&bc)*Zp*f%9Ope(Xf2Um=q1~NwLO&LBcHkB!)WVTpx{o_@hei<>aqy-vO z9(sHMuAlKghDBn57YdQGezqgc`>%S@UvX>a_Oh)Q!S!SLW!2%6H>D$xsx~h3WGYvF zQM+Z`J5*-kz?OFrw~FuJVApTyv~(IWPCIvGZYxK+B0btg`wzA*-`(k{z8+fIn>$wf zXT$Jok(LFI8n8Iz;Z>El^U3@hsNSv*2FSuPuT7>!RdC0@f$xOuEDaox_qjk@G!F-! zIInHjU16%CFi%EMoLV!Mt6r_Dtg-__NsN4e`L~pXIza zfV#~yvU~hppirdzRA<>m>aI93?4 z?&$%1#ZYd{J(!NY{Y%7m!5OmX3lW)m&jP5UJeRLlZie_`i(OOOCU9WBfJVz}FJ$Wf z5dT!|fo*Erj)k>x_h1P3mB>o z+t$1-;Nh$8$8j%{c5(0V)%V+Oo2ju7+TzJA^D@6dn@z z{nLw)VN`0~nwUoxTy{33T~z|xCylG0zJ35(3r!Tqrmw~}m%VkFGQrqS*;jVpiZ{NE z%(N~FW{~+DmoBk;Tme*nS#3XiL8u*iva)E?02T-+K7IFcBQ$lFUtFEWs9d&f#z3vssv(&+G z{Y1J>w`{u03ml*foOydx3i^9rF;^^kiKWs)aYuH|#iDAoe#Fqn0b?t;wfq$RE>@{L zcUL1Es^ZVyoU$Ocdb?Wxs>s9cOcOfQl1dh0#->N`=aWSYb^7tHCbo!IHGcH&$0iTS@pnGY zFn+}m@L7Bj8If~oPG|D5r%)^BxD(fp;)sv|EjAz>M_g^*Ec~$(2iJUr`K6ru24sOR zOZYeYp24g)eFe@op*;P|K6*D|W1C2nS=VWf-A%0DEG@j}c0O4k z#_y}3vG;NGpQBcX>^p4-S8|kthiMJa$`(l_&X}2%*cS36DRMlMq5@ex%H#^9; zhnAIZ`0Ap(kvW_fmtrbF=9kSTxqBAywFEjO&Is7e)xxX~@%FBJBXHEsX0&cSi964g1$yR3n!@lM zxr!N{J8&TEC{C5P!YCz0dX&-=?9Y7_Svl;EEQ4R~4bSD{fX9NGjL3Sjz(bGd=h=Q3 z5w|kge!V$Tef}uiJQjid2Tqi>JxoO`d&2mn$!#1r{qF6EVaAH+5Bm z;8V!dQ92%=P>jjpt!XRGouF2fesN=;3erdw$EPZfAxp}0@3DKe_;J|_YI{vAGAcyE zHca8U@5iC1@1pkOQja+M0N7!FcW9C7LS>v+EYHW+pnIE#7=N%VWBu41zmU#!T;C5!FWX}m4 z)@yC3JA9BVRGbs2yPx}>zCd%Uh4XVeM9&`ut0atZETwAlWvNv_dpvieV23D-8w$LR zSjKtR1dVSs&TA{gfo;Oox~slm#iGSGa_Jm9 z)6lNh6lSJA9}89*9)E!GWIoqP0ryB(xpa;EIHOIC^E+fgy|Wz)Ixj;?)2pxY+qaSV z&L1^CA;p~+bbpPi1BqqW?|!8E4t*BSI9Zd|1%*@^bK4JaJorV7d*7$$o57HgPAc?z zTAZA<_ZhS?T3u+0J=nNgozFFcJHAe}1BFg`a#)dn?t||AJfz$n<2P@giOlyd`jYSz z$Da>>$u4V0mV@)7xo-@hvaxz;v~M}~zgfWdX=w=d2ao4h+z!LCBQ0MOH_n3M^w4i) z=58pLbzPLwBTC}CQQ@YhV}&Ex!7IPWh+xHFL{aRK*U*@6x$HvKC^FCFNWHB{!M3Fv z92+}Jfz`9Xt9oL7_dHp!VD{oMprqBQkhR-OaoWhz30-vf^blDvQ+)bp z>oV*vXiRmPy%eb43aRg&UdGx^Nu$$O$xx%cJZG7&2zIx<2{JR|zH=Kq;AVK`9+@mW zXWDy}>0{{Q9=U1i+{57iQhT18OCnM}Rh5-hwPJY+HI2k<#Sx{ob7iv^1Eu(zS@_3r z>`yUWnr`?n=XwrZW{rnv&?Hyer8n0e2NI7}EC_B!ihG*%?5Ve~UNKGJ^PWQJ*b?XW z!+AfH4{RLSP&^k0>$G0Xk*t8G^X=Zk7kjWj$WrpW+BWVw!rFfJ^nC97y8hcDg%)WO z`n{U--phkr|6)}|@0s%}Wl@hvy>2z;VXBT|E2i*h5vS zUOadOyCilveAAe~Kf3;xEymB{P>Nf|u+dJk&<=%6q-XHdYg9VPSF+8GCw;Ki(eKXm zPmphXf12f1T{7Rg1qXh7K7d2_({0qMe_&5m?dNdo``DQBdza66Hgl-?aKxqNy|i&M(hLurn{v1S!;0dgZEl+5uxi6yv(oLz zVlHoz%jrVuo}Z&l`)Sw|kfwhl^&Ac_n?7@L(PQixNy)ioqk;XAhj8$iFm~*?6nZIs z2@bvrk%_-8f&ERMx6)$_ar``sf0MTkGWUPq5#__hvCK833$vr#plrmT_SR(1{Vymw zrr$p-cL`ryP?x`S{{xn9FD^5eJ_E@vl;d>&3pjAQoE}xO3fc_2r}sa803E-ISFXD) zh%cqHzU@6Ph7@P*+iR;h`~-c=D^ynB*qd2;u1TtFAFsw{U!QML1Bo??#)CN@<*Z^jH7gBtFwsCZF|3VXW-PRQf!=zm?-% z_uNjmKYdlm#gkdPb{!ewu`7aIy;IWq4=Dlt&2Fi5(OuXq^VR;oj2W`Fj|v=ap&`>- zG$*=W7D~>E^4(k*iC=%8ly%>!fL+-t6}0wGWPu0Az7#)kg5Nt!RnNP$92?UuN*9Pv4h63qr)jmjBcQ`a)Oc(+_ugY?mEoC&na7c-D5PBKB!xvM z=TtYebL`gOh~%;l=?AwV?Zb=D(Xa&P(qn>bdQbJny%$&DKe!GGL;9tbv!@`I$(WDfRi; zP2PQx)@51Yw;XuwPz^f0r~?;zA4s;uiC z+=Aman$8A(1`{}Mr8UWnd*0d-=K5y$96?}CnNfD+kT8zV`MxQZe>sG^z4$2@$MIu9 z7SI!VsnfoW%;!*AY$X1NEFzLx?40ldnp-b-AJF$A3)pv>ZEF3BX_wwbUv@4bi#3F~ zCOwWO3uiR$T6S;##5|Gm^LgSx=8ueSKNZ-CgL|GGI%12+Dm!*Hx5HyXAD!lY;Uaf` zt-$)b|Dk{J1J`>5P6!=4hBOB$JiBTRS)|@Kl@e#2f_i7&Nykc8bH_(LyzTwPOo(7=F1oYy46u&KUM4~&5iE6`JQ{PqR(ws9`op4vY40791SI1 zXcFDxvsmcbe{YAU-Ez&sL3Iy~{DBU=9nnY3PIB_UoNC;4U-7T|DgUwk8m(RBN=G>7 zIM5Zm{#i#H$G#F$j(tn3ZO`S>J4kM~h1XRQ|Md4V2X$mI`thJwvu7rDE|i@kWHA{j z4PyP4y{~aJ;*{~4A8BO4kfT+6mqM^%$GJo$Wo@!({=4Xp%9_|;?4u_bZUB@`Q3KUC zEAXdUjKRVMoO?momLgnnW)sIh0DLiYIoQ^=78qsc{9pD=gZe$5Zqd!NuwO+o>`nRz z7JON!YQN+t_8fU%qEYP!rJuL#&uebM@v%#PQq|r-$xp8}OyRlEy7Kdd@LmP%w<=we z9-0YFbKML|hB@bMP;>h1#Xqg{uuEXQ)`=1BJ!@KJgYgUL%_QNnft8B zy31!D1!>3JB<4X5WImp=B6&0j%1dFD_pHTQU_oXHVCXTFt%Z-)17_(NgS= zePC9`8c`<;oDx->p6h_Dmx&LOZyVs?-eUqi)IY4!Zqwm+nq;2HFAo6#dOFIP~>T&N(XS7J?lL zjvZ+d|MFf(F$3#m{gHh0%!=rK?!H<{^_%?iS|gx8smwMip1>2S;%+IS-0xMTuh)HO zdKJ{BbsD>0;j!z@)sJS(+!1H!$h%Hc{BPer#I<|h5mN4H6vXvU-Tu_!#xB`9l8~hH z>ZO6_0dpk$;`fc`RWOxO z==^ILXCDnJD{uLT^Gia;k{O$K_wm|GMBASH8D3nR-lppM(2%A{5D&MJ)z4X<)U$o4bU|2dAas*2aX_;kZmd}X^* zFd`}zx*Ny#Up*NEbQRkbJCYr-C9uUzMuqG5xh;IlX0eDUctSIJ>^pBEG*_L!|Kf-o zq@^L_e7`+V&RnCq{U6vx4a$i8;$RIB7CgI8!%h@fTmB^I_i*o1beW4;IJoiRJ;u>5 zOnoNTu5z}zYlo}CW?=g6tM!`1v44rwSE;)ie#|5Z9a6Ex=ixY%dE?uR1@)L2qHOiC z){G=PnQr!noCtKgM~6R~UjfF}sqJTa_@V3e!jUQI`+@f4v&X4hp}>-@p)@@k=jlho z!*4eR-xfm2>DCQKj3H$8(AyqYvXGwE8F0DcHv~+#DTy(Dij=vvIXPxlQ17t*!o^)3 z(Efc=WDu2tWi6!)wXM1&{!pu^6wgAOm>C+SMa|HZw`-xq`CGuqX*D?0D}||slE02> zaqdY%Nnl|}Ph%OjMf=Nc3CV$~eXi+W95S$}xcnmj-nCE~*7HRF^IGUO&@OI&#$*2v zb8Xc0Qy-*bRQL~yM)U(rIJBlZ?~53Ac1|chnZ3|6?Ri4;K8`;r=qg_5l&Z-j2|U=V zNE7DF{|S9lVOic`8rDzl*mPm)E0U0md^cg>X{EMxi`0NRzQT1&Mmcs*+w>@}{2`38 z>beRpt-e7s&Qm(U!yD3Rjdk*RMO)4JyuVcz|7Kf(75aM(q1pFKPl>4 z%@?IRIQP+UVA3BUbvGgCKKQFTv2qTiMYZh>*Ia?4vhTzlJk~(|{oNV@FZ8kXh`{%L zTVw41B_mhpI}HYHVX?|gPw2l--b(h?gR&yi6K7TlK>M;+W|4-bP$)yEwa$u#>Y?vU zC9No6ynF8#GtU{?k7mvN&f@5!6W`a|^xm9f$j}c7QVC0h&fkZv)|g2_^Vv1Ym1HfT zEnPT$_9QW=uqme&9?bz-KwHRo&P_=3ORY>vaKI|F(cM4pTVS2ow&Q23ldxsi0&;rE zCoG%2L@H>$0Vb+OhGi}Njh*Au9O!x8KtFuVGPNj)BycS3rRdK}V9A%POiGTyDxtZR zkyabAu_Q0g^*r}pbKmL4pJE*^fNHb)NNArFXB`vn_i{?uk%Ej0}P0O@Z?J=5pWV^u4p5aYi^9XaZm7-<|sjd+my+`B^Jtlj)s{S_@tQwJx+OPN}PLfz*Jc+v%WqMTK6vH z+w#j3dhf@5kK50AUx%$(MMuMPDOh}GiqclPAh3j$?-VQCf?f;%_zzyTBtf~X*>0<1 z@b@9zbFK@wK&?DY=ZdHVHa3($=o_TKFk`PId6Npz2Fo+1gtfvjv$U?myBnLGJGVQV zNdw(&_pF2;A($4nOjfnU51ZGTiOKgzV4PQ(i?3}1Xxe+tXuisQ$J&|myvyOW9@Iv6 zrp|I1gpSv)HB-p1fUzdqZ?c0OwC5b%zJGojRxB**nkpkof7^h3o|_KwGbRX!F2yXv_bU>*rPhJrC41cWHCp{{Ut9{p(}q(n@;OOCrnJ)!a9#ZI{mJ>tNX zdojjB^BOqkFqq^Vn`$FdhW++eXpX;l?NxQoy?O4_p+HFdaSd4azCfX($elP+A<%rO zYcGD&M%pJSm4fzatabF>BfU-p7*W%jd&6z8|3IRzd!z)9JT6SwuQD)43g0L_enXBh z#kQ`2jmdt|KrPG|st$fa5||@*!ZG6jBJ;qHM)Gp(IQe?xQnjBj=t!Akyqph`HP?qd zZVTqhZN=kj3fI>{)fZ>_sO~9X){6I}&waD7{QJYjF$?-p?2f_Sg;d*AI*Mr<}uVEemD-$=cw@>Jdy( z7KG8df38W zSNsRsUOun$Zsxg%4$RE!Gt54o!!$+n>*>=||I@g5>QkSSrl}m`+AC{WZQPyrT$9A_ zIcu3YX%pb|79;UG&ihQH%ll)=>I@jy(s?-b$a{F~aY0xmbrZ&13SQJqI*DYx)Oy`T z@tCnk`R?Pwe5iZgvh1+988qb08n$261A|P@Lj#ZJ0&8yb_nTH+yKIay4WSixPC@rY ztr=45NmzAl$xOY_Jy3sP+QAv$xprJ?CR-;pu<8q``GuS zE31)y)w;;yCkr~4Ctq1vlK}&E1MB)$aNfa!e|(zrv{&&!Tj=Aqkj(M30bPUJRFq4) zusC~HS&aKT9y;aUI`#RN#ZJ6?Z~WGz@7nOLd2;Fa!ju2o^rj0dUt<=&2gdV6z4o?2 z?zkww_}@fv&d-7NMSs(}5B{8Z;|t%z^Ukc`X|Lans^yt(!sQ3q&$~C)*&tQp_<-fj zG920%t>ngSK1n;-{fqyw`3-}MKZZh& z-?Z&J?<>HkBOA5~m~ih8kFvtHXgq#F5>VE$xbhfVh^l7!4SAhAO(cKKvDT(+YzB~uo_{rCkEM5R(m2g>L?``fpX!Aew z!hRK&+4M)5$y|i~me9n-S-j^C(UZ2Z>Y^v)NAsC0f{A2k4bMFQjV#Dp9eC-I97n%} z_8P4yaRGHKxI47nqIEWzZ`RMlVvE)Rt*ZM_q~>>MJht*^Bk3KNkNU39i5cVga{^Xn z!+~Aih8Q}QHqamW1X7^o3t37O7^iDjOp$c|E1kM~tHS2%odLQWU#apr1EiI1wv*m$ z0R2mMx^$RtgBn3qJJrGoeP`|iU)=`>NkYt|b{D>JZa)+<-|BpQiz(4%$4sY&L$0Fo z*(9m+*kCW_`Q!3&ZvXCkIh02jVCtI+-#=oGz$g@!pQEt_?`dB3xU%9Vey?2?W@g$4 zj3Q^zPqL|)>9cfK$n=SRc)HxL)RDlh1o3mrFFyg+VeygPQw*Tj9Q5el!F|VBa_i4i z{hnOR{{T%uvcGZQf!Gg@od;~(k~Qc0Q5h^!r(GYnXOQ_8CNDfCp$CmeK5jn`xmd7x zd^lkGX`oG34col39f#bnerj~xO5)q`weQ=K+tB!PyI95AE%MD6XsIbU8Jbr%^B0oP#OB#-VGo4k|ZC|~iE&9M3(T@kQ<%<1j_uqg0Jq}$O zTuyfxA8mlqgYgG0Yb?c9^RL?O`{YO>l5y`(X2w8+NtI&T4Xz&z%0csrBPzF$rRufm zK|c50JS8b?Rc328S-@^?RL8Lrj(#9CndIU!cLZ4v5?YfzxOlt`vU7ij_F&$(cCUz# zc|7!J($^U|Kbz~9sk6!bBee1ClVP($;?HNFqwsB&cGzL{DVS4xf7TV{WS(*jKlp4M zw(8@6WDDh7%oeg>(Ko)0N>8DsTjt2@%a_S~kw0be(!h1Ok1Ej9vgIu z<=kH<^POK(a;xb!^wG0(w?|r$MRz}9YT(&*U3L|}LYPbu9W07;&2_rLXjgY^reO1>!G!k}UY z4NyLeca=vzET1x=e|8m-R8Q7J&7|0?p(-2j&epnTcl}plO2tC2$CH!WC>T|;ken6Ae7bu>)A)~yseFE`iSq@BS2 zOUD}uw|k@GUW)wXSOe^T?zVca@j_%>V2Jt72!bJ4d_ zUvIM85GpR4>~x?z166tb_0py!XltF~cmK);puI~Iu#AYn+OPrFCtCm;@*YnQxtaun zt{oJQqyTLGQGGq_5)&TZxJMWA*bl4*y5Gfh#V}HGVcCvN>yV~U8=F#p4pUcO_8JZH z1=bp;tp(@gfWGI$I-`#)Y!edgP08Jk{R`r?^$+&pfcJ)owzsy(SahMtUVjf+_`!46 z!#R(Ejz)W;Zr4ML`l1yp?s(vv_u)4d-qD3d9fzwMEN4Rgz;Dm?e^km9_%A#yeMU10xEwg;n_yI)4#|1s~k?n|IO zn(Cp>FzbkesZ~tb<{4m>mtJYs}Gccyt z94c9P75iKjWNJ2d07FQ>=ui4O7}$P#k*;A3_B|C|m$AGB`ps;vFTTw^M=J7tC6yu` zk8$eM6&o@+`XV&5WX8YAFM~R#do8B8Imt zx1FTthh6I=Cg)xzVK3kQv)24)v9_{miZ(+5I*aE5*rhCb-9nZ0A=AL zxwd=(pnuZL+AzZw$U<+GX2pLb3C-75fBS$dmohIH`>s>yh1yK&sr|3t^6(dB_`#0o zbndxLSl>_K8mTd4<;mtXyZ+TLF-c&G%*?vwxm-J)&7z;* zuTTC1OedGT@(XD&ETuiN^EV%meInk^nUey&oiDnp=9d6tv5tCE&0ZK(=%CzrZUgDc zv%gd~Kf%htzA3^!|Kf+`RJG3wiZqVRXXP)h-2?HiBe$OJU5Ov;e!5$jN76M2O0B43v+hpBC~44`8VYWP~{?In&vkOjNx?x-;y(6PDDB~{oJdvjC z$1G3$0|R~@il(<4k=3*#z|2i<~F_rTO|1Fn=CD=X>`q6wfv93#^F6JOi7g117tn zI7QlclvPCH-?(&F;x*I0A2+H5y`_W88g`r{{{A51wqbKZRL^FDU8@yQC3pxpLhk zlKdDCOnlecFuIqE7a47Zhx8|L?FtR_>|gQdx+476m6+sKGXpxFX}y{Llh@mrg-*^4ga-)@fne^TD4Fj&=j9 z46`?h_(N~?j!nl8&4Iq#%KeJ(x%YuvcmLLDE>q!tPg)-}Q$rd8tLgD2={u21d))Ex z@l>GiZ1b{6lEM5{bJw{JNI<{I1v8n|;vYq2-GQ;YWB;SM z+ptAnp{1L22^ik*HzkL2{gjnm#j`7KJcLGFhetU>C$P3|9nFa{3jO{D^&2+XV&kkH zqnDdr8?QD+*Fi*8;CAZXLRxuJqyH`k9@juN|OW5_Zk-wS~s1`HZnr7B<{DYL<5H35mbFof0;00KZT!RS%kQ z{E=bJt72B`6=euLpof8i($FK8e&R@r*o3^fhQBJ$V294g?dzfg*nj(}jr21=pfBt( z&f2vYvvuVBuAXa0nqOI)syeS7IGTC$>@O#|`e*O$nRf*GR>44oL7Mp!G4Nln`m;%b zMB@9&zbh|(ImYGBHpqEe2?N1=R7(b@9+^*W?_C9<4xaDSPu5rxeN3Okud{T|QO|EM zOh2$`u_0H!(b^Qew^y+6_pueZ6?NQw3r6Li)t_a!?|Jg8R!tY#R|Ro}xqZGO`dB1$ zcC~W-Y;>o{&J?QQ+GD1v`W@ajx*H2*`+kVOd^3R)i}`dc+u_%(JsL*ae*i7o$sDt@m?5#L$zsKrfMyxUa)w9|$P}=il1W zm|VIA6u#A)U5OneHSPOyZ;_$(;g6TN1b*H9>`)_37lv;iv2vFkz?NG9V%DPi7?to; zp-hg8cSf%qx766WA3Cyr?@Ny=gC>XDrOh@+fTg0lldmihDWlOlscS>Ai5_z7*!Cu9 zl%*ElnYI?f9-KD1v2zG%F3A#?9N&@n?wUr!bSi#()($ElKQXzO|8BknKEL!?uHjZxHwy5#FKC~3C& z*QQx-!nUUIhpiL(TB_*H*K;m%>}Eql_$R0RCN4Pm-O1T+Q65$g%&mW;We69%;-zmd z;Df%bFS8%$+F|3{7`4(=FKnF@m?SxQ1(Y4>7XK|~fz5iUy1P9BF>6wWdxm8yqzPpU zn$HNw;`X|x`tdku^}1rvW5@N+ne-yT=uP%3==@`Gb(+TC^`GUDAObzA5UfOAFRByZY9vlCXWni8H^u#j%wV$@od- z+LxqCNir`RJ_D+NVrRGr=iU%BiWJGymZ!s(r=c=(zqK(sXVCiL?tJ(n|GV_;;xl-( zv+uKy4IeNR=U7Caa_dO3oZCn`e3?h|vHH>0{Dw9lfDs9jziygzGor-F20a zT2eW7)uc5k3?~CqQkSMno`*yC{r|W+jbO)->z#X}q=3$NcHVu;EU0&$7eH=W3BSHr z)h`L)`j=p|b}#22a01q?wX0#=6PPTyV(IUPAnU+aYZsLTm=N3KuXG=~2^;Otr zhj;_12=QF%cBCE~Z2c2uE%Sh$b5uEE%>*CaNlmBQPeSI-EVrM_jp<7&f?+$vyhna`dD?6jW1>jPJZ1oS!kvXH1|rK;22f&Gs!A8_N(#%{ge6=v--AX%#w zeqFM_Xrr(b-!`~n!>(?ZS%=GjeqrY;L?xyp!TYjcC_ z*L|K0w55~vg*S#Gb*sk3wZ|4t$g`dM9CjEYvyW_IAw)rjm$2DhhkwCql*b!~zAQ9)ZnY<z67)yXNJt+2UgP^O$L{P|7d)ny^r(r<0( z;ky>=2bR5a3Fq3~p*5e$$k}g?WAdAYzCIttQSY=m=^Fz`k7?R27Q(fUN%_D(skvll&PEZnTqx4|KI1Hb9D9kU@12peTaicH3lZ96E8M!xq!=evn z<3|iPkVS?l2Q+0Jz;A0)_reEUdpH?MlkIlh)5V&SGvlw96=Bm^%Hz=AoO|Ke6yS2b zu-6yrEd#H8)-*?Q>Yb2Pw&M_d%+h?mK6k&jbVP7u^L5VsGqO+$X?Nj<3Hq-3eeFV( z2vVDm)-O%o4f#XcFYr-Fm~YIlqo3J@1M~Nq?8;aO&7aOmFrr@pL(}`3+lwfmtZTBo zro+A8Rqvu>ty8@Vx@tb>9pBl8Q8%U=8z_^3x#VWJ)z9P5yWo!7bM;8<*MBm+ur?Z6 zk1yuCI*IF-u^@g=>O;j7K)rp`=8kF@#!jcV(EW2UUSsm(Kp~D_3X;gf8skalv$69s zALeUSKy|mj*K$%M^vgbAIWE|SjDk7>H%f{_Q=7AC`@>;a8dcDt5ADYVP5jv z2Iyy~`(1QK==IMz7P?du$}6fCsVBQa&v<%4!eSGkRam!mgtbF6=Dji0b%tKC{wqlx zI@lAgRQFc%BGAMOdQR7I?LUl|MHW{*djgFz?k_7=9K?F`vr-Zs1nNcYe9NHO(COv$ z)ZLDXMHcI)+TOlE5{`N@rgxfSPZrx9_Ad-*kp)RM_m0(8v_S7KXCqQfH1R0x~(0HKu z#o>i(q4%`Bn9@)uHay?yCA>2KzsmtGo{U)A*5XwJS;>aeBsM;S*1cQl?MA$I`l;8o z&c+`cg5;bt{9>6bthIDbFYbGX{YK9;t)e|4uy^2B_OYLkEPlD}genD?iT!I7pK|ue zz{7vJSBk{nMe}f-1SraUq(ZhvClojt_l4(Y~4J+hLzBKQMXkj!2>GK?w9xP zkN}#NNBajA0bmrUx*NP%2(&wrzgK5;L2^>tl%iMXp#LOGr@4(EYD{;SN365s^e5{F z&eNTu3df(c@#%!s12yjaZRwR?z}Y(}#j8#XN;9{Ztk-shCh4Lt-O_@ zPVL8z)6=#{R9yl3Gr{>I1@nMW_`{@xtc~@Sm2D45T>D-v>00HgA)?S;x%^7kMeci# z`ZKC&pWo22Fwv_uC0%mjyA>_ZvFEpysP&$W=P$cVHw?tKbvfyg)s*)KjD1+ zypTks$2%mLJ1)fz*HC}b=1<)Hu+^s*km7QoXUn1dqh`mDDnT_H^u7sQ-LWB?j;_Qi zCpVkxx%t>E(B!_d*c2&mBH!rG|AKvdlk{J&xDM3zq<6<dY{wUW#;rWM?;5zmN4Ws_YGxz^dUa z{ahP>;=Wxi8jW0kY~5nNQv-Ic#qh!Xg=?2_&grlyy)x8itq&9{b$;KwwwTP9Jv$+8 zUN#QS)je|k=qsoT2*@mrJ`7Y7&AisDx|p*=^-S)rOE~!OWyI%7K9Uf>DA+wW28P!s z$ClV9XC)}J5#n3>iG zeH)gy^#}dSJI0gS`PIE0uwmz=jvl9}m=x@?@PG&B{tT9iY+l<}vjE$p_F9)sOUG~9 zhEyi4O@UExn3eU_76P!l%){d_G(5XE@@mKsIzP?6bz*!2ng5r)PGhqR&@Pr{?^~e@ z?FYrD76@?856Gg+4EgMv9|PTf$>;9nBhWIwc!8Ur6t;Kag*hWKB!P7QmCoWE{Th2; zG1ONVC1LW)xks0Ncfqc*;VmsY4Y6K$=1j611DJci3aYH+zUR;PBO67xUByO*LLn_D z4=AR2Z>+>Ep~?M8##PIDq$-U;M_f73=BF*y`}7`Idh=uBYwlve*d|hpnjzeO+wDIt zeix{j&-LbM2SPym)nx@Uz9ZA?n_ss6PGmfCbq|=m2P(mwZ0q9eR4e@#$bis!upH1`;ZD*TMWhDFz>&wW6GZ&zlgh{fx{u-z8vHx^vKdQJ0n z174p@z{DI^;VVVIq2uY%&Y$U?n6_2PX8GNhkbfq6e7(zJsBF70U7tMB&-#jc!SCyU zej}_!W9T?kuUg3G_}&crUY$A=ajgRCvPBg0Jol(e~d&L;yDrx-v)iz97TR-D^{vhNF z?;rUSz_klK(EjX^%d`MU@vDcGSGezci|EDK7N1;3{)2kyFT`(6GJ<9WkNIczbKbv0%jNPWu}q%*PwqIA zSE@ymz&L!bHQ3iNa~i}xWflI}Hydajn?%ah z_@S};_pN#MHP{_7D0?>E1?#H?;j8ZnpsgBvCX&M4musD#`!K9M4_G3_Ikb{((7WW9 z{JI1^ES&FK;wiLC0%wfDz) zR~ELA{Ij3mn#$cD+W6teh$r`*4PEhKqlxumsQH-UzkL?h9!o{L3OUAI1oOXVpI7Z! zLK0FRyCPQS2$YZ!-%~RzphK&+d}_RZbq>!w9514LyD;tV_ib1&dneVt zQ{a?ut>MrA93h8YJN}eJcJ_d0PW^+=U0a|qZ2y^L@o8A+I9u<^&h=!$vURGC*K=T0 zKVtLAkIE2zaLwUGr%9Okb;pT)ier%cW^VF%<%2NLOYPhhe40DnoR;vXsRyCqps$J# zH6QQ1NUeUhGauT19!hgsuLP{jOBKD7ig0}KLt1&S6DFQgh}}HX7;6)w|MWD6lLd3E z(@ivQK=m$(%+=cY5IL^qE%KJ@8I)sXegEOTYs1*C4GR(8I5f;B52NX;kbBlR?8&DcH-^!_|Es$t9>SweHH zrJYhC;j)UhVP_jKGsI#ig;;^z-KX?wzhvltcvp9OujPbXx^{l=ofA+?4IJ0n$-Ph2 zu{eG#v|k>3|0E?|@#gx&>DiUkMgBgF&8%A|+grKk_RPK;?@e!V?5ZO3qwKFOAcNFT z)1L(!NkGFAM>oeM8b~fvxpB$k9`+y6Os?+O3hC05GGp#=4c^k%hDtcI>4{? z{?B*h-+!Ht+VlDFwFg8@ z6E*v5xq9a`O~Y%Rbd4vU@7Vd&F-?W-PBp=1M}k(Dp+{jk>OG3E)WzP#<$ z&9&02a0Z43En#I|8MH_&VE$}2f|eY&*I7bq;OAV=Q2DrO%#lgEGg`DByNxFsDd;yr zA%EJtV|-W0BGyG~pMMgEzTaViyX9XZz2);azw|TM_#>b8BW3t%M-P5+I0n4X5^kcCi3~Jxq6Z7W)P=gmd&Nbt{CuLfdCD7yDVK6sByleSK ztls0Eq@B}=1F>}`jm)($YS$&U$g&uj;>T(Zh`VFQL4z(OTg0fL-H*4u*@?r8dOLHg zSU|tIVPV~{JyILK<{hzb!`0bFW>ME z{z8S;h4FWlzA3<{*zOydZ#eIdv1`fVMF%@mG3iTSp3$y$Xqk0BK*}x+`xlFb+-f%g zN~(95uQrbzVGdmzHBP=dU7HQbWd}AhH*@TiBV*|YtNSH0A?cHW=53o{{Q7=C@=sO> z1b$Mi%R90fyH$NAXPkY5t#|h1RF^scWoc()lwLN_?io&A;VX%>W;@t2lgI7|)ahK_ z`l0c!eoukEV}_DfpdBz~Zc$xU(2AKVd7bgq6LLFee&Uo9QrLIpG$09@$K4hd=B35-({`bkq=FueMJhg?*QWkEiX25A#}|N8CtaVEDv26YO~){ zR`16C_c4lR)YGA>)nxbOhsS}IuB<`vu*89%dlb_DY{GZNH>;bX2Z1{I(Xvy`zQ{1Q zPYB{?V7pJ|lsDUoG4O+Jvd`;tKz_Gv+}3C-WTvYRpE(~G z<%IRr!S7bQwa5?dIU=+Dn_AygGKv3iVUx2Y$6h$3h@CrrI{OgNx0smB>-dfhPQpQt zHac?h+um01TpNpqqRA<|emCqAQ54u02^5+z+rGz<@8Z6*`kAitu57cx0 zr`I?vj;*p3fD~~IlH1w{mBFl4%OtsW>B!wj=Ia z;r_H1toakK(CKlVQ+^<4qtH`9C;S}{60$Hl9jI1uiw}&Sng})TJwDFGqjAM1U!o-S zAloy~dx0YZQ{`nxYF1{zukS~GUfkyn{R)~sRf@}jId5{Nx%ob99rX!|)m8+WVey4( zEq_c?^sO~aWdP&yiMQPuT3EF(aUb*B6QG>=^)*TD57bxWd0w9M6sQiCe0*V1kpJ3M z?RsVnCIvs&7B2I^TA4S(w=H6kZhmLm6RXLPkjzha6P$+i(<Eb=_EI>j70g z;-|}!UjTibyvt8rWylou#MhT6@X5zt9a9=9%@TzngbZh5vIWky5q7lUl;r7O@m z#zdI1-U=(?bMnQXaqP`PQgW{1Zc;n9y-Cu->bDE9`@!*nhqG4bjZtPBV>PWT#~6F3oXfuI<(H%0oD7W!z8K| zGVa-KHTv@$Uz>`*JO93&Bsv2QP20E|n$nvx`cB!v#l_=Ohh}+0htk>Elw=ufXXp>g zp3H#0qHim=uM&Wc%^MHxZ99c^V?XuWUN&Hjw9t!Qem$YNYDRj_l zmvkJR#6#zdeY=dh8!E9k#(eEw#jQy9sw91AOhYO#a)nydfO%dv>-|m|)_|BwV5c^w z-EB(faN+o~!um?l@E=Jg5Ps&DScnDBxyN7E;ijQKX(fxf_W&7n{Kx#4S3>+%b32m^ zDQu%Xt17X01_QJ2zL~zW5$MNajycEdM%rlm&t(<*z_@+pEA^fV_DH^1qCQ;>`ycda zFaG!xvJMv{2>S{_<8Hp7?KKl}rh0jbi%S8}Wz0{FZ43d*R{nJ<`#wU`fzox(BNga< zh`(MWnSvc3A4{BlR)t*=%0dM?l5-q%}! za!9XMJlu4R4~Fd9#oHpK!Bg4g!hC%ppu91A-1hq`G>QEla!BT}3&Fcj0rSqiboPVa z-y^SOR&no{)&U-ckg`Dv~^w!Ps8&#VxXh z(ARR}j?$}td9V0!%Ipgz8?bXm(7pw4xPH{eBc2yzd(1?pn6$dDEccxsld#v3BVFlcQ;UL)M`ld57k`2kp|c2 znZMTHO>491v&VbMB4%4eE$CZ+Gde zGkc4a`)1_>x)RVVM(UgguJLDOeo~vuluUsTT7Tj!mK;V`(_BGEv^)@h&EJWffJiXl#-Z(cMR$d*7 z#T7nfd!=ikD`1=Sq9sny-9vSLtI`RLyU5keN`t^$T9&bES_KXmXb0Tfz8hK}s|h80 zI%4lB`Hy8aC$VL%`ZL!)?mai=_GdYhg4!|tILW`B&Up`n9&ux)vn~n(W0Aq{MPj_~ z{{PrscARd>-6x_f?pmd#s*Z`5ROgj_bOGkw^i4L5J}eZN{ayZ+1XfSeD+!n3=uKpS z3fK8`$x<9TWzZU>XHFKM<2gJgtqF|D;zKGI#jx?i$npBbyVz- z47|oXFSpBJ;R<^{_&&iR&K%19q zHLtptXCAUHC5KK+Vj?5W|4hEO4^ov(ZN8N3`0wRN{9?}6UQzW(!mB&&j;%V(^E(-j z!sfk_`-UB&Q<^3hv5;0o^5h@B0L(2+|CSO-65kJV3B&o~STElho3!g3(A!l!UJ5io zze8lo(*75m{&dPp#M!Md!`SIMm!-@vLBFiznw4)gpss3WPt)lD>@wII+~&gdJ2cvK z_R&rq?!8E#B8QnKUNS&SakPFaT@Bc_j%n)O2ek91S>=wASYI?{p0}q8QhX{F-7@b7 zhMnlQQ;AWW`t2WIpLiz3wQp9WqITLkEz#Kh!xr0G&E?P_dAq_g?fPHqF~=+EqpSrJxaH8elIniQJ@Dn-*9tnc z&N4niHx~y+%iWks-?l-oLb27ky`1-9Q2(XdX|=&UXz&gGtm9$;L+gS)rY3s=_3T-y z;8asAKR95P)Vut@{~lSSV#k)}^A}?qebSrWji<3Yv+?wlc8>o$tXY>gb$pn+pG;G{ zJL+-tHjHWoO;K)Y$6T%5XNO54K=vdr8hGoDz4v}GgoD&L{Tfs?c+l=(29b(RrUIF_ zfuZ9vb4rXX4BRbTc3t&8R26&qh^k)2yu8aPik&PR)THeA_-P@Dzk4H})R$@Cepx~}1&;eSmWF&`aMQXOgtUGhc zp>gLe`!`ED_w1oxIaYG*a}5|-QgXyz(BVIhwlcFzBSJg-^QX}g#e0xgO6Q%#IZ zp)_s9wcqa z6-Qv5b^U%zP9K;{s(Oyg)dGV&b7q@qHAJ7&Gb!w}B#9=^Zpa!YjY-WmDc=6)tvPh3xA}HTohT)Q;)pByexAA*FP!c z)9cvAYZrOS4cXY=pGQ|i`jQF@rK$_iAH}!%0W5{kQ>G3hr2){f;>MQQZcacy{-vy< zP!dx%>xT?JvxYR~=*uM;JoZPif1fLVZ0Ke(|D+uzGlo`S^#S@|a$f>A-03iWcaghq zH|p3$^`E>FXq#$I7x$II5Ox^6urh#F0nBoj48ze?_vTMo&*Og^X;(g*&2o|9@VTJf z*P5*DBpeH#YuGtF!T0&LCC?ZB)!znpek?vczcJwN=TO749+d1av&h1B$#cVw>mqaG z^_U%Y+;fc9%ipFy(wX3^>1zIR&XcgOgLctq)&%`-ljr5}*&sdll;N=`4$i_%HPxg` zH^`7ung3$)3wf-!i2Z^V-1m}g+wZG|9XJlj77a_mx|1Z#%G!STp&W$z|9bDlI)YK7 z!Ab8#VxhtQnqFx0GM?|)zo6vAED07*I`gbr{ET^AzgW~k^Q)~t#mW3fDYIxSt{qa^ zPUC1JU7q*5(55`~cF3_$P^aOw`Qnx}T>6=Y8vESj(F=I_Ykm3ig)W@+kR+hfFGm^s zL+1ObFnFWp*I()1F!Q8s{tn*r7}`23Ve8r6nACo?mby5I=l6OZj`*(Rli^9H`mgY@ zG5;U0Z!qF?`pcnmh5yUzWMx2Rf%r!Ol7J{x@$TZM7&WV5pQrg4w%7YRNv0o!`oKM^ z{#zeG5r40Apx;5DHNf|0DkL2Ka%ktDeZ`P{zl~(`fOCF_ulKDjQ}}L&Y45juB9)zo z&T9hx7Z^Zt3e#%RCqZrb~gB%=;g(I7! z9x9Yz;kOZ|d-etJP3&<{PXYISc?bWGn>A+Kdw(4z9khkY9JvkyB2FJDhhmXde6Xm? z_TXQCH&xww&y*m2=)azNccj}0UVJxUBziA~K@s`O0~H*9+ZgDvxqNwuBd1;obMvFn z1$mHW`8)Ca)h`&U75l)>l55Yr>2mzNWKSMD&i|1Ptox(3SvC9rdpb!(D$u&cmHU3R z@4eaF%D$sOdv;@ZR?8uvj>7PWUh}445l`!@+&7pxrD#g zs(vSKy+%iH&i||YHoJv7>Lp|K3Twd-$3R& zecu*X20iEOY1a=ssi_j7!*W{Yc{5rTuOF^*cz#L;6j(reOQB z7)o)^6o@UX_id;)0aU89KTr^ctxC+N?l;@9-83_cv?m3bx1Ktv7-V5j?VR&^t18L- zuB@rQrHqi?F~4Bl+~=6@>Gn2Rkn>&*3OBtwJ+r?4zxRv8*AOdwZxs#e{FM9O%e^D> z_hgIAS}?}#w@fQ*@(+y(9OF&bS8Cg%O(ScP(*&p&C3bPgd3XdE&Nh%Nq}E9ed5$qpGeV#5VDOpI>C~x}EOtRp*C(hT81VVns;#^7;0L7-=_3o?jXu0zI|d6MA6}yTYn05lrKw(6?WXIuL9B-0{y|DQYHf-Bd zKQB5HNB1ujWsR=Dk&CgQ}r!gu!%T?wb_g&`8Ti3Nk<4oZ5B3D}#DFIIU(Ea61ZJ9&R)u<;b ze0@3O^C$n9MVkg>@8SV3W-4^4sWv;-)?x3y6|a6z`2^h((hGjy-wSjIxjQY$5X;1= zCA$Yz$b#B*xpVKFXFpFF>#P^(QpObYr%<{v&-DH6s`Q8K3&F=RMJe)ep6gc-l>gl2|3g>j{u&od`*ZvW`z-k+pc99Vq8&NrYGTXXYE zru%WvpCa}Tel=>*LRxUF$H+b&e^y9w^D*t8zZO~g#!_!vpOFOn;~!W^8$*7(j`qU} z9{cv3dgJ+`28Z<>v3!e!f6lK%*u3pqx9^qLK=YKaxpL(N3~0)08u0PhrRTv9`~zAA z4KH88u)Lba=$8jBy{MPGH09i0@mc zyJFgA&rf&O3Ls^5QU^w~YVF#56>sYg**WebHAcJSbIY(21S-l)uDYVA9vVwU*mK<~Leb zy69yzlm*0%_ig`#^yur?<0yw9$M#j_qy%=@FlDi{5oabRw@NSJxi zU-_7!qHeO}%%A^#dZpcG1m1p17M1a-(fIp$AXk2gCO;9J%d=lf7U&a*^BCpaYs7}b zyA7gKxc0+q#}kegc=6Q38vDlS>1TNH0j0RB@zIpkBmsRJJv&c%e7KDc2{8J zBR~JAHE*!lR^+$eG*?I{q@|r#L>n*v~w)u4CB@ z95K?0>i@yP1rvO{SuOiN2x(q zmjC3d2DhPps`d-b0X1M`{y65Dp$EO2mn_omvE*rQRO#HZnL$(-keaRXZe$wt_=Z|P zbI65$S8B?wxvChUf2PRbF; zwFdkre?d1ZGS}yMGd8>$os}8D(T|~d5ve`t6X!lX7Pxt=xhuoe!_#{4bWF@YGy9tx#=lOZWO!^ zcJ6T&MS1d1;jXuyJEN-ueOPKv1@13yX-0^_AtH`{Rn3Gb>6*q)_gL0xk5Xx?=VzU} zCP)xw3he+h!n`5S@?`k4_sNjHjsPeH@bwQnw9;JG^0gSSAByCESUE$eEl-6RyL$<{ zb33`syk#)Y@2idsJcb?SQ!ze&^wpoUr%DyA)S zU3mP9+Ea8sHo`Bb`1}lIQ0SagF_RH}Eo@1#*_g3BYsfty;9pSIEl|{=PX?d(v}S+d z$c0>*d*oiNEzQ;{^LMKh0f7|?B?WrJI+qTiY?QZ26s$nzxq!z;ljy+uiUgpKZJ#HK zrk_@MyD@&`8+u_Fa=aE*KEp1hKqa`ufqouZ5-^B2M(Cd*$f%HEObI!lL@H_$aM|*V zR9nF#qf$h^ycf!vYyp|BzI4fJ+3B5syC-c62eC4iU}fs69sBKStR1y}m?Su#ea|Re zyvvv^cK+!81+ES&^cVlHj+-&X&rYQITkjY0wHHw{+)CQ4t*@>Ne~me&cT(Ly+9>0| zzB=o&w~5p6&CKj2t;S($oS}|WnoT}N-A(UI=-AbsWR+PikLRvUjh(9o(^g`HCwp9NJvdBuhhlPg77r-O>P&&3w9x!r8ED|OGP!Y zlk}H#yu0Lv+7xci}4my4x`LpzT$Yn-{-Gsf+>Z?L(|K2gJR*Z@MQ% z_S}A7twZY-Q|bKc1n!!&K24d?wLgXRs-VXcl5e|Vq=-KpqbNF&9tGt=Hu#hdymY;*t(EFgP@}SS2 zMTh4qALzXxR`2uQszf=vDw+1ssnvAotTuka19)Us?Xc&n-=PJzj{1>>_kQ#>A@xz> zZQ3c$%;sAQNh;|%74I4JO)TF#2PLEEi#A;;u}3vgGRsrZX+)C@0+*h;a9!U!^EMf3 ze@q{$+%d*|TCw+Xu#ZXcMA0+1v?QCo;bFFy&i37hHhQ8lzi=m$ypW9Hht++9~Wz3_>g}oYNw3MAKc;)bFK$N-y zW}FA$ZZETJeG&X{N>b{>#QA$@E-d5}HBxnd9#k>EVRO-!xBo&%je2mqZv-z%!}vo* zT$KFs9+i-L$L`F4J^(y`C`Sa)2g>LkrHk(m3RxO$wT(Cz#q(ZncfFNjg9$Mm#KF#k z)%Qlem7!7{!l{okua?PVpw;guNZf9|FYl{1xYF*RabWLw?QMRc_RJgyWYd1jFrD?s z=gJ9bS&6m-pI%3Dnb}X`P~TvoZd3Qd!q(&i-W1x?R6|u>TX|Py7)C!g07J*np1aK? zOE$^_dZS(2qw)%4q59OgEIbb&!E7pk;5hYlAdg2n3x7-G@Yu-jue2h-#rlFfU(JK# z9#)&&atD{YRy9_teq0>Aet)?>Ca6tfOs(H1fi!)(LlhJgv3y7P8RKwG$B7&fB7)-Myt%q3MU0xB3L!9+=;(2|DE4U}fpVR%H{e9OXkT>SU%}p84KX@#;~Qq{ z`y?@MY&3i1u$_9#$FWJdskc(>@8NBmb0BigqjSk0va(oVW3PpF1lAuYzuMY3k)r*w z`dy&O<+jhqL6R+Q){LD!TWpa(-rrEPV-W4mI*10DGoI`$b^;F?gl^nAtNBnHc{6O+ zMeiKvks9FBLm-BEXi7vICBRmu$K`WQEoa#lzl>lcCQ22=i>)axmqBi<)+|U5!}xej zFN^1G-IDixhAuF?JljO+cYLC`{$|uq|hG6lSr-6skJ2?rr&ML_;&B-#C7^d=+ zvOFq&(`p6s0piUF;7G0H(hTXj-h!P?NXff_1-6h%#)aWtvE$m%rgw5f$sVR{7jv#u zUz71KEv!k=B&Mj2DlOiq&N}tsOphGj>Km=v(53_~{1KJGopHnctZ zeM0Jd2{MAunK?&Z$Yf>w1(%)=7PO5KCM6DDxijG6%2Jqu$U_7+O*>qo0MfRqdQh$> zA8&lK0?}o8jtiE0XqPtH_*m6FFLLAUUucV@L_fFDU++0@_e>x__kqTz75r}oh9y=k z-{p^wKEAcval6Te7Gmi-tgP38aX{7K6%Pnej~yuaBoh-AA$2G@>wJeEqqyIV5|s73 zhH?bsy5UZU=pv{BDMm%7qGOoyGoWn1fN9!|o`b}Vg{Q$65Kd9OEy&)^P#qhV-~8aT z9k$>i^;yx(hHyvw`XkkA)#Co=gVh;t@9yx}g|tdZ)Qp|jGub;h)}&2(*Y9H1$(N{* zNZz_7Cdhr|5N^gi3pW`*^^n@^u+qE)jcCOzE$uc+8$tg-a_w>e(t&$=$A{QN&cmTx6!sAQPxPq^GtHyvrGiR!ADbRoLT*l$_&S7t@n3j!1!05OLiUim+Z^5_bd-ve_y1XM z?HaUh4@2}n9SAB=S4S!K9=}77umt5WSblHh_olpJRn!sWtR$^nsvU>!H+1o^53NW= zqWGDo%J2k6^FzRU`=DLf;!Nm@h%u@V$XQdiA2M@Q58OkItvPD}d7nUXL4tKH?MzBD zI@dqfe1jF{lGd>hbBxMJNn4HT?}CMn*3V-j^PMl-&%;(;IT{HO zV4fM5Oq)UoN|m{6amdqlv-3ztM=f&PpcxLp54ekIsLFkviIMIfL|uI~01NOU)zpB38x@E{MbC`WQM()2-uyq4*J=ih|3LQUABtC`~*j{ZOt z=&Eb`f+hFaO@qb!eqoo^;@5_H#Oq0(X}JRF=+3yJqKM~LMFsbFHf2Kh+KDq13s7PV zp-4c0d2Srah4)(#k@|1_#Q_5UL<%o_b~@49PyJby-uh>{;N$EYy%xfi4k7*BxGyy_ zR`yQmX{{;abNCqc**TXOw#hh(3nD>JPa4yw$2Eu$7*;>cj<~((C=@?Kf;HP(Gr`1N z1>?Iu6ftv+Y1NLYZdKt^kplv)a8=2}_sb4(rWXN9G;7ycJAJxqh-n|y zkDtoJdAUVT40cxNLMP+|jhgjw>{bio@Iz;XCl-f$phQ;D4VVFT(AxPitz zWI~A_JLTLe~;{6PNrp(5#FSE2$8~%KpqA=H+9Ov zS^|iy&+awZAFZKoDZaX38PZxWX_JRVVu3_GN{*6X+~7m?=Bl{xo#1jp^74(A+a$aj z!5Y$14kXI}t(_@1TKb3-55&m{G&R(Sd>2-P5(-cyQ+JTZgxoR}|4tUXmSq1jRIku! zk{=%!r30?dmC17lM!^VhFp0879!u~jX{l@yN8D92h9JP+?`G^U?Z0FCgxCqQs+EiG z!>pP>VNSN&rXOAuYz@x@dBn+(11G|6Voe^fX*>uvT01%TCD*IUz+96@W#>|4^{>}tAfY$?B@OwiMh!5k-QyX>z{`7BYkZXPZ`@3O%6Yg05=G0kk+~?+esDewb(v9 zwtWPaGRY4f%ggoMg4PY6La?&Q)EW}!ugec&$No~OWVR%~IKIgm-2IMA{bVPu-6=+h%X911n%cTe-KVPMc@Y(@ZaO)_J)Eni*>NL*QBqK2&6u^ z12zcIpjqbOKnpCC75dL9Rxki-wNuzq_9?Q5t*$e_a5a z_n{MXV*|H{#+-VT>#_j_haM+~$b&mJOPog)5D{GP8La7d3;b!#g<_M}OC+|fnV`?l zp-9&^RHP2LF4-xMRi_`Q3Y65n*F+$9;KF6Wx%uWZ_>zE+X4@PR3qQMZL7@kQXw3km zx5L2%_(wQE078gLQ=@hLJ;doSVu3y`TnJZ(Sl3^#T$PFsN1Xi?IZ1Ntt?hX6Gw{P# zMdCP(ND9_;ond#dQByBbM)F3BIdM|UwFW{cj{Xg7J${G_z)}wE+;;NLZ;(hi;&y6K zn39u+BrJF4;dy|7fuL9%Fp5$}TuUGM=C!`uTCvy#PiabYN^a0tB~6nUk_C8CIy!m` zS-JY1Q_{=^=WQOO{%<8eB+KPmpCC3(ge^msW@K*E3^n?sB>Db|gPC!tuFQX8VJakN zyH&{M&@+I6k86dOU@3OxvN-l0MBRRJ>NZcbAIKjBl-`hRx@W? zEvmPp`9H0(3x%tVG-7q<_MN*ndO7FEuy&ymAsw z^mAJeiDyZeyjJ-C(Ymdx@iJR7^?UJaFaKGargA7*O>{N-Z?4ueEZ8-E{`4o^ zF8oaBIOpv`&V`P@4Ch#!+B)YLP|3_gv|E2?&J@ZBY7zBeh67XJ+f@MKHc9&qE8GCE z$i0Ha@^tR603q8`i)D5nfIzqp*52}AvFfi|J6trR6%wbE#nMiHEY=I0&M?h+zM$IW zIGdk;y|X-dP)%Nsa<4TV+uJ>-1P-(2zWX*&H*M<$7Ed)pfUF zXzmze{pNk6VL-mPZ)mRnoAKgwCKRBDO&z3APd?pJRra&#M^vTz$1Q8PKkaHnXgeOzd5JhmYZes_pthdSx?s}i zv0I^6y}v6?JvOkksVUkloQbR zO-ON~0VX8vf@cgNGPW3|(LYS~^aNkd%Hq9qDL}XPQQyg$>Zg%~OC8VH z6HJk>IQNe^1xwC+g=p6o>leGcN@b6fCiIG92F}^1#NPH6;^`j*sp}wkR7xpjhqzmG qF&}@EP&j5k^Z2Pny+B&4.3.5

    4b%2yPGCU5LxULOCq?%roDSe?h}jsq?7-+F|dQT=8r@6%g*u;2UT?tE`b;|)Tc zd5q^PE1Lchp1yCQ+BEwty>fYpx(wCI*98-xMJ!ZF?(PPjC~8kjI1wyI!zrSX zpS5%s?Xh!wAjC@A###y1BZo?7wqW|fu9f12?F0h!uTP->WRCV8|QtnW?x;6)B!Y}RW&9aeFqKPJfTs0 zdSkt0rts;!+Lz5}{uQC^S=%J4yL%$-Y99+)jX;NMZ5B{FePY#<%KsSGE8NCU;+0X& zdYM@R1E!1SA7ZIQ@+6?jrDKGg5-Buob|t8D5YzvzA7c#ebEBe(YiHK#er?nr8+83_ z112|cOi7O^3e8eLZF!V4Y0d+kXOp}~1w&!T;r%4f`ZyGhWD$j=?n9l2h%%Xb9nXF8 z8s~679keXnkyGYa1w3WmG+Oo@s?6w?y;Zyhql@$!lgSwGtmlbqMx4wTOcJksl*-~m z+1D+@)qd8((x-CyXvH^ZZNK`b!`!)u{F<-SN;wy(5^-hUgO^axmr;e}5C6^!rzo=v z^qf%Rr{sF)QWFfm<5p6?^c<=@O(R0-u)6A4yqjhMg9)0ruA-%yU&ztj)=! z0}Wl;bl9}S0ySQAAZ+^*iDy2j;`c#8360*doYZJ~gi0g1k~Mm2At&yIOXHnI)Tl%! z_l?yGwHf%nX#9~1lQm5OxvU%TictTIxMNAI;lKZmt&<;SzE^qup{>-EoRf#`;{=Yo z$1}UUVVYY?7hLc#QFJGxdxRd(tTWZ!nD;NPl&3fFU8CtC6XF^`Fr%_8jr{aRg;-lozU!X zPrq^xEw2gPjaQUp4cVa2Y45lDxkfO<_u+-KX9ASuo9>*5Bf^-_fN7tPB6Nxr@!L2) zho<(F;9%Ysm>v$4zMx$SC3cdhynJbT<1iL-t>dF55k~sO-JQ&7=L3#gx~a~6Y#xpF zaBpHMrJZX)xC~a)O2^^62KObgunjyMG<$Y|1 zP(=5(%(R0jXH<`?h#jjd;z6%5<)ICp`O$5Go^NMRr#thsQpa_)?oaWF>OGIfKFy8k zkDr8vA$WKoeIK4_^_ScbrY8?%x#N-tR(GLpO|ENv30NI}eK*}6_BJ!r>0u78K%7lp8iCH07_g&^CO?W{Q52kExgXz zbVjGpqQ6LZnl{Z(!m%(0Z|jb(N1e@)(h`Svppkoiha6{4!pwmKH-+8rp;=SzT_PNt zU^Lw)iy}lo%dc+TWYTp(>p{$2N$#6)bh768Q7Qi$KiwxA-RE-mui-VqwBEm_Y9}y1@X)^P4a=?%uuP9`j!2GMR>YLn;da|`iQhg+1XsA1x?Fa zi$-PjLStvcZ-U?>ApiLl?&%?nh`d*}dD7iMQ`v%|CgXV2pnrGLIr2J0{rvP;jg1IH z@sqZir|u$x0Gkj!Zvf10H$HxYn~dG}ahX)yjt{8(dTk!_^)qN}&1wHDIvlEd9~N!z zO2cuW`g!Z`JKA{hbk0=KCT1*N8##TM<<?o7TjlsT1R5uSXF+6;n+fMf7Un@n;&UPoe_iKi}yR!)X6|TwV4k$i0zvxk6-O! zQGJMF;=Z4H!xkk+uoP7hDE`=d6GcT z1jgf}ZgEnPT*k(&yS3q@&sMDO$C4^;;^q-YAnt!=bN$O{+V}d%i;AOR2T^Zw2p0~?M~^K*k88o> zl(lPU%+h9br%NVuy?n*!ek>XFNIlqPP+tyX)HRDk&$!X3BzjlQbYR2zH2k(oCDE`8 zl^qz|(pOrII-_c`Dpa2!0%=E4mn}OQH8{ApMpFv9_vbv)Q@;gdw!>7VF=&Dy010q}Cnk`S-lH z`t1OuBzZ;du{bouu}@QEIvK}A<~bF0H5S!69yL9Omxba^<6Sz(<6v4wt^MT+E$@a% zGS3WL)B0g8eB$;ieT)}9Vw7BTENmPNa6Go!%iRn!+QLo!M-*XcaGam8N`R?3spAfR zN6?^+y0BML3{01@3GU}rK%F6iQeO5oQ2J8IP}5Zh`g->=7#y&L>Q^5ur&P1hLUUsI ziKtC5v6D{6{!klg9`?c+WZD7cgiDS2G#3ms_v*9#h({xrGwr|cvV(pdQdRSj22>N> z=tWm?6ezm7St5s`VLGrxVR?hr@_V~r76SalVXMM5uqUuL9+{JBp=VlA4zn^$8DOvQ*wl9TpBe&pNl;`422 zsUhM(>nUR#!&r;u+?PW@bh4=OS8ReQd5#5qo-$e(Kf4tFGz^VMPKZjXX`opK-zBizM@zTdLx*cK@C>`9wO+WLhvvqv_vU_BpWEh1{@?T}bFd_` zTtunF__-ZJJDmAly*8}#nThxDw(r8xqDP;&OR)_W+~MMS#VabOU+r3uiG5%5OO?DCz-i%U6TSuG68l`~Goa zj|y>2g+^-assH>$zU|5KC+1MY{jrJeof3%Z?)4&e$N~nJZKN;oegJarpOp5MN+`RW zx}dVI4&7^~7jk!4!@}xeE*qvT(8RU;rThpJ8o{}Ikl43D=Mf04{)e6oxZ@dv&NKad zE{7<&QsVTbr7&#Dc;}-lmd_ZDG5T$h>xU>zrqg_TPNJ^!^9R@t@}efskbXt}jpzSk z<^ZjLbu8|jSnDKSJ98KYp6Cu9w!-S+Q~iUQzG_u4S8~9Sj^K(qMmFbqe2hmF>z4P2 zW(#0ef%>V{=gCI42j|tiv7J_o@1=K8c5p0O6X@ z*{Sw>sAZRve3R26qSy~?KDi0=JKKKC^rgzp!^pGH*F$a#8=jx6zOR3=M;pd1`?9mo zXTt2Q=DxXcFH}0UXYI#*gq?R6*RNK$n!sX=*rAcM478d|dH1sd!&P*hRjCuPwt!Dj z_NR6~Cc%R3bL%Ys*FfqYXEYvQL47hoYPTY<0oCgJ5su;tm_FHP%S*=ct)x;XhSGi5 z`%gLMUF`91#xvIv{Tx|n_<20tJs&M(6blu~LkxBz8~JFX!!SGjOsD^NjtkUn8BKV`F0qMcn$l3N)tXbky9tfkkuU6T?L>|IQPX*9Q;6{C zxaLUm8|XQkO|SI~>%TiYNL+YQGKXi#ZGPRzB8GC#$9ngqyCEW+5DXVMbtJfzkZVV7*;*5+`eOl^`}jJ zFdU}RT!>1o&zAnU90W6;zFx9X*a6gzBvZ@R!mwI!-{L_30!qJVtCjhy0?lmk?&tYG zIPltSitnKtl@@248_)7165&+7j~>=%lKP+^xY{5QO~6Rcr@Ba(V^?2FY<-TS+sE|h z`O{cDV_Jh)ks3eL{H|*X5XN}^lYKcYy&qQ~agSS@|6?O)(W_XiO7lWgsY&jS3@gw~ zSx)|=aSs;vJ)q<-{u?)9Jda&d9UA%6TWEfo!ZXW2rWbrt=(lUGCBPLZZd530!aXTR_(3En=fwTF{f9 z_>^Cow!ToSM8<288&$BHHXVG{q!3&8eOWlkNB+?Bm7O}C5Q|HA0bq&_VXtd35-)ikS|q=uHX zc+?EMexQl#j`w8NT5*gI&r`J@V0nu9%JcnQw{AlJb^EBX(|d6Ynk0(@0te8b{?$ZL zE3Cf`#s1cHrP^=-?sm z->u{Y6M~}FOVJ5HHSoC5MX!vecPUP&mQBG3%2TaU5dli|>ZZhXZbYi^*1qxYJYRO2 zx^ljJ1j?V??I4O9z;JtXL-Axfj%D*H-gCy!5TTYT#Bwta&0098M~Uu+raI-H*7PHI zX62Z^PsRMt0|vl|K?zROne~2iF_%Te{ZPus+{I$;F$@AH~iDEbE@N@=C&|1 z4OEclQLDn+ie_)bUU*tn1|-Ftyd)ht)cC#ZIDH8lkU3}C1pIP#R>znq@45+Dl zPUyFlLzi*)Wg9;sJe~1=>-0z(zY3nG7-@Uflkf~@lWoqII)lG4)mCfY8k*hgq^+Mi z0wjwTpIm2qJk!fZngKn#@pRdB28vH^!rZAo%_kL3u$Y+BI)gY-uY+L4vy-%Z5uQ=o ze6Q!RQs@oRpVHhfg#G>4Hii&l2%h<5>~6`2$^Sh+sHkY~MGkAM&j6Xc^VIyuDcbWA zDvqchbVuOlvU0Qz^adVgy8M# zFz@|&b7A&ODiq9o1`bn>Xpj;hA4%Jm8(rf*z~EX98-8K+##S zPw=DZ5~ALQxdSFvcF>hv)+22yfm&j^D1bs6UE zQjOA{-arMe$I7adu$T%Q&E_!{POg?EFSfi{a9Qa%SQDCjVu0Y zOh9#WaJRIPgW*fsnl(b|FyQv8C$AdAEt8G?J`($Tpx=4W_h@l3TC9l`T>=6So;cgf zr=CR9{&)!!mL(|kjBr1A><5tVzR6E+=tND5BY))tsv+0hfON0q5ll9*TyBCT2ojuJSE}f^>04@49gn?Vmy!}4L&=S3;C#Rd*!1K z3<>|e4=_K(SVvq@ISu;7A1%icauC^D*Rk)A91L*BTR3u)p{1x%L9Z_n)hNlPa77=- z(Pfz%h%j%0p-brkd)zqyl(zSBDhdF}7##EKyI4?l+^?vC;72o5T>H@1XV+~<&N%NkO?gJcS)Y)>mszD$dzIr2;vyA4q z7}`px=>ti8nc78;hlM_niFY;xYIDpR(^VSY1d(`ut@nmu`kIRwfiz4eNmMDBx!4xq3{O#)%3k8pjxE~{n-<=8-XhS zYdxT@5dGseivqOYwR4nDQB-L7T=aX5Qma$53D1;%Oe@^%1sZqhW7>8D%hyu}wd3tx z(xINl)u27n7+-@l*WtiD+Xd$ACV5WJXUJK^nEDz^`54G!t~yz zRMw8v)-J<>bI$WT_y73&)nCy^e`5TgezX1LoqCwAZ@S}+k8#Y4VDaVZM3wMv84f7hJ#?fG`(m7fY@~ zZ{eAa95y2u4`!2R`>BNGD?-Slc5E2$xrLE>ZL+iyrt%&?U+PebE?bVvKr=Tn#ObEkN)0&E4n3yJ0NI=YWz!Fg(c)j4@gigIQ^A(m4+T zOlp!P6-u*EalDU^moTP>pUjwJG%R23IjrCp3Tl!!lK)rmEWs^Xz)h*q1s~? z(A><{-!Fdg;uwQ+BCj;hnF$g`sEP$B?_eSU=6l=|61I zVvIoMVQ-vzUI+cLE}o%JPU9H%21{mo&r$JY152HROm4YizpaBt> zbF%S*xdHLt;@e-s!mNW%Hghg&W1VomVJ?GbDlLqbZO3q94a~a|@4Am7vhkS0(**2% zZQT0vEUyDU6z$=OSmbL$qyw3uye{o{Iy2`(V<(+Z*YlN4!se+!nF%HQ__{*7-i*D> zq(HVYjJIvi3^u4of15-(Ci_0XqW#@z_keXo5>!e%7%PUBw$&IHTqy_g*&p9i6EWQK z6!`CW!9uI+{EIUd-|jhx<>~qb8SCox|ASwy28UAPyP)N=eRKqs7B9dApW5b$)^3>W zwcF7#K7`0F!qkA>-x2YySzjMMa?ht>>BezRDCo|yI_o+L&G>uO`6h3Omh`pNHIc`wXu#7)(?+7CtDb%*&zA6e$T2wC=WuM)0KV?V_txOE55ABNw?R~@#jxRdTe=Ru4VGd;k6!7w_V`2t;kNQAS%>Xtq`xLxu|H`(Ja9z|av6isoIS(wvT{CW_E>3S8e~-eCbNzwphGVGH!>0T4i!5j+`kW7zt%X79siflT zvv|f>|87z@LqL5W@g^{98VuH@e5j?nPTmy4wa>svejnfZypGL~=4i+|o9 zkdFCX@gp(fLk+rUc0hOp$E6QlPw9f~5->fv)B~ob_*SgGS@_a_viIaon3y;r^QLs{ zU;H5GJ#v9o|3ig7 zpZ-qizu)Ws>A!du=Fuv~-9~u!-`$XHgw?}la=%>`YJK^?euv2A&3DvPHsl{RjCYIS z_`H^9A6n9py%$2cjs6gLG!Ob>{k9i>9d}DJ-iY5Elj`{DyCVDl@B09Wt~P6Yn5GK@ z%Mas^EC^Wr&-Z9w{laaXu=)RXT}0V6taFE%h@Ja)47{jSbcn3A)i#v;KXtqQe}3-b zEIF3-h(gp@DXSavf#?7F{n$|8spXn2sH9b=uHh;Q#u3b2?#{18tdL z%vSZ=Pmj$&9*i)1?7JJ)_EVU>tG6Hu_v5$^o5W#+n6+;bUyO!Mu1c_}iJ9h9qOSM#ZpmV#@oUTkk`QkMMNE zro1`+w7d+yTeIGXH_~7Y0B;4t3Fnx)+YWIjy3z|M0o=KB=I)Y>k zvvCjF`r_GPw(Q&z-F|U2c1F5Z^%w`5o@e&`&5;f*NlW!gWQ-p`dGq=EG}mV+8;MY} z5}?2`Zp9~Te*(0hinWih!1xBYaVqWmwqSZQD+O6>I-L$1#=Uu0b93(fr$Ai$z0g4C zfriG4hHjlrKqUD4tVW>&&y->jwo1PPd#}E~%h6gQ*k^}O$?*#{h%8p?=N{9Ax}H^E zvAui)t&09u5YkUam0}L5n(?)0CAQ*$_oqs5ql6J3*I$Jh$F`W1MsowVUWXYe{$uBB-$6$J?=sH%ED!}E@1Gu$z%guzlHKo` zfTIt~?mxPT8?_G}z4(#M4AKPhRE>`wg|%Qg-e0Whcve>5>q+i5XjYxCdW<*nzuzCn z@JU?CxFHFex?(HUqMFfA*>^wZb9q29@)Nw7dI4qP%JXgR8vPpwhqQy5G`obxt2xfI z`RSse)WhV2EKfwfR@H2pxC<5vTp1l@b)na*^OUwZmN#sYSUjyrcM3+l_87knISLDn zVgdX0yr84c_i*F^48Pm9QcdxyN-Ey?>w z&Y-_L?;X6(qyZEXBjeM}AMx~U!KdfYepsxbtUThtqr%QRE`Fv`IL2!c5C?hzw;p!(tw&*o`sux%1rAMm2APscym)$d>4%4HF+Y*@hkoq#5lo+$x;=vH z=-snuyoT3YH3Z9NjBETkUpqlX6s|@8`;jm44Ab9@v@DOIdXpD-pVe%@rK82}cj|Oz z9YD2V7Fy*EM+GyR^XRT?!{p$%M)nR*JTs&JuCD9zuyE|<>zAgE&~yDlir0;F)VuPY z>J&-wPyR+ofES4TASb~54cW2|pR zmy(9~(a#F7JXz{gxMGi*Mq1B3D5Lo|Fv?m-8Klzs_@U9gVe`~47{0wUG}vTFgBi7S zmOQn29S;?oWgY89vHt0CE$TA257OcTJVWu6n>w8WwAt}WNkmc6N>0JN)BlIFHxH-k z3)_IrWG*E_hE!4^sYF_*NRyDMk|c#RNRM@^W`*AU2CTT_Zh4m>f-nM zS1g67XDMS1F5MP}*_-Z}DjT4Fwb#Ej&*?({4KJJPlKwEUb)R4qIS6J)#Wm8$YfzhG zg-W+r9gx?KzFZfyIzDdgwDivhI1FdMp+AT_MicY5)#Bf@(G-->$^<@X@51;?7b;$|2=4hGcu9>uE)rt&xC^u|)bv?h)Pb6szAyLIon5Uz(4qLG zyX&z9>i?Z!IC9%#wfivrs(t7N&G^`=zt0bjd=7ti7wXqt$x1NCc;b53sV;83=YjQq zS#9?CJPa4rsE#et>$p++RHF6<-h3b@{Z!vSrn_oCdXmye z=e9B8(fs_~?RB~%5iLaV@c!b#@FrTn9Gs%28bOn+q;IdwBI-HYeeV7Lh*JzKo9z)} zwt)QmlU#hZO{ia7=iAlw+5a^kDC6?IgDN?q6cI@!QDD07tzpi?}&yCmiIp;J<)wp469S6jIP|4wk&2`Cvzj}whG*Ty&o2u zDo;C&sR(8zbUn*HY`=(zOuY-cPE&CEAgUtMzs zdV)_3hupsjNydp+H1=SAhaQ6u!bcviM-Nn!$=1A5kn)!F=-~_-%DbT4QV~0YXS!d- z{!A(d$an9G{*p0=!Q#u?RGKr<1an20K->U|X& z@oWABh5_YIF^_dAg`vNW8+Ww(HiW_Kjz=s4H|TZ|baibN(hO$M%b~mr?dCb;hz5ke zIvD#kK;tIy^z`2g&|6aYO+m;RX6)EP-cYXrB`??Y75^9v+7FKL*V*IPwtS1;a*BcP z1QrI=%|5++jQX3RX0yB-8A^5BWGDxn}3FN zn+_tDZ)C1Rvv>WAxHvIAXkbynrf9?xwd*bUh8?HZ!9sh$P$Opv_W#hO&=0>JVt(?) z!tYFLbs6ec(X7MIw;tvj(eQ9e$noJm9D8HfyG!avQ9f6JSQoVcEye8+j6b>)8|S#U z6x&VAj;l}MJAV3>1fp*8TQjVR)hP%2#`24^e1hSu?R%L>!mEyxApd0hpDW){XPWR6 zKC4nhbw3n;zy2P~K4JG*B;;e`KBm>3v}-@i##S4e$nQlx?V7mr;o}VLkso@7;krH{DCN2j6b~OmEun$4}L@)6tTl!qmK)vqzoNM?b!jOBZx;YlJxJp zyu7emO&$oHr_Mc~Ve^{TF}}4_K?sUnRKiA?$S5zdn&*K0S|CrDT;6zI9@~s*Rf0m;>Hbfc39HHMdS(E` zJL^hAnVtV@|DdJUB2RwCU~!uH?)_udGetnwB@8HCrRN=>{s-p5KB;CviWMtAmFNt0 zr;e2zjo%1!D!n|ziLFn>Ypm|D-_7jBQBiM(d7ov`SaONYK_fcPBWzB({a62B z=FZiv5x*SJZ)&3$`2ZQwlCO$f!u7()lj_sr`;`IOJSlBsYG^`drt)IneyG!}Gr1Wy zh5FO{sOAFN@&52T?50L zY%G}keDW9pns;5k-NkB8!9sd<3`=q85UF#V7 z`x;`4zdmJ<*Yw{f90Fcz2wI~8+331GQDQ)|c_G}~BZTL8cktAab4EaM!f)TD!UUb~ z9oM^TVB}{XYkoMqRVNl&awhp}e)aS@fZ=NU&!uiX%~bg|2&ln%OmJ-Q*b|X13z3GZTYwrk<^`8j+>f}M)yM8xrlX3^L+cp_=y&AN3_i8+Sg)PDw@{F{Ok6)5T+Gk zjTOw@V0>f!1oMqJSa4dGFU!~u`tv1RyzE&PjP~u^D;2wlc?{QT>dK?bOp=&hH@ha1 zru>lwQ3YgFX$To*~^@ zFHge2Ypbtfx&c5Uzn0z+e+6bGV$E>I$8lUSDnGXOrz4`wR6>WmK1{toQu)Wi636^^ zFZ!z%tB!fDyy97Yw0n~cSVsej=3K~B6-0J;qx_n22eDC+I zAv1iRXy&c)H%FTeL<#*VwylJLFXlhx=%B}NweBHaR62;;Id$_k416fr5)}RlC8chW zR-BwdRW5OBw=0^U0W0R~u09fIm@DJ5lQf2FJwi#(?2*Up{OZHilq-vCv3>P${`*lJ z7Juyg)@x`vDuM_!$ykXWBQTjvwY|sgfQIj{)fIS1-)E@y_-n4~y-!f_rM(vFzW+M($tm$u{6ta7&aD<(KhL0Aoi?R1wG=eio7y6I zbRNz3bQx~hza2=~5BAMlwIUMJctN0q28_z)9*%#5&`fnwtWr%9CLc?_DbS{_qk$NS zX2&<#sQcx;@b4iQzffm{`yfSN8+3dPx&6#S8BMO=Eo5eS9vbZ=TpPo$qn!Sd?+=RPN0-UX(crQy;4rs)in+;S(CBH+wJ~8^YNDck*iMj$@Yo zbd{us#dn(oTz`pWDL|%I(McU}gTB2xJcOee{JQk}Cl23ay{Rh&BXvidYXla7u${X? zlNF)qk&K)GW*6ubs()4U>L|2&m~V)R`-3LH{>JvZ5vbjOw9srnf$EH-j>vfEKoReX zLHCP1G%CMk#(JFYM}%G>xxTi7b!blM>8`}U6KKRB_Rij)4DmT^U!7{$*k~AxXtG;p zeP7&W=x@h$1#XYE7++ZTzE0T`^Z$`I6C+{N+lE(K^U~&Dv&vNS5Zo30SDOXP4 zU$rp#z`U~hqc%*xv$`oG{RxQdiAws%u0mH8lj;pUj1PzOC+o!%RV<&mm*`W!=raZs zwHs!f(#Ns>(Z(;kD4d5ScHyqbz;;-=dN5V^S~2t(r~LGG2#0YI_3p+pjDNMEus*S3 z?Z0)FII2y|!~B7DPq)5p+q(@BYEFWY^ENawevV710psIa>CIw}Wo?5Mrb&%EB`?w9 zE7=3pk5m6^J`fQqE}CazxS{EV%zcjbiSr#eNDtMFMiy7B`%2GZcT1p(Ns?| zkp4E>)utPQZ}LF??)x>US4*+wo8lCZ)}=RQ!DASsMOehOSu^yrY*3P-WzY&#pP#4J zm5(CAhJ@$E8qH{2`lx&VPCP1up<1oX1^5vzs$}BkfXH03SK??G&*DV-JH8{OY^dYe zScv8v4#%3b`Sh6^7$3!O!FyNJUlmYxkbL3Q^HQkyZqqwN@kRZ6basc;Gx+@&<;by& zrtyx`|GEDrw`ZItdJ)l>nT8${6K0ooTW|)xd`hpE1rn(^n|EwgxdknqwQ@3FwSgwq zWAILUaP{@I+PqJqKpoLOQ;wcart4Y%?3*on_6F0>O-Ne55jQT;0`(l6+E@SfVe;(+ zNtUwM1yL`Ku*C5$qO^zGlLT?8Fn-wW$0chFS7t!8W1nS6e*hzL9Nb6N;=+*~yB=tKDisw|@LkHnk)9Ru>t zc%0TFhI}CUc#``rDLJQD0QELM8W(dwWvCc%w;hb?KDAAW%}o;CdK-CzOg|9Sl5M}PmmdszW! z{&%%6-o$u}2HZqWw+(P2f{i5Z;er%&`S9JfQe6)tvge%#^W~xc?LA$gmeuWD{7VUs zuaRqEMq+>{8c2isciPtTF<5_F=ilfa$-nTQ{iFHbcfIIz0ur0{$6H0o{@49DrljOQ z3MNk(z6;V;Ty(9fBkGJ}-^64+hx+PinfU9N9#MMgf|@7`Kz4*+c)XajcZ^ zlOgtx5y^dD*@`Q@E(R8qUVKD$41Q`9T`W)^+kw@CH|U((eMO`db$q&iWn;#1m<`=% zz)n8QFs`wudPOfTPQj8H%U7>KK4`>=z1p~L7yLc3^Y+;YOy8cPv4|dP>VUZ(e-9Vd zRplIX>4{(2P1Spdq=zq|$^CR?m?E+(Rdl0~n;XJ4U>!#7D>B?rj%Ye_ZlEug6N zw1>b!`gx9Lt1>MexIBifN4&axUHX31Bzel3|MLiR`o|oQITXn7-oez89kL(;5k`wBY_essGv^^!o>T zIOcwfFw~!8@?L!Nn8tl>7>X))%=nfBD?{ENWtKgG=DntI=hQZa_8PY3TF7hmHMT1u1Ia&EUXkh2T>^E6bsL|x4C9?%) z4-yV(kBRN4pN}wk_K>Mf?f}dYulannF+hDFxZhz9#>1MK$F%Ie#~lr9l{{;_GaF4G zuvG3`s&weZFWU?^~o+z6rdmak354>R$;!Jo>^z&yc zJVbpf7ne;&>GjWObl^>y+JSU56da^zd6tgDhhO9Cwk};O5H;EA2|5K~^T&QjI6O5R z_47mhf&?UXbZa0q3f=$hl8}r z%8h9P{itLG$1{E>3woaT_>&6dP*qFZJLk7-P%^ORqgPe~B zmUrr-i3BV!r-_9?obt^A;%lQ^g`NPQYNjc0xAHUW(?REt4}L{z!AS4M41t^9&_dXk znY4@olzLSC@ArrN5Y~`4Y#4A7%{fm=8Z|qhPI6KK=j-cu7LGB^kjbw=*zwve^m7o7 zrO72)ss0?yn)_yY29_aG9E^n2o1w1M^t!02hd?mqd3EPLH>ywA^Jk6TC?dHu$>KS? zz$aN=k~NtL&GuVK9gX+`l(wBzOpyw3Ya3AB(o96vn|l1>?COz zBZb;d>tFwR|0jJM7eu@v+e;qoDww=%~>+s1Ka=-c7kgZ~txR!<%UjoT2Hk%USJ^0p}ygSRG0L zUqGP0m$C5^>x+(vG^$hZW33<>(&dr;dZiVUGkx_^h9j8V=`)#p(yFZwtwHI(U(=32 zXXw}&rJop%P=9r9H>b`v$YVXF-8IYLZ$Z5|9qnWH>2Z#K-_@lTN53=Zp(woM=)}#N zm_Cqln>qi{4Rb`W6kPL3t`Ge+7^;`6J_+g1(=2#`bx_~2l#lPMPhjs+_#mK!@te@E z_wTwJaa{<@U#n_*=(Mwc6I!Vv%AAs;<02qx{>?G3Q6ip&ZD_#1suU4*QssuN80>L0 z8g%OW%qRMLVbpj2?7{{!DBF0ulK(bVUp^+5zgg|A)BpMW$?H4&m8Pl(S(`m;E>G+N ziqT(<@HMVz_!rJfU}Ox<4Jfu>06S>lOEBKSegTb(kl0PjF~9GSPHDr9S+1K=uWOcrw(wUVI(=7D_dS5kdsv(pe=~-wF&tT? z7yhRY2%f5{mAso^dGB!A3jwVESTVwdyl&6xG z*0(mJnXYx|Ol27F`GCbPgZG6>FdWanF={2B{{D$7@#Sy#$8;gJ|Tf<7w%-Fc?hY<_L>%1KNwQu<}Lv^XQjM z&GsELn4L9K;VE#2MuVA*Sufs(eDrjT{m(MATBc`(}{ZUDSJ5s_`egGWMRjudD;_!cosUC5PhE z(hTkOIYjCfi|KPxV0fvG=zeOm7ltR_zvl=` zRc} zgWm-bDHqp?-NyXx-PgP|9Ji}Oxrg15)Za=d{VMS!&R-9;dF)}|CWgtw;%GgQljmyx zbG|!%yqY$tF+u~5^Fo6E&i9q|Tt>Odo6)3)s?3D)0Mt<4=+z2J1F@?)Z>*~ybsZ@m z*r&R>UwlmCq1V#8>kRK18BKFQ?hN`k3^XJgm6h*7?Y2oY>gWjkc`Hgm*aeW*Sev}o&4gl7Z%a5 zK%Vw*`ApPT{f9-{00$%AYKs4S$%P=l@pqhws%Yua=z?%v2@tdI+P;5l1C8xJx|4Qp zg#y;aNm7M9YT>+d%H&ElN;q>Z<;k;j)WhM`NQ-AU9~t^Fq+cgfuc?KmdV@}>&BrqQ z-fd$rbL>VP>Z@y6o(j@}>d1pTgn#P6l;g4Gp`dUm-h5c-gqt4pPscyn!G`4pllNIy zx}{)#tiKAy={GiD{)G`%nTxS2n}FI>^{6OxAGEpEw-&OB(C<%PdC_)Oy$^x;4`Jc4 zw=h|F*fHeHXVkjCQ+sa*I~s0|bp6Q9jmBs9ZoV8$|BfJ^`C-dbBaF|e?`X8ydlKXK zN|5xQ))IXI!)JmPw%|^orBorwX11& zX{Nq(ygW4AN{qL-jzjaaIL)&&B)HnaJ6>~@!H+~`3o_ZH7j4`1_o4wvo^2_y1j2>y+!&f(GPy*NKMF^QjT>_@(PYId@?) zwZ878P$8Od+AI2Nlp#M6jhb1_U*Bp6)3q~dVNJ@2;HwdQaF_*#loKw^^V0oO&=qMv zx}*9n^q*1L9+!#f;T^HjA?#~YpxcG*tE2mF)S_cGOe=RoSrN@8t^Pysns?EfqvmSk?}egii59}% z{r~po>vd`AA)zRbZ&Ucj6wLnXu?zh3O7RlvHggggx-fzsj(*T->Bxlg&f9f1*`H9y z-Xnv*y#E4ajaEm?fjua#EAd#bNHI*B43xK>#_F2L=eo11Un;|B@1{2&CRL$FnKIwK zmjSmAwWhS&M}&sK++ra|+C3j2eHZsU<;REcEw@8$Rd?fAZ=Av{TGPAANs_{lVTawrnMd4u_=bJC@G^unvZ8W4@ul4ud@T%<+nI4$$Pr>rZD05G5 zzg1d8KuG-hn!Fq)&hYt#SWWtf?$^o>}H>5FrN z*X6(Y?gCnh?~y08QS>`pA)wvB5o$9G^Y;Dtg=Ys-xx2^nP{hIA#{2@epuImckk>#J z)h|7`Mf$+S@czE|3r_^NFq{Eey0hZnUIzREhH(z6v%Z-*`5sCyrE#TqF~}D*>&tDf zyN)i;aZFbS+18HTf`-4gQKT$kn5|s4OW&UcE!m@go4Vp*urp+rZyQ72IG%a`i~R92 z42P(Z(`N1+VFg-zgUW+qZvZh$iQH+mpJ7}JZJivUO7bwGYthL@2!ZKMV!gKp=za@4 z+egbJ{*##AMa-M?HHnCU(O)Y258AS!ne>V6I7>~``a2`JwZETXe9M(ryU(N+Lce>N z>pJr!7?et_Vt3L)!(W64O7Cj{!8_p#4=?6_?XlL9cOjbrt?+J}+~syiNy~}BFXf{d zLms8~It8dnXM-2dQ6)Ub2R-tn^L1z{NVf~nIR}JV+*_jL3J~nmj|OcwL-E$8#U>K; zI1*5W_-wmgufuZ`3C-M!T!fjYLsY9@)-ZNfo?2W!3zVk@dzxg-S6fGzyMyc7l6@DE z)N*25&6eonE8guO{*sGHa_Vy#e)ea4^PCGn#HXcY2Eq!@A6I{|xsZv<3I89bSgEb}+vw^G?Q` zFwBW0TzE3L2FIez{rlQ?IhYr4k2T-D4aV(aabfanR_*_6xt}cW6yyF{Y>wQ#aXZG} zJAH<-o4d6J#aAf(_1hGTYBFVhnz3$!!kvV5XD_aruL53AQ|&)Y5TMJd@LX3NofiPl zYFCJc)6PQg;+sl|x;=QNU3)4gLrG8@`|0j%EnP1H^4#_Dvta`2=Y)+3G0kOw&hoqA}P>*`DCrn^G6_l1d%AjpI3>2rn9(kyP$T8=i z#hhq{#>@!w)5S|LCpl3+5Z4MthOJ8iIDM==g6wXSFIc>N{Ji?1Q%uZoc6M@3;K6%^$~bjF$?IZ#Vp!55LWS?8uG6?2;i379kDXKB!5x z@epB=h9%|aJHGVO^V3ny?k@rFAJX*?JX22QwimPt=wF`XOG%AHsfEuDuQ|ACd>=X} zERRFB8UZ2n(b%=1B$$5%9klQNi6eo@(`9oxFXqR1X0aLNd-t(AleuNFn|aF^zT5qE zvF9g z&l1DMH5X+L4d2wAJYwj0#CR5^+Z^>dpMmI-#CG9qDk8^<^9JT&Jon^$g`bO@PoVsQ z(k{yrxiB;-FC*YJk0PEO+k9j|1bc7hK8-`I^DxV1-{-rO3=3PHb3TsvgL;EYob}T7 z!^(Z`U$QRCF!%7QUBS(3Fcr77lM+Xln^2u*IvJF0iK5F~!|V5c!gEG#JIvOB*}1*z zqqh@IeT5mlAiUum=HD9QDx<;y7Z}V76ua)H!!UmoT!4A-z7LRU-^P0~gP!+*`a@Rs zetPmA^`(r=cqL(e?w(h-4Lk-nqP|lzp|6HQVgBh)34zyiUNuzN>*{#Hn+&v4$@9-z zRR7s={rv*M&S=WfH~nMCCq$9`dTeVE0cwBT?sRT4fZ3Bqk7hc%IQ%LnV34A!@@l9S&!oD};&f0X8Z{3n{4_a)MxC`ElbUQ` z$mGS_iwa?=URA&4mJOXZ6wfL1NzblE82S{;?zKw(29oFgv%A;1qu%t13BEO_q26pZ z@}g7oKRwo$%lI^=XLmXIN@w53zSlP2M)d=EB=qS^h(vc-HjHmhN_zSN+Yd7idph&D zn1L|X>R;Wlf*ORc4L=$QfLX0plc+ZszsF?6RpPc_OsC4(6q=Z0U*7 zhW?f-FFNLkYG&k z;ndvTC}Xud_q*>;}rtdDOzHN`>h1WBs2hlEn;$Xw+fj#ka4f&-s`ajn z`{2)GMV)}th%osm`is;*yOsIa*y3e|e3#Yr!9y3e#my=~pBdREY5yJ|EVTP&$R0;6 zgrT)66RY|&fham?+rL8RMFMim7K?RNFBs1ije6-Ub4J`7IApX==wy4ag9kVuh?hq0Ltgw5=h#FXEQld zpU~I^l+?-ZUF;1&yKa8-Y7_&W5iH94#{XtFgn?VTG%26x{#fX#yRAC4QVBINDb}9f zF+b>JQP+8?{C{@CC9ar#*{P@{<$&dFxu$>L*Nq?4TNu87$tCmUBv}WTjXHg2=cz|9 z=v~>pJx~eJI%rn$9uqLTHzcojeJc94@TW4hG!6)S83%svW55$&=xqK=% z9E8Mc{S|Q~E-3Nd*tUlXpaPRC&6~^zCSD>+Sl^?@h>;yvH<=qVNFqj=H{(@Q# zq0Wmof-|yvV8}D)i`NHC&L_Pmo}auLjlK+B=2-F=L8I35MJ_KG{4qegd_946kRhG~ z8Rk1}CeLELmV}yhvLo%x|MZ$`%2Tnwfwq^t)jW;`rmkooWeyNP6FcJG+j3%dCT+oH z9a3R|qO3g8@E6!Q+vnfBlE>Rg|DW1&Ab*3NJdh1ON@ogoBg&#qf^?oHG1c;1@*P+-fGj z`w%RfwOn7jm!6-3XU*1gRiSKyiftbyIo?d5eA~GC4?G0EPd29xe~C6j$D^!sZABSnlPJ78fKt4U?bA@s|3r}QjU4W{5+ zWYL5PfW`cg);n5wj`HTp0%cFsFrc{d%{if!QKp^Z4xb_0$-AMm@T=Vb=H5mG!;U(Qb#~p#vxS2J}ni!tU z==J|SovUruD7eg!#}Bh=d!~*@Ze_SWe6DjSKP(L8 zNl&xP8A`#VH9?#B;2BIQCdcTAh@+L=I=Uw3!AdIhWRDQb-Xz_d*sphHgj2}1`C)x&DuIF zh{?0Wt{$ob{yGpJ>UJHF=7#x}gRiX)AsCReD>Xgch$;#)Qi}31K90$)y?dN&vj54m zor`Ph^I`bCdir+B%RtkTZn94H0~#fho0_PF+7&CsCjRYHO4a0(DHa3z`{wI=$Ufr1 zFt+)LQdj#~AfFvyAE1=8+IZ;Cbu6^#zR{!Ce**0-`Apn_XD~Y|bDmZ{4^5AF&VQk9 z1ZpZ_?*cgw{*eC!@8UWSi>f=ePV7yB^!Bl#*gxN4`Dw%G4O4p<2(EAK*o*O0)^FT- z-lvz2+lm$g1y8t~E{6Vsr*O!<9A>WfO~-tZg^}*veh*^ip|{>{x5mLfOnzSHIagtH z4#%YZeAL$+!E$KafR{67A5f2!+G$^E1WLxbBV7c3XekW(Q2clY1~qF_zagC^wrc ze}@|u%|)XMzMVppy6=`&hkJmwS#QQU6!U8}U4DL}`7GV8LlnOJT?a?QAy+?Eokaa7 z4`QQlS^oUMC{LL1pH5(UFe3E7bl=&t24<9Ck3_W@P^-FAdJ;`=OoO6-Q!G4y_$>VU zYv*Us%LIqQoeto7|&dDc>J(E#&-BrY_3%;f{*`I~R6pW$ycfXa7;Y+u0wJ4=rj>JKjVZqlkSur&PM>ypb@& zcD!+;2Zn<_v-?Y=P3$b5dAlK>7H>1u`%{9X9M-~;z|$nj@?cmpI&q-2AG1%L@`&C}aZncyRr0~iT@U5NIcu2x`0eMyGsgT9>X@T}{q{=u8*;-Y zO%FR~rf#R21U)H*Me4}ick+4AAGUAQl^=^&PFL!ta?=&##q4f>y;*oakHi=CctjH1=; zFRahah6UBJb%syh0OkU` z-L1SOSBV}MhK97R!)y^`n*3*dA{t^b5?Hg{0Oh^q zW3oE*0L@u_Hu!0Hh2CDTTCxXcOdz~>TKPoqRYH^ZOrzEf7opCP9%*6eV6;@waX&-6 z3(YQWy2-l|2Bp$0zn@(AfQEV6Uu;^&_zv3Y=Qlh(pny{CE(XiF(D{o{XEEpPh3dm- zZ2s*)fRQXJS_v~53Qbnh4y;sls8{ALqDZI;odkYKO1EZ*c%V=@U!1@xvaE9t%XC$!>@iMI2_yuo2 z6*dMv@BjK2Ek7)~z&!Y$@|oeexl1jAz0rmY{hwp`CzvBY0r`aW0nW z*KnLa?ynq$9@Jj+{y^UQooI0FVAt4a99rVu;QsX5T^y^f^{&R1KxlRdQ2#t`jN_1d z;ZH-DpMGro+N(3_%81&M|MviLLd{>~-IU+B~jB|9{ta)%6?G zW_!^<#qytrI!X6l{=)2wqPka}{kJ_=o!`Hg4%D3;Qv<@)iJ;4SjnMqH?{TL$)0Z5$6m6LsSFu@yNe*D+k7<(vY-!isXQ zNe8q%93rZFm80S|f@0-bKOBoE@$so_6DZ#Fdvn-tW;B`FviXJ%hVR<0#Xj-&>L8k< z+O-*YszSH-QSaAlkDz|p$Za9w@j&7|-!%WC76u?JD$g(71$t+=#UU>fHV8r0s>HC})Nw=_-NoBQZVJOt>y5i{rQ` zu<6CIpJ+U3-PL0jrHD+6+#eqvidtAgYCDb@p&Y!jcvF}aBEP=wd%^N0np$LQ)pV1A zj>U&PR&{QutKz1{r?ONa7=DQOwfPXr=QZAXe^V#S+)0i~KAHo4o4pP`&}E0J?-yTp zeIY`x!z4clT0pOOqGP1KIS|h=Q@B2M<5^jh8{&=XpvK|ihVjG^hV`A%I8yuNw>s+F z;zQustd4rK%9^-;)6Wq+^Mm~AQ$BT=9{v3VciP8yIJTHCoz!dF5%r!A@;sk~W7>aF z=(>(Bq7_s9&)ui%p*W_D(wwtahj1)RA7pKJYvP#LW3+pUZl~i|dw!%c zo$x`!K@!TLW|-WjJwLE%rR_ReEG#zH(5!`KV-Ym|Is;~ofAW2$xd}yhIbA6~jM;s? zY6tf2+=k(Iklx(hn_V;tbKYtxb|x2KpmaC)#?$9v>eh;mnT9f=4E=bnZ~hg1a}d87 z{VEcz)N77Dzf}u)1@AWBx}k#_Uifax$vX#>;g^~gjZ~l>I<2!q+yGiW#JoFw2+ITQ zH`0BPF2M}_K!1AL~lWt!Y(SCwDgaW=)a*iP4v-9oz9l?n7wD1bQ zbpkXmui3Q4{SZC< ztMe_}7QIcK%L@AU8`w`!F9X?Neaw4hQAF@_t!hjvAA9P1XT4OmCmWu1=`fr zW}g6h-3HLq?O4NdT+!gV&C>3F1fjRcb8zwd5{z9c9@;)W3>~DIfwLcrQNuo&2X|a_ z(aN^ro2@UhV8nEr>@f*$n9wgw{F$4HdPEiuiTH4%9w$-CfvTU-$4e9LfQ>-;_=%_X zR1q3-?PcD@p9{lDxAxyT%#cs^pYuW4f1qOhIfgux|9p;U=`Gv6<`c~5364!Dd!l&) zzUNKQ3aZ^*(Mz@4jM^7Hes+Y3qglU+lqXtOpuX<*?F7yR9CQ8N4R&&Le+E!)ira3d z(c@Y;rtvAksBMh;4B_n48_uu4qsEhacE0{r{7*hEwPoVz=P-`NrQPPh9vi4XAjB6l zL(j87Q(28C8n&#Sr2@d@kP!UjAj1m7f4EV^Q+oapOg22b%zRNF4akubMwpI5 z{jZ0YJ0hDQ%VOfK%r_=&JM|Ip@y7*;3Q$9~SgHsl)`tiFG=PJoU#gvcrxHIW+_>ZM`$E62G| zHHN%A=rQ1zOMSZzO?GT?K141=)3FxK(Ou$*0-^f;9vEL&UrWK*Huk${bkq34)d(>( zzemf7u%iIR&X<2YZY_yYzP*VZFb+a-q^~C$0)3%A(PjL}c?$HMd#n|7>mCFh+fsPi zB^3=GmZqw}wiY_{rIfblM+2#+Lf)&(4VAps%5BFvp>Zqw2RB#_0L91a z3d#Nr`rFVjZcJQ@CLfm+?!~`>{P4AUTgI{xIcBSZYN-$E@32N^WAzZ>1(nOcgAEN= z_wu}wn1kZFdcAK~Y|%nd`VP+DJ7L74!QrEN4eGhx&>yz(9IBqStLPz4q5*-U676bq zKM)$abTK$?#RzCGFMc}iwh#4gICk{6vOkWy%1>-ig@CGVr}{`~97C;h{`RW7eg;O4RK$0kS3+aDA$qef$DkwX z7Pr*ID?IzIzt26kRzPE9dP`_C5fPu~MO@qZ8b&Ss#a5_$5PACMMTK`RC{fKq=*9IG z)Teyon*w?UIV9_A`w!|u_uuy0{!Md0G?tdqZ)5^eW?%iZPAxQyrCk=vpMa^rTY?7_ zyI|Pn;r8ARb(m~=nq2n70y^d+UmL|c!@$wG71i~a& zWAE|v17~3NRN-3(g45cjQGy=I)d8k9Q3deS1OK1Zh6C&0}eZ1?dg+^p>1xlv5s7L$t z8x}D#w78v+;K$=pzd-%=4dYmxxUFl$$piXbX!>EoGV6sgAe>bkU%S%;HOrJZC|cje zF-dg&%xejUraE7>)=qU8u$Yo_5w=Auf!1G++V`QTB3au3NqQUxjm)*#n{=z7p@@kn zJqIjrODV%?K0=KNQI-n~6W^(#S++O&QhW?`hNw@b^OPhl3=tOMs~(!%K$$wapL46-zU$cm z=y@gKdtapn5$6B+x0gx+b)(z6LRJPo7Rc-=<#7)-f)w$CQYP%7n0(~3&@W@RLhW6J zcd~9z!DNcgX{n6?K<#^dr><}jhyv+GE&9c1%ALcZw9^u1&%R^H;E4rNYSgq1{uJ1#?xcp0-n`)-Gy}P z0_a)fT;#LWK;!C}fw$Js#|^YE#v#;LB`Bn5@qHiKj%T~c`}{`lSyb%iw$|&z0Myuu zY5fiPr>C2(d8D`p%jbyg8`~YBAr8~r^VwOq6H)%rvGs-dSez%TMKxnmS;K$_gq`zxU4o^<7XfA#m>0*ebAyt%J-n5h z38Gv!XzP66#qeBLLA_iXM&8Yqd2$&*3<~1r*1wJ&Gc4l3OlG z;LyCf8(Tn{#J}%8Y|q5=DnGUMNIP$)$`S%U=U+yudHI>E8nm@jNM&ZzQE^=D0fL78_#wjN-Z3R zv)9o`SoAIhmli}FN5NemEYRWw1))O*bvWjad-bnOiy`8z;&(sSeZcV!Nreb1(DQ51 zvfYk@3XgQr(EaF}#VX%%+>gsgoHTTCtm0>hjh0uS`}rNV2c%CZ>0yb`t1T`tuo6T` z65WWXSM@u|hZ*wvvHkbUn8W31EF$q$MyP0F-_MZymXOUQfoQrswprr?Bfr_gaoJPC zQY5Iga%vho6^c4edtMsW^@17F^{};T=U~SDby}E*77(hW9Ap@I>{jhhN>taowTj2_ zEIQ`<-|oMIrWBNCxYiY7`!`xbbv&Dn2aIN((jpu0WujcSjju*V+)#e@++DO$68&8f z%6d2Kfryb&4?LyG4C_Q)NzT8165|srIYXI>B1og9!$D=u2}LkU;GOjSOV5W!vyWfh zZfM2)Sv>;dHkq1?b;E=KVijKxhHEmo?pk{Fjf2pzaZZ9$3d?_K7Jd3SDJU5Fk|s?p zdbhyxBcC;4k@UJpp!~d~C91R*$d!^u`L9=@$@D+-jfPkrM4`F;{Bh$eK)Kbo=br02 zdb^7?^}}&b!(i5PFD@yFj$Dl`M zM{HKxK2%;Ou&XBH3>rH0@{+c$J*wGVIG3r~i)Kt0O2?ojWuW3?SO0uj0EX)M`B5lSv`(|XYXUv*C(4ezb3!eyeWszq{HRUWaF;xuAJ4LEeaWPx7f~Zdru*_R zUhr|59`TCHiKzB8%Qvy(I;dC4RU~j_Ga8ovA+0us;kDA@({4T8^$Vq%ahR=l!{VX6 zRZ*`VZudgT`m~!=Lj!pCG@$tIx5=VVv%kHLVZwBquME@)0MX8x2K19|qBE~V%e z!8ak5^N0M6(8|FR`OmI8qNe1M15eDNp=WOWwQj#}*gm~8xu)+qomURpJBHmooW)^G z+I>#Ft^>trpL4ZZcOK@ov?DXE_QI&bMQZDNy8jwzk8ZFk+$KQ&^LcaY*?FjcyL(0b z?Pip{EuS@0Ck1MjK8N{rW4KW?xv!%-%@}XooZM`te%My%uYCGycl&Q>%Tg@jy&Q## zcM#04>HR>>diOkN8>ndNX1J9@1cuvMRl{ErwhQC)uHBjb=BYse_CK%F#=9q2QIEhn zBZomO&w~>D#DLrBIP~0KAIY?+iQ1I+g^LD!$1$B&cD6q75r&I%4|qxmK!xk3>cE*Q z)co^zkQ@ibH=C+iB)Eae7iL8*TW@ZvL8+1F2igxbp}BD-;rv<*pO+Sk|JqPVuWLg~ zMdtTC*Jz^I8SxEcH_LI%JM*VEGfCi>5>>yg6a}KOw4FHfXf;GmExhx&-yN;^HbtuM zlz_rTj*DLe+|XZ{>xz?E3edm4u>|Ht(Rg;B)g&j~zk_3I`QY@B%>|8K)8BTx?-!2c zz^H5GlVG%vTlkk>os1U3`z9?M5h4Y>{qwx-5z5*0S)q4qADRjZ{cJ}`h6aV4&$MVc z82Tb;=Xxy0B^7LyN^qy*xILDR=4ffIbzQ6ZEd|137j){o#9u7ip$meL&$VoI2cNOJeE;dj>sl);;9|%wUvAh{ z>ixv(9|i^nAK3cF0AcZC`Q!02XqlWi|amp z(9^>~Ft_|uMS&5t_$op<)zJiq5(lRaX{V#vE6SVsW9j(osMefIl}7z%w=d#5lAlBY z`HJ7vqk$oy2IsbQL{CEhv(KlTi>sjVqtV+?ReF3L#@kDh?i+oAeh*sK1B5^~d+f|DGrM`&&fU2$niwdhhgIp?hCx+hMfqbx@S+zwgQN zr0+#sH0p>NOTT~Z#;Wa*8m-@KshNggzD)<^W9ajb{x+=CPv2n7uZ`a?s5P3LjD{4$ zE6SAZ5KUG*EjcLeKc1gj;L#_6K9|ryp;unSk+H zkCi(-J-BaGzF0x{PZouI7;2Z+1@pDqfv<^?LC1;P|LPt(GzWm zTEuGzVPT(fT-y>{uc8t}IH@f^q!oxJcW>Iuad8_Oj@h{O-o7@}e7ae+r5T5)+OBy` zVwy1XJx{|uGXnj*&nk1$!4&d*npJo>F}XYLV|eBkyh1CkM2pV`|DJo>_|`xt9?gyN zmTr=zjBo`

  • 4.3.6
  • 4.3.7
    • -
  • 4.3.8
    • -
  • 4.3.9
    • -
  • 4.3.10
    • +
  • 4.3.8
    • +
  • 4.3.9
    • +
  • 4.3.10
    • +
  • 4.3.11
    • +
  • 4.3.12
  • 4.4 In-Class Exercises
  • 5 CFA @@ -371,30 +374,30 @@
  • 5.3 At-Home Exercises
  • 5.4 In-Class Exercises
  • 6 Full SEM @@ -411,23 +414,22 @@
  • 6.3 At-Home Exercises
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/efa.html b/docs/efa.html index d86721ea..1f6aac55 100644 --- a/docs/efa.html +++ b/docs/efa.html @@ -333,28 +333,31 @@
  • 4.3.5
  • 4.3.6
  • 4.3.7
    • -
  • 4.3.8
    • -
  • 4.3.9
    • -
  • 4.3.10
    • +
  • 4.3.8
    • +
  • 4.3.9
    • +
  • 4.3.10
    • +
  • 4.3.11
    • +
  • 4.3.12
  • 4.4 In-Class Exercises
  • 5 CFA @@ -371,30 +374,30 @@
  • 5.3 At-Home Exercises
  • 5.4 In-Class Exercises
  • 6 Full SEM @@ -411,23 +414,22 @@
  • 6.3 At-Home Exercises
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/elaboration-tips.html b/docs/elaboration-tips.html index c55fbf6c..8ef3b46e 100644 --- a/docs/elaboration-tips.html +++ b/docs/elaboration-tips.html @@ -333,28 +333,31 @@
  • 4.3.5
  • 4.3.6
  • 4.3.7
    • -
  • 4.3.8
    • -
  • 4.3.9
    • -
  • 4.3.10
    • +
  • 4.3.8
    • +
  • 4.3.9
    • +
  • 4.3.10
    • +
  • 4.3.11
    • +
  • 4.3.12
  • 4.4 In-Class Exercises
  • 5 CFA @@ -371,30 +374,30 @@
  • 5.3 At-Home Exercises
  • 5.4 In-Class Exercises
  • 6 Full SEM @@ -411,23 +414,22 @@
  • 6.3 At-Home Exercises
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/full-sem.html b/docs/full-sem.html index fd7fb573..0d2670ee 100644 --- a/docs/full-sem.html +++ b/docs/full-sem.html @@ -333,28 +333,31 @@
  • 4.3.5
  • 4.3.6
  • 4.3.7
    • -
  • 4.3.8
    • -
  • 4.3.9
    • -
  • 4.3.10
    • +
  • 4.3.8
    • +
  • 4.3.9
    • +
  • 4.3.10
    • +
  • 4.3.11
    • +
  • 4.3.12
  • 4.4 In-Class Exercises
  • 5 CFA @@ -371,30 +374,30 @@
  • 5.3 At-Home Exercises
  • 5.4 In-Class Exercises
  • 6 Full SEM @@ -411,23 +414,22 @@
  • 6.3 At-Home Exercises
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/grading-1.html b/docs/grading-1.html index a1099f31..7013f8f1 100644 --- a/docs/grading-1.html +++ b/docs/grading-1.html @@ -333,28 +333,31 @@
  • 4.3.5
  • 4.3.6
  • 4.3.7
    • -
  • 4.3.8
    • -
  • 4.3.9
    • -
  • 4.3.10
    • +
  • 4.3.8
    • +
  • 4.3.9
    • +
  • 4.3.10
    • +
  • 4.3.11
    • +
  • 4.3.12
  • 4.4 In-Class Exercises
  • 5 CFA @@ -371,30 +374,30 @@
  • 5.3 At-Home Exercises
  • 5.4 In-Class Exercises
  • 6 Full SEM @@ -411,23 +414,22 @@
  • 6.3 At-Home Exercises
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/grading.html b/docs/grading.html index 474fe11b..f695584e 100644 --- a/docs/grading.html +++ b/docs/grading.html @@ -333,28 +333,31 @@
  • 4.3.5
  • 4.3.6
  • 4.3.7
    • -
  • 4.3.8
    • -
  • 4.3.9
    • -
  • 4.3.10
    • +
  • 4.3.8
    • +
  • 4.3.9
    • +
  • 4.3.10
    • +
  • 4.3.11
    • +
  • 4.3.12
  • 4.4 In-Class Exercises
  • 5 CFA @@ -371,30 +374,30 @@
  • 5.3 At-Home Exercises
  • 5.4 In-Class Exercises
  • 6 Full SEM @@ -411,23 +414,22 @@
  • 6.3 At-Home Exercises
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/in-class-exercises-1.html b/docs/in-class-exercises-1.html index 0a21871b..76983644 100644 --- a/docs/in-class-exercises-1.html +++ b/docs/in-class-exercises-1.html @@ -333,28 +333,31 @@
  • 4.3.5
  • 4.3.6
  • 4.3.7
    • -
  • 4.3.8
    • -
  • 4.3.9
    • -
  • 4.3.10
    • +
  • 4.3.8
    • +
  • 4.3.9
    • +
  • 4.3.10
    • +
  • 4.3.11
    • +
  • 4.3.12
  • 4.4 In-Class Exercises
  • 5 CFA @@ -371,30 +374,30 @@
  • 5.3 At-Home Exercises
  • 5.4 In-Class Exercises
  • 6 Full SEM @@ -411,23 +414,22 @@
  • 6.3 At-Home Exercises
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/in-class-exercises-2.html b/docs/in-class-exercises-2.html index 03d7e89f..dde403ae 100644 --- a/docs/in-class-exercises-2.html +++ b/docs/in-class-exercises-2.html @@ -333,28 +333,31 @@
  • 4.3.5
  • 4.3.6
  • 4.3.7
    • -
  • 4.3.8
    • -
  • 4.3.9
    • -
  • 4.3.10
    • +
  • 4.3.8
    • +
  • 4.3.9
    • +
  • 4.3.10
    • +
  • 4.3.11
    • +
  • 4.3.12
  • 4.4 In-Class Exercises
  • 5 CFA @@ -371,30 +374,30 @@
  • 5.3 At-Home Exercises
  • 5.4 In-Class Exercises
  • 6 Full SEM @@ -411,23 +414,22 @@
  • 6.3 At-Home Exercises
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models diff --git a/docs/in-class-exercises-3.html b/docs/in-class-exercises-3.html index 64f47b12..b95b394e 100644 --- a/docs/in-class-exercises-3.html +++ b/docs/in-class-exercises-3.html @@ -333,28 +333,31 @@
  • 4.3.5
  • 4.3.6
  • 4.3.7
    • -
  • 4.3.8
    • -
  • 4.3.9
    • -
  • 4.3.10
    • +
  • 4.3.8
    • +
  • 4.3.9
    • +
  • 4.3.10
    • +
  • 4.3.11
    • +
  • 4.3.12
  • 4.4 In-Class Exercises
  • 5 CFA @@ -371,30 +374,30 @@
  • 5.3 At-Home Exercises
  • 5.4 In-Class Exercises
  • 6 Full SEM @@ -411,23 +414,22 @@
  • 6.3 At-Home Exercises
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models @@ -465,22 +467,22 @@

    4.4 In-Class Exercises +

    4.4.1

    -

    Load the ESSround1-b.csv dataset.

    +

    Load the ess_round1.rds dataset.

      -
    • These are the same data that you analyzed for the At-Home Exercises, but the -processing/recoding that you should have done for those exercises has already -been (mostly) implemented.
      • -
    • It will be helpful to convert cntry back into a factor.
      • +
    • These are the data that we saved after the data processing in the At-Home +Exercises.
    -
    library(dplyr)
-
-ess <- read.csv("ESSround1-b.csv") %>%
-  mutate(cntry = factor(cntry))
    +
    + +Click to show code + +
    ess <- readRDS("ess_round1.rds")
    +
    -
    +

    4.4.2

    Kestilä (2006) claimed that running a PCA is a good way to test if the questions in the ESS measure attitudes towards immigration and trust in politics.

    @@ -494,13 +496,12 @@

    4.4.2

    Hopefully not. PCA is not a method for estimating latent measurement structure; PCA is a dimension reduction technique that tries to summarize a set of data -with a smaller set of component scores.

    -

    If we really want to estimate the factor structure underlying a set of observed -variables, we should use EFA.

    +with a smaller set of component scores. If we really want to estimate the factor +structure underlying a set of observed variables, we should use EFA.

    -
    +

    4.4.3

    Suppose you had to construct the trust in politics and attitude towards immigration scales described by Kestilä (2006) based on the theory and @@ -540,27 +541,24 @@

    4.4.3

    -
    +

    4.4.4

    Estimate the number of latent factors underlying the Trust items based on the eigenvalues, the scree plot, and parallel analysis.

    • How many factors are suggested by each method?
    -

    Hint: You can create a scree plot by supplying the vector of eigenvalues to -the qplot() function from the ggplot2 package and applying the geom_path() -geometry.

    1. Eigenvalue estimation

    Click to show code -
    ## Load the psych package:
-library(psych)
-
-## Run a trivial EFA on the 'trust' items
-efa_trust0 <- select(ess, trstlgl:impcntr) %>%
-  fa(nfactors = 1, rotate = "none")
    +
    ## Load the psych package:
+library(psych)
+
+## Run a trivial EFA on the 'trust' items
+efa_trust0 <- select(ess, trstlgl:trstplt) %>%
+  fa(nfactors = 1, rotate = "none")
    Click for explanation (EFA) @@ -572,10 +570,10 @@

    4.4.4

  • We also don’t care about interpretable solutions, so we don’t need rotation.

    • -
    ## View the estimated eigenvalues:
-round(efa_trust0$values, digits = 3)
    -
    ##  [1]  4.555  3.054  0.411  0.134  0.071 -0.119 -0.177 -0.258 -0.280 -0.312
-## [11] -0.344 -0.406 -0.412 -0.423 -0.444 -0.492
    +
    ## View the estimated eigenvalues:
+round(efa_trust0$values, digits = 3)
    +
    ##  [1]  4.980  0.716  0.482  0.165  0.069  0.014 -0.066 -0.092 -0.182 -0.207
+## [11] -0.284 -0.296 -0.319
    Click for explanation (eigenvalue extraction) @@ -600,19 +598,19 @@

    4.4.4

    Given a vector of estimated eigenvalues, we can create a scree plot using ggplot() and the geom_line() or geom_path() geometry.

    -
    library(ggplot2)
-library(magrittr)
-
-efa_trust0 %$% data.frame(y = values, x = 1:length(values)) %>%
-  ggplot(aes(x, y)) + 
-  geom_line() +
-  xlab("No. of Factors") +
-  ylab("Eigenvalues")
    -

    +
    library(ggplot2)
+library(magrittr)
+
+efa_trust0 %$% data.frame(y = values, x = 1:length(values)) %>%
+  ggplot(aes(x, y)) + 
+  geom_line() +
+  xlab("No. of Factors") +
+  ylab("Eigenvalues")
    +

    We can also use the psych::scree() function to create a scree plot directly from the data.

    -
    select(ess, trstlgl:impcntr) %>% scree(pc = FALSE)
    -

    +
    select(ess, trstlgl:trstplt) %>% scree(pc = FALSE)
    +

    Click for explanation (scree plot) @@ -629,51 +627,51 @@

    4.4.4

    information about the number of factors. We’ll use the psych::fa.parallel() function to implement parallel analysis.

      -
    • Parallel analysis relies on randomly simulated/permuted data, so we should -set a seed to make sure our results are reproducible.
      • -
    • We can set the fa = "fa" option to get only the results for EFA.
      • +

    • Parallel analysis relies on randomly simulated/permuted data, so we should +set a seed to make sure our results are reproducible.

      • +

    • We can set the fa = "fa" option to get only the results for EFA.

      +
    -

    3. Parallel Analysis

    Click to show code -
    ## Set the random number seed:
-set.seed(235711)
-
-## Run the parallel analysis:
-pa_trust <- select(ess, trstlgl:impcntr) %>%
-     fa.parallel(fa = "fa")
    -

    -
    ## Parallel analysis suggests that the number of factors =  5  and the number of components =  NA
    +
    ## Set the random number seed:
+set.seed(235711)
+
+## Run the parallel analysis:
+pa_trust <- select(ess, trstlgl:trstplt) %>%
+     fa.parallel(fa = "fa")
    +

    +
    ## Parallel analysis suggests that the number of factors =  6  and the number of components =  NA
    Click for explanation -

    The results of the parallel analysis suggest 5 factors.

    +

    The results of the parallel analysis suggest 6 factors.

    If you’ve been paying close attention, you may have noticed that we need to compute the eigenvalues from the original data to run parallel analysis. Hence, we don’t actually need to run a separate EFA to estimate the eigenvalues.

    -
    ## View the eigenvalues estimated during the parallel analysis:
-pa_trust$fa.values
    -
    ##  [1]  4.55517136  3.05362593  0.41073627  0.13364902  0.07116952 -0.11909803
-##  [7] -0.17729996 -0.25838504 -0.28014401 -0.31230246 -0.34437047 -0.40604399
-## [13] -0.41192101 -0.42303666 -0.44425206 -0.49233019
    -
    ## Compare to the version from the EFA:
-pa_trust$fa.values - efa_trust0$values
    -
    ##  [1] 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
    -
    ## Recreate the scree plot from above:
-pa_trust %$% data.frame(y = fa.values, x = 1:length(fa.values)) %>%
-  ggplot(aes(x, y)) + 
-  geom_line() +
-  xlab("No. of Factors") +
-  ylab("Eigenvalues")
    -

    +
    ## View the eigenvalues estimated during the parallel analysis:
+pa_trust$fa.values
    +
    ##  [1]  4.97995262  0.71644127  0.48201040  0.16517645  0.06885820  0.01422241
+##  [7] -0.06606777 -0.09225113 -0.18231333 -0.20740917 -0.28415857 -0.29573407
+## [13] -0.31877470
    +
    ## Compare to the version from the EFA:
+pa_trust$fa.values - efa_trust0$values
    +
    ##  [1] 0 0 0 0 0 0 0 0 0 0 0 0 0
    +
    ## Recreate the scree plot from above:
+pa_trust %$% data.frame(y = fa.values, x = 1:length(fa.values)) %>%
+  ggplot(aes(x, y)) + 
+  geom_line() +
+  xlab("No. of Factors") +
+  ylab("Eigenvalues")
    +

    Of course, we also see the same scree plot printed as part of the parallel analysis. So, there’s really no reason to create a separate scree plot, at all, if we’re doing parallel analysis.

    @@ -689,14 +687,14 @@

    4.4.4

    have narrowed the range.

    • Based on the scree plot and parallel analysis, we should consider solutions -for 3 to 5 factors.
      • +for 3 to 6 factors.
    • We need to examine the factor loadings to see which solution makes the most substantive sense.
    -
    +

    4.4.5

    Do the same analysis for the attitudes toward immigration items.

    @@ -705,18 +703,18 @@

    4.4.5

    This time, we’ll start by running the parallel analysis and get the eigenvalues and scree plot from psych::fa.parallel().

    -
    ## Set the seed:
-set.seed(235711)
-
-## Run parallel analysis on the 'attitudes' items:
-pa_att <- select(ess, imsmetn:rfgbfml) %>% fa.parallel(fa = "fa")
    -

    +
    ## Set the seed:
+set.seed(235711)
+
+## Run parallel analysis on the 'attitudes' items:
+pa_att <- select(ess, imsmetn:rfgbfml) %>% fa.parallel(fa = "fa")
    +

    ## Parallel analysis suggests that the number of factors =  7  and the number of components =  NA
    -
    ## Check the eigenvalues:
-round(pa_att$fa.values, digits = 3)
    -
    ##  [1]  7.841  1.474  0.748  0.534  0.327  0.152  0.123  0.025 -0.033 -0.064
-## [11] -0.081 -0.098 -0.121 -0.137 -0.145 -0.179 -0.196 -0.204 -0.214 -0.222
-## [21] -0.246 -0.253 -0.331 -0.425 -0.432
    +
    ## Check the eigenvalues:
+round(pa_att$fa.values, digits = 3)
    +
    ##  [1]  7.895  1.449  0.734  0.533  0.313  0.156  0.121  0.019 -0.001 -0.064
+## [11] -0.083 -0.103 -0.119 -0.131 -0.150 -0.175 -0.185 -0.200 -0.212 -0.233
+## [21] -0.239 -0.247 -0.334 -0.422 -0.427
    Click for explanation @@ -735,11 +733,11 @@

    4.4.5

    Based on the scree plot and parallel analysis, it seems reasonable to consider solutions for 3 to 7 factors.

      -
    • Again, we need to check the substantive interpretation to choose the most -reasonable solution.
      • +

    • Again, we need to check the substantive interpretation to choose the most +reasonable solution.

      +
    -

    To evaluate the substantive interpretability of the different solutions, we need to estimate the full EFA models for each candidate number of factors. We then @@ -768,37 +766,37 @@

    4.4.6

    Click to show code -
    ## Define an empty list to hold all of our fitted EFA objects:
-efa_trust <- list()
-
-## Loop through the interesting numbers of factors and estimate an EFA for each:
-for(i in 3:6) 
-  efa_trust[[as.character(i)]] <- ess %>% 
-    select(trstlgl:impcntr) %>% 
-    fa(nfactors = i,          # Number of factors = Loop index
-       rotate   = "promax",   # Oblique rotation
-       scores   = "Bartlett") # Estimate factor scores with WLS
    +
    ## Define an empty list to hold all of our fitted EFA objects:
+efa_trust <- list()
+
+## Loop through the interesting numbers of factors and estimate an EFA for each:
+for(i in 3:6) 
+  efa_trust[[as.character(i)]] <- ess %>% 
+    select(trstlgl:trstplt) %>% 
+    fa(nfactors = i,          # Number of factors = Loop index
+       rotate   = "promax",   # Oblique rotation
+       scores   = "Bartlett") # Estimate factor scores with WLS
    -
    +

    4.4.7

    Repeat the above analysis for the attitudes items.

    Click to show code -
    efa_att <- list()
-for(i in 3:7)
-  efa_att[[as.character(i)]] <- ess %>%
-    select(imsmetn:rfgbfml) %>%
-    fa(nfactors = i,         
-       rotate   = "promax",  
-       scores   = "Bartlett")
    +
    efa_att <- list()
+for(i in 3:7)
+  efa_att[[as.character(i)]] <- ess %>%
+    select(imsmetn:rfgbfml) %>%
+    fa(nfactors = i,         
+       rotate   = "promax",  
+       scores   = "Bartlett")
    -
    +

    4.4.8

    Compare the factor loading matrices from the models estimated from the Trust items, and select the best solution.

    @@ -812,103 +810,91 @@

    4.4.8

    Click to show code -
    for(x in efa_trust) print(x$loadings)
    +
    for(x in efa_trust) print(x$loadings)
    ## 
 ## Loadings:
-##         MR1    MR3    MR2   
-## trstlgl         0.683       
-## trstplc         0.600       
-## trstplt         0.594  0.198
-## trstep          0.725       
-## trstun          0.738       
-## stfeco                 0.748
-## stfgov          0.109  0.638
-## stfdem          0.253  0.509
-## stfedu                 0.644
-## stfhlth                0.628
-## imsmetn  0.811              
-## imdfetn  0.875              
-## eimrcnt  0.768              
-## eimpcnt  0.878              
-## imrcntr  0.808              
-## impcntr  0.870              
+##         MR3    MR2    MR1   
+## trstlgl  0.839        -0.115
+## trstplc  0.763        -0.218
+## trstun   0.579         0.161
+## trstep   0.554         0.198
+## trstprl  0.444         0.342
+## stfhlth         0.656 -0.125
+## stfedu          0.695 -0.157
+## stfeco  -0.102  0.704  0.146
+## stfgov          0.593  0.226
+## stfdem   0.183  0.476  0.150
+## pltinvt                0.813
+## pltcare                0.808
+## trstplt  0.330         0.526
 ## 
-##                  MR1   MR3   MR2
-## SS loadings    4.214 2.339 2.098
-## Proportion Var 0.263 0.146 0.131
-## Cumulative Var 0.263 0.410 0.541
+##                  MR3   MR2   MR1
+## SS loadings    2.299 2.016 1.970
+## Proportion Var 0.177 0.155 0.152
+## Cumulative Var 0.177 0.332 0.483
 ## 
 ## Loadings:
-##         MR1    MR3    MR2    MR4   
-## trstlgl         0.707              
-## trstplc  0.113  0.620              
-## trstplt         0.580  0.211       
-## trstep          0.709              
-## trstun          0.726              
-## stfeco                 0.758       
-## stfgov                 0.651       
-## stfdem          0.253  0.510       
-## stfedu                 0.639       
-## stfhlth                0.622       
-## imsmetn  0.705                0.149
-## imdfetn  0.908                     
-## eimrcnt  0.213                0.864
-## eimpcnt  0.942                     
-## imrcntr  0.447                0.515
-## impcntr  0.984               -0.110
+##         MR2    MR1    MR4    MR3   
+## trstlgl                       0.797
+## trstplc                       0.725
+## trstun          0.656         0.113
+## trstep          1.003 -0.175       
+## trstprl  0.121  0.455  0.200  0.112
+## stfhlth  0.663        -0.106       
+## stfedu   0.704        -0.110  0.100
+## stfeco   0.729                     
+## stfgov   0.631  0.175        -0.149
+## stfdem   0.501  0.107  0.115       
+## pltinvt                0.855       
+## pltcare        -0.103  0.863       
+## trstplt         0.479  0.340       
 ## 
-##                  MR1   MR3   MR2   MR4
-## SS loadings    3.458 2.334 2.107 1.061
-## Proportion Var 0.216 0.146 0.132 0.066
-## Cumulative Var 0.216 0.362 0.494 0.560
+##                  MR2   MR1   MR4   MR3
+## SS loadings    2.161 1.952 1.722 1.239
+## Proportion Var 0.166 0.150 0.132 0.095
+## Cumulative Var 0.166 0.316 0.449 0.544
 ## 
 ## Loadings:
-##         MR1    MR2    MR3    MR4    MR5   
-## trstlgl                0.165         0.677
-## trstplc                              0.725
-## trstplt         0.202  0.631              
-## trstep                 0.810              
-## trstun                 0.653         0.128
-## stfeco          0.736               -0.108
-## stfgov          0.643  0.272        -0.174
-## stfdem          0.511  0.203              
-## stfedu          0.668 -0.149         0.134
-## stfhlth         0.637 -0.156              
-## imsmetn  0.707                0.144       
-## imdfetn  0.912                            
-## eimrcnt  0.166                0.896       
-## eimpcnt  0.952                            
-## imrcntr  0.421                0.531       
-## impcntr  0.995               -0.124       
+##         MR1    MR4    MR5    MR3    MR2   
+## trstlgl                       0.935       
+## trstplc                       0.810       
+## trstun          0.505         0.168       
+## trstep  -0.138  1.128 -0.108 -0.154       
+## trstprl  0.359  0.250  0.140  0.201 -0.104
+## stfhlth                              0.557
+## stfedu                               0.752
+## stfeco   0.710 -0.118                0.172
+## stfgov   0.973               -0.132       
+## stfdem   0.556                0.153       
+## pltinvt                0.882              
+## pltcare                0.855              
+## trstplt  0.288  0.308  0.313              
 ## 
-##                  MR1   MR2   MR3   MR4   MR5
-## SS loadings    3.438 2.123 1.678 1.129 1.071
-## Proportion Var 0.215 0.133 0.105 0.071 0.067
-## Cumulative Var 0.215 0.348 0.452 0.523 0.590
+##                  MR1   MR4   MR5   MR3   MR2
+## SS loadings    2.019 1.716 1.655 1.674 0.936
+## Proportion Var 0.155 0.132 0.127 0.129 0.072
+## Cumulative Var 0.155 0.287 0.415 0.543 0.615
 ## 
 ## Loadings:
-##         MR1    MR2    MR3    MR4    MR5    MR6   
-## trstlgl                       0.826              
-## trstplc                       0.851              
-## trstplt         0.432  0.353  0.110              
-## trstep         -0.153  1.092 -0.155              
-## trstun                 0.470  0.191              
-## stfeco          0.704 -0.103                0.155
-## stfgov          0.963        -0.123              
-## stfdem          0.545         0.156              
-## stfedu                                      0.653
-## stfhlth                                     0.666
-## imsmetn  0.708                       0.143       
-## imdfetn  0.915                                   
-## eimrcnt  0.168                       0.890       
-## eimpcnt  0.955                                   
-## imrcntr  0.423                       0.529       
-## impcntr  0.998                      -0.122       
+##         MR5    MR1    MR4    MR3    MR2    MR6   
+## trstlgl                0.980                     
+## trstplc                0.655                     
+## trstun                                      0.911
+## trstep  -0.116  0.739                       0.163
+## trstprl  0.197  0.577  0.138                     
+## stfhlth                              0.614       
+## stfedu                               0.771       
+## stfeco   0.689 -0.123                0.144       
+## stfgov   0.891                                   
+## stfdem   0.513         0.144                     
+## pltinvt                       0.816              
+## pltcare                       0.778              
+## trstplt         0.706         0.193              
 ## 
-##                  MR1   MR2   MR3   MR4   MR5   MR6
-## SS loadings    3.462 1.948 1.563 1.529 1.114 0.921
-## Proportion Var 0.216 0.122 0.098 0.096 0.070 0.058
-## Cumulative Var 0.216 0.338 0.436 0.531 0.601 0.659
    +## MR5 MR1 MR4 MR3 MR2 MR6 +## SS loadings 1.606 1.417 1.442 1.327 1.014 0.879 +## Proportion Var 0.124 0.109 0.111 0.102 0.078 0.068 +## Cumulative Var 0.124 0.233 0.343 0.446 0.524 0.591
    Click for explanation @@ -958,204 +944,204 @@

    4.4.9

    Click to show code -
    for(x in efa_att) print(x$loadings)
    +
    for(x in efa_att) print(x$loadings)
    ## 
 ## Loadings:
 ##         MR1    MR2    MR3   
-## imsmetn         0.801       
-## imdfetn         0.756  0.111
-## eimrcnt         0.837       
-## eimpcnt         0.813       
-## imrcntr         0.854       
-## impcntr         0.771       
-## qfimchr  0.233         0.852
-## qfimwht  0.135         0.712
-## imwgdwn  0.308        -0.163
-## imhecop  0.380        -0.143
-## imtcjob  0.615              
-## imbleco  0.691              
-## imbgeco  0.681              
-## imueclt  0.558        -0.209
-## imwbcnt  0.731              
-## imwbcrm  0.634              
-## imrsprc -0.487        -0.119
-## pplstrd  0.259        -0.413
-## vrtrlg  -0.261         0.274
-## shrrfg   0.529        -0.105
-## rfgawrk -0.368              
-## gvrfgap -0.615        -0.151
-## rfgfrpc  0.449              
-## rfggvfn -0.478              
-## rfgbfml -0.541              
+## imsmetn         0.802       
+## imdfetn         0.754  0.106
+## eimrcnt         0.843       
+## eimpcnt         0.814       
+## imrcntr         0.857       
+## impcntr         0.769       
+## qfimchr  0.235         0.858
+## qfimwht  0.132         0.719
+## imwgdwn  0.293        -0.181
+## imhecop  0.371        -0.162
+## imtcjob  0.619              
+## imbleco  0.702              
+## imbgeco  0.687              
+## imueclt  0.561        -0.207
+## imwbcnt  0.732              
+## imwbcrm  0.637              
+## imrsprc -0.494        -0.125
+## pplstrd  0.249        -0.413
+## vrtrlg  -0.275         0.240
+## shrrfg   0.514        -0.111
+## rfgawrk -0.386              
+## gvrfgap -0.601        -0.148
+## rfgfrpc  0.432              
+## rfggvfn -0.489              
+## rfgbfml -0.545              
 ## 
 ##                  MR1   MR2   MR3
-## SS loadings    4.811 3.943 1.667
-## Proportion Var 0.192 0.158 0.067
-## Cumulative Var 0.192 0.350 0.417
+## SS loadings    4.819 3.950 1.683
+## Proportion Var 0.193 0.158 0.067
+## Cumulative Var 0.193 0.351 0.418
 ## 
 ## Loadings:
-##         MR2    MR1    MR4    MR3   
+##         MR2    MR4    MR1    MR3   
 ## imsmetn  0.788                     
-## imdfetn  0.734         0.144  0.115
-## eimrcnt  0.854        -0.146       
-## eimpcnt  0.788         0.161       
-## imrcntr  0.861                     
-## impcntr  0.744         0.191       
-## qfimchr               -0.141  0.859
-## qfimwht                       0.735
-## imwgdwn         0.601  0.237       
-## imhecop         0.656  0.210       
-## imtcjob         0.671         0.141
-## imbleco         0.607 -0.157  0.156
-## imbgeco         0.635 -0.123       
-## imueclt         0.368 -0.241 -0.187
-## imwbcnt         0.536 -0.262       
-## imwbcrm         0.430 -0.257       
-## imrsprc                0.605       
-## pplstrd         0.219        -0.396
-## vrtrlg                 0.220  0.295
-## shrrfg          0.300 -0.277       
-## rfgawrk                0.447       
-## gvrfgap                0.765       
-## rfgfrpc         0.228 -0.263       
-## rfggvfn                0.483       
-## rfgbfml                0.635       
+## imdfetn  0.731         0.153  0.110
+## eimrcnt  0.855        -0.143       
+## eimpcnt  0.790         0.165       
+## imrcntr  0.860                     
+## impcntr  0.743         0.182       
+## qfimchr               -0.122  0.853
+## qfimwht                       0.723
+## imwgdwn         0.638  0.264       
+## imhecop         0.680  0.217       
+## imtcjob         0.633         0.136
+## imbleco         0.563 -0.212  0.153
+## imbgeco         0.604 -0.168       
+## imueclt         0.392 -0.236 -0.168
+## imwbcnt         0.526 -0.282       
+## imwbcrm         0.397 -0.292       
+## imrsprc                0.616       
+## pplstrd         0.231        -0.378
+## vrtrlg                 0.279  0.264
+## shrrfg          0.299 -0.271       
+## rfgawrk                0.452       
+## gvrfgap         0.123  0.774       
+## rfgfrpc         0.193 -0.281       
+## rfggvfn                0.467       
+## rfgbfml                0.619       
 ## 
-##                  MR2   MR1   MR4   MR3
-## SS loadings    3.831 2.867 2.460 1.671
-## Proportion Var 0.153 0.115 0.098 0.067
-## Cumulative Var 0.153 0.268 0.366 0.433
+##                  MR2   MR4   MR1   MR3
+## SS loadings    3.828 2.778 2.570 1.602
+## Proportion Var 0.153 0.111 0.103 0.064
+## Cumulative Var 0.153 0.264 0.367 0.431
 ## 
 ## Loadings:
 ##         MR2    MR1    MR5    MR3    MR4   
-## imsmetn  0.794                            
-## imdfetn  0.733         0.156  0.117       
-## eimrcnt  0.905 -0.139 -0.229              
-## eimpcnt  0.781  0.110  0.197              
-## imrcntr  0.908 -0.117 -0.179              
-## impcntr  0.734  0.123  0.231              
-## qfimchr         0.114 -0.156  0.864       
-## qfimwht         0.165         0.735       
-## imwgdwn                              0.717
-## imhecop                              0.695
-## imtcjob         0.532         0.129  0.207
-## imbleco         0.691         0.142       
-## imbgeco         0.793                     
-## imueclt         0.560        -0.211       
-## imwbcnt         0.711                     
-## imwbcrm         0.557 -0.110              
-## imrsprc                0.610              
-## pplstrd         0.224        -0.406       
-## vrtrlg         -0.222  0.102  0.308  0.103
-## shrrfg          0.228 -0.266         0.121
-## rfgawrk                0.450              
-## gvrfgap                0.778              
-## rfgfrpc               -0.317         0.165
-## rfggvfn                0.503              
-## rfgbfml        -0.110  0.562              
+## imsmetn  0.792                            
+## imdfetn  0.728         0.169  0.113       
+## eimrcnt  0.910 -0.150 -0.237              
+## eimpcnt  0.779  0.126  0.213              
+## imrcntr  0.910 -0.128 -0.187              
+## impcntr  0.731  0.131  0.236              
+## qfimchr         0.109 -0.156  0.882       
+## qfimwht         0.139         0.736       
+## imwgdwn                              0.740
+## imhecop                              0.700
+## imtcjob         0.543         0.124  0.182
+## imbleco         0.682         0.135       
+## imbgeco         0.799                     
+## imueclt         0.572        -0.202       
+## imwbcnt         0.712                     
+## imwbcrm         0.545 -0.124              
+## imrsprc                0.620              
+## pplstrd         0.207        -0.396       
+## vrtrlg         -0.198  0.151  0.285  0.116
+## shrrfg          0.208 -0.263         0.139
+## rfgawrk                0.457              
+## gvrfgap                0.783              
+## rfgfrpc               -0.338         0.156
+## rfggvfn                0.477              
+## rfgbfml        -0.125  0.538              
 ## 
 ##                  MR2   MR1   MR5   MR3   MR4
-## SS loadings    3.974 2.799 2.193 1.693 1.131
-## Proportion Var 0.159 0.112 0.088 0.068 0.045
-## Cumulative Var 0.159 0.271 0.359 0.426 0.472
+## SS loadings    3.970 2.790 2.215 1.693 1.166
+## Proportion Var 0.159 0.112 0.089 0.068 0.047
+## Cumulative Var 0.159 0.270 0.359 0.427 0.473
 ## 
 ## Loadings:
 ##         MR2    MR1    MR6    MR3    MR5    MR4   
-## imsmetn  0.696                       0.175       
-## imdfetn  0.834                                   
-## eimrcnt  0.236                       0.868       
-## eimpcnt  0.943                                   
-## imrcntr  0.447                       0.526       
-## impcntr  0.956                                   
-## qfimchr         0.140 -0.120  0.858              
-## qfimwht         0.174         0.723              
-## imwgdwn                                     0.723
+## imsmetn  0.705                       0.166       
+## imdfetn  0.833                                   
+## eimrcnt  0.249                       0.859       
+## eimpcnt  0.946                                   
+## imrcntr  0.456                       0.517       
+## impcntr  0.951                                   
+## qfimchr         0.134 -0.122  0.875              
+## qfimwht         0.151         0.725              
+## imwgdwn                                     0.748
 ## imhecop                                     0.678
-## imtcjob         0.551         0.127         0.202
-## imbleco         0.753         0.150              
-## imbgeco         0.813                            
-## imueclt         0.567        -0.210              
-## imwbcnt         0.744                            
-## imwbcrm         0.601                            
-## imrsprc  0.158         0.514        -0.103       
-## pplstrd         0.222        -0.402              
-## vrtrlg         -0.224         0.305         0.101
-## shrrfg          0.217 -0.281 -0.102         0.125
-## rfgawrk                0.499                     
-## gvrfgap                0.786                     
-## rfgfrpc         0.100 -0.279                0.156
-## rfggvfn                0.530                     
-## rfgbfml                0.610                     
+## imtcjob         0.566         0.123         0.175
+## imbleco         0.753         0.144              
+## imbgeco         0.822                            
+## imueclt         0.580        -0.201              
+## imwbcnt         0.751                            
+## imwbcrm         0.597                            
+## imrsprc  0.146         0.527                     
+## pplstrd         0.204        -0.392              
+## vrtrlg         -0.204  0.143  0.281         0.115
+## shrrfg          0.198 -0.275                0.141
+## rfgawrk                0.517                     
+## gvrfgap                0.784                     
+## rfgfrpc               -0.294                0.144
+## rfggvfn                0.512                     
+## rfgbfml                0.596                     
 ## 
 ##                  MR2   MR1   MR6   MR3   MR5   MR4
-## SS loadings    3.289 2.992 1.988 1.649 1.091 1.102
-## Proportion Var 0.132 0.120 0.080 0.066 0.044 0.044
-## Cumulative Var 0.132 0.251 0.331 0.397 0.440 0.484
+## SS loadings    3.304 3.013 1.994 1.649 1.065 1.133
+## Proportion Var 0.132 0.121 0.080 0.066 0.043 0.045
+## Cumulative Var 0.132 0.253 0.332 0.398 0.441 0.486
 ## 
 ## Loadings:
-##         MR2    MR1    MR6    MR7    MR5    MR4    MR3   
-## imsmetn  0.715                       0.170              
-## imdfetn  0.854                                          
-## eimrcnt  0.255                       0.852              
-## eimpcnt  0.965                                          
-## imrcntr  0.467                       0.516              
-## impcntr  0.978                                          
-## qfimchr                      -0.231                0.625
-## qfimwht                0.107                       0.718
-## imwgdwn                                     0.735       
-## imhecop                                     0.715       
-## imtcjob         0.573        -0.163         0.198       
-## imbleco         0.709                                   
-## imbgeco         0.852        -0.108                     
-## imueclt         0.435         0.288               -0.110
-## imwbcnt         0.576         0.235                     
-## imwbcrm         0.423         0.225                0.122
-## imrsprc  0.160         0.480        -0.100              
-## pplstrd                       0.648               -0.109
-## vrtrlg                       -0.527                     
-## shrrfg                -0.212  0.324         0.149       
-## rfgawrk                0.517                            
-## gvrfgap                0.744                            
-## rfgfrpc               -0.213  0.237         0.182       
-## rfggvfn                0.542                            
-## rfgbfml                0.599                            
+##         MR2    MR1    MR6    MR3    MR5    MR7    MR4   
+## imsmetn  0.700                       0.162              
+## imdfetn  0.821                                          
+## eimrcnt  0.245                       0.879              
+## eimpcnt  0.935                                          
+## imrcntr  0.452                       0.523              
+## impcntr  0.938                                          
+## qfimchr                       0.751                     
+## qfimwht                       0.720                     
+## imwgdwn                                            0.700
+## imhecop                                     0.172  0.624
+## imtcjob         0.574                      -0.120  0.174
+## imbleco         0.679         0.108                     
+## imbgeco         0.832                      -0.145       
+## imueclt         0.531        -0.191                     
+## imwbcnt         0.649                       0.138       
+## imwbcrm         0.464         0.131         0.290       
+## imrsprc  0.146         0.440               -0.100       
+## pplstrd                      -0.274         0.392       
+## vrtrlg         -0.121         0.190        -0.297  0.115
+## shrrfg                -0.124                0.437  0.131
+## rfgawrk                0.538                            
+## gvrfgap                0.616               -0.237       
+## rfgfrpc               -0.131                0.437  0.135
+## rfggvfn                0.504                            
+## rfgbfml                0.526                            
 ## 
-##                  MR2   MR1   MR6   MR7   MR5   MR4   MR3
-## SS loadings    3.462 2.275 1.830 1.175 1.052 1.184 0.991
-## Proportion Var 0.138 0.091 0.073 0.047 0.042 0.047 0.040
-## Cumulative Var 0.138 0.229 0.303 0.350 0.392 0.439 0.479
    +## MR2 MR1 MR6 MR3 MR5 MR7 MR4 +## SS loadings 3.224 2.467 1.456 1.305 1.105 0.901 0.984 +## Proportion Var 0.129 0.099 0.058 0.052 0.044 0.036 0.039 +## Cumulative Var 0.129 0.228 0.286 0.338 0.382 0.418 0.458

    It is very possible that you selected a different numbers of factors than Kestilä (2006). We need to keep these exercises consistent, though. So, the remaining questions will all assume you have extract three factors from the Trust items and five factors from the Attitudes items, to parallel the Kestilä (2006) results.

    -
    ## Select the three-factor solution for 'trust':
-efa_trust <- efa_trust[["3"]]
-
-## Select the five-factor solution for 'attitudes':
-efa_att <- efa_att[["5"]]
    +
    ## Select the three-factor solution for 'trust':
+efa_trust <- efa_trust[["3"]]
+
+## Select the five-factor solution for 'attitudes':
+efa_att <- efa_att[["5"]]

    4.4.10

    Give the factor scores meaningful names, and add the scores to the ess dataset as new columns.

    -

    Hint: If you’re not sure of what do to, check 4.3.9.

    +

    Hint: If you’re not sure of what do to, check 4.3.11.

    Click to show code -
    ## Rename the factor scores:
-colnames(efa_trust$scores) <- c("trust_pol", "satisfy", "trust_inst")
-colnames(efa_att$scores)   <- c("effects", 
-                                "allowance", 
-                                "refugees", 
-                                "ethnic", 
-                                "europe")
-
-## Add factor scores to the dataset as new columns:
-ess <- data.frame(ess, efa_trust$scores, efa_att$scores)
    +
    ## Rename the factor scores:
+colnames(efa_trust$scores) <- c("trust_inst", "satisfy", "trust_pol")
+colnames(efa_att$scores)   <- c("effects", 
+                                "allowance", 
+                                "refugees", 
+                                "ethnic", 
+                                "europe")
+
+## Add factor scores to the dataset as new columns:
+ess <- data.frame(ess, efa_trust$scores, efa_att$scores)

    Kestilä (2006) used the component scores to descriptively evaluate country-level @@ -1163,7 +1149,7 @@

    4.4.10

    time to replicate those analyses.

    -
    +

    4.4.11

    Repeat the Kestilä (2006) between-country comparison using the factor scores you created in 4.4.10 and an appropriate statistical test.

    @@ -1175,89 +1161,89 @@

    4.4.11

    dimensions. We can use a linear model to test whether the countries differ in average levels of Trust in Institutions (as quantified by the relevant factor score).

    -
    ## Estimate the model:
-out <- lm(trust_inst ~ cntry, data = ess)
-
-## View the regression-style summary:
-summary(out)
    +
    ## Estimate the model:
+out <- lm(trust_inst ~ country, data = ess)
+
+## View the regression-style summary:
+summary(out)
    ## 
 ## Call:
-## lm(formula = trust_inst ~ cntry, data = ess)
+## lm(formula = trust_inst ~ country, data = ess)
 ## 
 ## Residuals:
 ##     Min      1Q  Median      3Q     Max 
-## -4.0439 -0.5990  0.0774  0.6546  3.8306 
+## -4.2295 -0.6226  0.1171  0.7194  3.3061 
 ## 
 ## Coefficients:
-##                  Estimate Std. Error t value Pr(>|t|)    
-## (Intercept)       0.05722    0.02380   2.404  0.01624 *  
-## cntryBelgium      0.24797    0.03461   7.165 8.15e-13 ***
-## cntryDenmark      0.89049    0.03709  24.006  < 2e-16 ***
-## cntryFinland      0.66876    0.03293  20.309  < 2e-16 ***
-## cntryGermany     -1.02160    0.03082 -33.147  < 2e-16 ***
-## cntryItaly       -0.70295    0.03910 -17.980  < 2e-16 ***
-## cntryNetherlands -0.10502    0.03233  -3.248  0.00117 ** 
-## cntryNorway       0.07686    0.03331   2.308  0.02103 *  
-## cntrySweden      -0.06433    0.03464  -1.857  0.06332 .  
+##                    Estimate Std. Error t value Pr(>|t|)    
+## (Intercept)        -0.09028    0.02445  -3.692 0.000224 ***
+## countryBelgium     -0.28923    0.03642  -7.942 2.12e-15 ***
+## countryGermany     -0.05966    0.03211  -1.858 0.063205 .  
+## countryDenmark      0.75509    0.03882  19.452  < 2e-16 ***
+## countryFinland      0.59235    0.03439  17.224  < 2e-16 ***
+## countryItaly        0.10991    0.04071   2.700 0.006939 ** 
+## countryNetherlands -0.05357    0.03379  -1.585 0.112893    
+## countryNorway       0.36922    0.03493  10.570  < 2e-16 ***
+## countrySweden       0.28560    0.03613   7.904 2.89e-15 ***
 ## ---
 ## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
 ## 
-## Residual standard error: 0.956 on 14315 degrees of freedom
-##   (3863 observations deleted due to missingness)
-## Multiple R-squared:  0.2656, Adjusted R-squared:  0.2652 
-## F-statistic: 647.2 on 8 and 14315 DF,  p-value: < 2.2e-16
    -
    ## View the results as an ANOVA table:
-anova(out)
    +## Residual standard error: 1.029 on 14769 degrees of freedom +## (4912 observations deleted due to missingness) +## Multiple R-squared: 0.082, Adjusted R-squared: 0.0815 +## F-statistic: 164.9 on 8 and 14769 DF, p-value: < 2.2e-16 +
    ## View the results as an ANOVA table:
+anova(out)
    -
    ## Post-hoc tests
-out %>% aov() %>% TukeyHSD()
    +
    ## Post-hoc tests
+out %>% aov() %>% TukeyHSD()
    ##   Tukey multiple comparisons of means
 ##     95% family-wise confidence level
 ## 
 ## Fit: aov(formula = .)
 ## 
-## $cntry
-##                            diff         lwr          upr     p adj
-## Belgium-Austria      0.24796876  0.14060941  0.355328097 0.0000000
-## Denmark-Austria      0.89049075  0.77541622  1.005565280 0.0000000
-## Finland-Austria      0.66876487  0.56661216  0.770917590 0.0000000
-## Germany-Austria     -1.02160451 -1.11721519 -0.925993827 0.0000000
-## Italy-Austria       -0.70295083 -0.82423523 -0.581666425 0.0000000
-## Netherlands-Austria -0.10501583 -0.20532400 -0.004707667 0.0319748
-## Norway-Austria       0.07685797 -0.02646120  0.180177134 0.3373046
-## Sweden-Austria      -0.06432725 -0.17178456  0.043130061 0.6435559
-## Denmark-Belgium      0.64252199  0.52477886  0.760265130 0.0000000
-## Finland-Belgium      0.42079612  0.31564634  0.525945897 0.0000000
-## Germany-Belgium     -1.26957326 -1.36837964 -1.170766883 0.0000000
-## Italy-Belgium       -0.95091958 -1.07473883 -0.827100341 0.0000000
-## Netherlands-Belgium -0.35298459 -0.45634331 -0.249625861 0.0000000
-## Norway-Belgium      -0.17111079 -0.27739413 -0.064827447 0.0000211
-## Sweden-Belgium      -0.31229600 -0.42260635 -0.201985661 0.0000000
-## Finland-Denmark     -0.22172588 -0.33474178 -0.108709970 0.0000000
-## Germany-Denmark     -1.91209526 -2.01923451 -1.804956003 0.0000000
-## Italy-Denmark       -1.59344158 -1.72400698 -1.462876176 0.0000000
-## Netherlands-Denmark -0.99550658 -1.10685803 -0.884155131 0.0000000
-## Norway-Denmark      -0.81363278 -0.92770411 -0.699561455 0.0000000
-## Sweden-Denmark      -0.95481800 -1.07265047 -0.836985529 0.0000000
-## Germany-Finland     -1.69036938 -1.78349215 -1.597246611 0.0000000
-## Italy-Finland       -1.37171570 -1.49104866 -1.252382745 0.0000000
-## Netherlands-Finland -0.77378070 -0.87172036 -0.675841051 0.0000000
-## Norway-Finland      -0.59190690 -0.69292818 -0.490885631 0.0000000
-## Sweden-Finland      -0.73309212 -0.83834193 -0.627842317 0.0000000
-## Italy-Germany        0.31865368  0.20487064  0.432436712 0.0000000
-## Netherlands-Germany  0.91658867  0.82549312  1.007684229 0.0000000
-## Norway-Germany       1.09846248  1.00406162  1.192863333 0.0000000
-## Sweden-Germany       0.95727726  0.85836444  1.056190076 0.0000000
-## Netherlands-Italy    0.59793500  0.48017717  0.715692818 0.0000000
-## Norway-Italy         0.77980880  0.65947582  0.900141779 0.0000000
-## Sweden-Italy         0.63862358  0.51471938  0.762527777 0.0000000
-## Norway-Netherlands   0.18187380  0.08271813  0.281029475 0.0000005
-## Sweden-Netherlands   0.04068858 -0.06277190  0.144149065 0.9523456
-## Sweden-Norway       -0.14118522 -0.24756752 -0.034802920 0.0012799
    +## $country +## diff lwr upr p adj +## Belgium-Austria -0.289225482 -0.40219224 -0.17625873 0.0000000 +## Germany-Austria -0.059655996 -0.15926604 0.03995405 0.6429963 +## Denmark-Austria 0.755089552 0.63466911 0.87551000 0.0000000 +## Finland-Austria 0.592348290 0.48565882 0.69903776 0.0000000 +## Italy-Austria 0.109910185 -0.01636587 0.23618624 0.1476635 +## Netherlands-Austria -0.053567808 -0.15838407 0.05124846 0.8131104 +## Norway-Austria 0.369224250 0.26085692 0.47759158 0.0000000 +## Sweden-Austria 0.285601197 0.17350905 0.39769334 0.0000000 +## Germany-Belgium 0.229569486 0.12386351 0.33527546 0.0000000 +## Denmark-Belgium 1.044315033 0.91880537 1.16982470 0.0000000 +## Finland-Belgium 0.881573772 0.76917165 0.99397589 0.0000000 +## Italy-Belgium 0.399135667 0.26799745 0.53027389 0.0000000 +## Netherlands-Belgium 0.235657673 0.12503199 0.34628336 0.0000000 +## Norway-Belgium 0.658449732 0.54445381 0.77244566 0.0000000 +## Sweden-Belgium 0.574826679 0.45728417 0.69236918 0.0000000 +## Denmark-Germany 0.814745547 0.70110863 0.92838247 0.0000000 +## Finland-Germany 0.652004286 0.55303505 0.75097352 0.0000000 +## Italy-Germany 0.169566181 0.04974170 0.28939066 0.0003895 +## Netherlands-Germany 0.006088188 -0.09085878 0.10303516 0.9999999 +## Norway-Germany 0.428880246 0.32810453 0.52965596 0.0000000 +## Sweden-Germany 0.345257193 0.24048642 0.45002796 0.0000000 +## Finland-Denmark -0.162741262 -0.28263218 -0.04285034 0.0008579 +## Italy-Denmark -0.645179366 -0.78279052 -0.50756821 0.0000000 +## Netherlands-Denmark -0.808657360 -0.92688442 -0.69043030 0.0000000 +## Norway-Denmark -0.385865301 -0.50725174 -0.26447886 0.0000000 +## Sweden-Denmark -0.469488354 -0.59421139 -0.34476531 0.0000000 +## Italy-Finland -0.482438105 -0.60820928 -0.35666693 0.0000000 +## Netherlands-Finland -0.645916098 -0.75012357 -0.54170862 0.0000000 +## Norway-Finland -0.223124040 -0.33090264 -0.11534544 0.0000000 +## Sweden-Finland -0.306747093 -0.41827017 -0.19522402 0.0000000 +## Netherlands-Italy -0.163477993 -0.28766412 -0.03929186 0.0014719 +## Norway-Italy 0.259314065 0.13211649 0.38651164 0.0000000 +## Sweden-Italy 0.175691012 0.04530545 0.30607657 0.0009794 +## Norway-Netherlands 0.422792059 0.31686740 0.52871671 0.0000000 +## Sweden-Netherlands 0.339169005 0.22943659 0.44890142 0.0000000 +## Sweden-Norway -0.083623053 -0.19675232 0.02950622 0.3462227
    Click for explanation @@ -1288,15 +1274,15 @@

    4.4.11

    part of the analysis, we need to subset the data to only the Finnish cases.

    -
    +

    4.4.12

    Create a new data frame that contains only the Finnish cases from ess.

    -

    Hint: You can use logical indexing based on the cntry variable.

    +

    Hint: You can use logical indexing based on the country variable.

    Click to show code -
    ess_finland <- filter(ess, cntry == "Finland")
    +
    ess_finland <- filter(ess, country == "Finland")

    We still have one more step before we can estimate any models. We must prepare @@ -1307,9 +1293,9 @@

    4.4.12

  • We have not yet used any of the independent variables, though. So, we should inspect those variables to see if they require any processing.
    • -

    In the ess data, the relevant variables have the following names:

    +

    In our processed ess data, the relevant variables have the following names:

      -
    • gndr
      • +
    • sex
    • yrbrn
    • eduyrs
    • polintr
      • @@ -1317,20 +1303,20 @@

      4.4.12

    -
    +

    4.4.13

    Inspect the independent variables listed above.

    Click to show code -
    library(tidySEM)
-
-select(ess_finland, gndr, yrbrn, eduyrs, polintr, lrscale) %>% 
-  descriptives()
    +
    library(tidySEM)
+
+select(ess_finland, sex, yrbrn, eduyrs, polintr, lrscale) %>% 
+  descriptives()
    @@ -1342,7 +1328,7 @@

    4.4.13

    -
    +

    4.4.14

    Apply any necessary recoding/transformations.

    1. Age

    @@ -1350,7 +1336,7 @@

    4.4.14

    Click to show code -
    ess_finland <- mutate(ess_finland, age = 2002 - yrbrn)
    +
    ess_finland <- mutate(ess_finland, age = 2002 - yrbrn)
    Click for explanation @@ -1362,112 +1348,81 @@

    4.4.14

    -

    2. Political Interest & Orientation

    +

    2. Political Interest

    Click to show code

    First, we’ll transform polintr.

    -
    ## Store the original variable for checking purposes:
-tmp <- ess_finland$polintr
-
-## Recode the four character values into two factor levels:
-ess_finland <-
-  mutate(ess_finland,
-         polintr_bin = recode_factor(polintr,
-                                     "Not at all interested" = "Low Interest",
-                                     "Hardly interested" = "Low Interest",
-                                     "Quite interested" = "High Interest",
-                                     "Very interested" = "High Interest")
-  )
-
-## Check the conversion:
-table(old = tmp, new = ess_finland$polintr_bin, useNA = "always")
    +
    ## Recode the four factor levels into two factor levels:
+ess_finland <-
+  mutate(ess_finland,
+         polintr_bin = recode_factor(polintr,
+                                     "Not at all interested" = "Low Interest",
+                                     "Hardly interested" = "Low Interest",
+                                     "Quite interested" = "High Interest",
+                                     "Very interested" = "High Interest")
+         )
+
+## Check the conversion:
+with(ess_finland, table(old = polintr, new = polintr_bin, useNA = "always"))
    ##                        new
 ## old                     Low Interest High Interest <NA>
+##   Very interested                  0           144    0
+##   Quite interested                 0           785    0
 ##   Hardly interested              842             0    0
 ##   Not at all interested          228             0    0
-##   Quite interested                 0           785    0
-##   Very interested                  0           144    0
 ##   <NA>                             0             0    1
    -

    Now, we’ll deal with lrscale.

    -
    ## Save the old version for checking:
-tmp <- ess_finland$lrscale
-
-## Recode the extreme levels:
-ess_finland <- mutate(ess_finland,
-                      lrscale = recode(lrscale,
-                                       "Left" = 0,
-                                       "Right" = 10,
-                                       .default = as.numeric(lrscale)
-                      )
-)
-
-## Check the conversion:
-table(old = tmp, new = ess_finland$lrscale, useNA = "always")
    -
    ##        new
-## old       0   1   2   3   4   5   6   7   8   9  10 <NA>
-##   1       0  25   0   0   0   0   0   0   0   0   0    0
-##   2       0   0  62   0   0   0   0   0   0   0   0    0
-##   3       0   0   0 148   0   0   0   0   0   0   0    0
-##   4       0   0   0   0 183   0   0   0   0   0   0    0
-##   5       0   0   0   0   0 599   0   0   0   0   0    0
-##   6       0   0   0   0   0   0 199   0   0   0   0    0
-##   7       0   0   0   0   0   0   0 296   0   0   0    0
-##   8       0   0   0   0   0   0   0   0 217   0   0    0
-##   9       0   0   0   0   0   0   0   0   0  79   0    0
-##   Left   24   0   0   0   0   0   0   0   0   0   0    0
-##   Right   0   0   0   0   0   0   0   0   0   0  59    0
-##   <NA>    0   0   0   0   0   0   0   0   0   0   0  109
    Click for explanation -

    The variables polintr and lrscale are represented as character vectors. If -we analyze these variables as they are, R will convert the character vectors -to factors and make dummy codes for each distinct value; we definitely don’t -want that.

    -

    We need to convert polintr and lrscale to numeric vectors, but we cannot -naively apply the as.numeric() function, because R doesn’t know how to -convert a word into a number (or, at least, not how we want the operation to -be done). There are many ways that we could go about this conversion, but the -recode() function from the dplyr package is particularly useful for -arbitrary recoding tasks such as we have here.

    +

    Kestilä (2006) dichotomized polintr by combining the lowest two and highest two categories. So, we don’t actually want to convert the polint variable into a numeric, Likert-type variable. We want polint to be a binary factor. The recode_factor() function from dplyr() will automatically convert our result into a factor.

    -

    For political orientation, only the two extreme values of the scale are labeled -as text. So, we only need to recode those two levels.

    -
      -

    • Since we’re only replacing two of the 11 values, we’ll need to provide a -value for the .default argument in dplyr::recode().

      -
    - + +
    -

    3. Sex

    -
    - -Click to show code - -
    ess_finland <- mutate(ess_finland, sex = factor(gndr))
-
-## Check results:
-ess_finland %$% table(old = gndr, new = sex, useNA = "always")
    -
    ##         new
-## old      Female Male <NA>
-##   Female   1040    0    0
-##   Male        0  960    0
-##   <NA>        0    0    0
    +

    As with the ess_round1.rds data, we will be coming back to this Finnish +subsample data in future practical exercises. So, we should save our work by +writing the processed dataset to disk.

    +
    +
    +
    +

    4.4.15

    +

    Use the saveRDS() function to save the processed Finnish subsample data.

    -Click for explanation +Click to see code -

    Although gndr would be automatically converted to a factor by lm(), it’s -better to be explicit about our intentions and create the factor ourselves.

    -
    +
    ## Save the processed Finnish data:
+saveRDS(ess_finland, "ess_finland.rds")

    Now, we’re finally ready to replicate the regression analysis from Kestilä (2006). @@ -1476,8 +1431,8 @@

    4.4.14

    analysis.

    -
    -

    4.4.15

    +
    +

    4.4.16

    Run a series of multiple linear regression analyses with the factor scores you created in 4.4.10 as the dependent variables and the same predictors used by Kestilä (2006).

    @@ -1488,10 +1443,10 @@

    4.4.15

    Click to show code -
    ## Predicting 'Trust in Institutions':
-out_trust_inst <- lm(trust_inst ~ sex + age + eduyrs + polintr_bin + lrscale, 
-                     data = ess_finland)
-summary(out_trust_inst)
    +
    ## Predicting 'Trust in Institutions':
+out_trust_inst <- lm(trust_inst ~ sex + age + eduyrs + polintr_bin + lrscale, 
+                     data = ess_finland)
+summary(out_trust_inst)
    ## 
 ## Call:
 ## lm(formula = trust_inst ~ sex + age + eduyrs + polintr_bin + 
@@ -1499,27 +1454,27 @@ 
    4.4.15 
    
 ## 
 ## Residuals:
 ##     Min      1Q  Median      3Q     Max 
-## -3.5724 -0.4492  0.1260  0.5941  2.3547 
+## -3.9499 -0.5102  0.1337  0.6638  2.5919 
 ## 
 ## Coefficients:
-##                            Estimate Std. Error t value Pr(>|t|)    
-## (Intercept)               0.3289965  0.1151344   2.857 0.004322 ** 
-## sexMale                   0.1505241  0.0409380   3.677 0.000243 ***
-## age                      -0.0055811  0.0012592  -4.432 9.92e-06 ***
-## eduyrs                   -0.0001833  0.0057951  -0.032 0.974774    
-## polintr_binHigh Interest  0.1237345  0.0421741   2.934 0.003392 ** 
-## lrscale                   0.0907595  0.0101966   8.901  < 2e-16 ***
+##                           Estimate Std. Error t value Pr(>|t|)    
+## (Intercept)              -0.057518   0.124294  -0.463 0.643595    
+## sexFemale                 0.004091   0.045170   0.091 0.927849    
+## age                      -0.003071   0.001380  -2.225 0.026219 *  
+## eduyrs                    0.023223   0.006388   3.635 0.000286 ***
+## polintr_binHigh Interest  0.166860   0.046448   3.592 0.000337 ***
+## lrscale                   0.058951   0.011232   5.249 1.72e-07 ***
 ## ---
 ## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
 ## 
-## Residual standard error: 0.8355 on 1694 degrees of freedom
-##   (300 observations deleted due to missingness)
-## Multiple R-squared:  0.06695,    Adjusted R-squared:  0.0642 
-## F-statistic: 24.31 on 5 and 1694 DF,  p-value: < 2.2e-16
    -
    ## Predicting 'Trust in Politicians':
-out_trust_pol <- lm(trust_pol ~ sex + age + eduyrs + polintr_bin + lrscale, 
-                    data = ess_finland)
-summary(out_trust_pol)
    +## Residual standard error: 0.9321 on 1734 degrees of freedom +## (260 observations deleted due to missingness) +## Multiple R-squared: 0.04155, Adjusted R-squared: 0.03879 +## F-statistic: 15.03 on 5 and 1734 DF, p-value: 1.78e-14 +
    ## Predicting 'Trust in Politicians':
+out_trust_pol <- lm(trust_pol ~ sex + age + eduyrs + polintr_bin + lrscale, 
+                    data = ess_finland)
+summary(out_trust_pol)
    ## 
 ## Call:
 ## lm(formula = trust_pol ~ sex + age + eduyrs + polintr_bin + lrscale, 
@@ -1527,27 +1482,27 @@ 
    4.4.15 
    
 ## 
 ## Residuals:
 ##      Min       1Q   Median       3Q      Max 
-## -2.85480 -0.50639  0.06368  0.63001  2.58685 
+## -3.03673 -0.67306  0.05346  0.69666  2.38771 
 ## 
 ## Coefficients:
 ##                           Estimate Std. Error t value Pr(>|t|)    
-## (Intercept)               0.097827   0.121803   0.803 0.421995    
-## sexMale                   0.163314   0.043309   3.771 0.000168 ***
-## age                       0.013771   0.001332  10.337  < 2e-16 ***
-## eduyrs                   -0.050447   0.006131  -8.229 3.73e-16 ***
-## polintr_binHigh Interest -0.258279   0.044617  -5.789 8.43e-09 ***
-## lrscale                   0.033794   0.010787   3.133 0.001761 ** 
+## (Intercept)              -0.165989   0.126840  -1.309  0.19083    
+## sexFemale                 0.015572   0.046095   0.338  0.73554    
+## age                      -0.009112   0.001409  -6.469 1.28e-10 ***
+## eduyrs                    0.018476   0.006519   2.834  0.00465 ** 
+## polintr_binHigh Interest  0.463763   0.047399   9.784  < 2e-16 ***
+## lrscale                   0.054932   0.011462   4.793 1.79e-06 ***
 ## ---
 ## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
 ## 
-## Residual standard error: 0.8839 on 1694 degrees of freedom
-##   (300 observations deleted due to missingness)
-## Multiple R-squared:  0.1645, Adjusted R-squared:  0.1621 
-## F-statistic: 66.73 on 5 and 1694 DF,  p-value: < 2.2e-16
    -
    ## Predicting 'Attitudes toward Refugees':
-out_refugees <- lm(refugees ~ sex + age + eduyrs + polintr_bin + lrscale, 
-                   data = ess_finland)
-summary(out_refugees)
    +## Residual standard error: 0.9512 on 1734 degrees of freedom +## (260 observations deleted due to missingness) +## Multiple R-squared: 0.09806, Adjusted R-squared: 0.09546 +## F-statistic: 37.71 on 5 and 1734 DF, p-value: < 2.2e-16 +
    ## Predicting 'Attitudes toward Refugees':
+out_refugees <- lm(refugees ~ sex + age + eduyrs + polintr_bin + lrscale, 
+                   data = ess_finland)
+summary(out_refugees)
    ## 
 ## Call:
 ## lm(formula = refugees ~ sex + age + eduyrs + polintr_bin + lrscale, 
@@ -1555,23 +1510,23 @@ 
    4.4.15 
    
 ## 
 ## Residuals:
 ##     Min      1Q  Median      3Q     Max 
-## -2.8921 -0.7112 -0.0618  0.6912  4.1660 
+## -2.9118 -0.6860 -0.0594  0.6904  4.1044 
 ## 
 ## Coefficients:
 ##                            Estimate Std. Error t value Pr(>|t|)    
-## (Intercept)              -0.6771470  0.1489234  -4.547 5.83e-06 ***
-## sexMale                   0.4853926  0.0524999   9.246  < 2e-16 ***
-## age                      -0.0004773  0.0016248  -0.294 0.768968    
-## eduyrs                   -0.0254016  0.0075576  -3.361 0.000794 ***
-## polintr_binHigh Interest -0.2148018  0.0541602  -3.966 7.61e-05 ***
-## lrscale                   0.0940049  0.0131295   7.160 1.20e-12 ***
+## (Intercept)              -1.690e-01  1.438e-01  -1.175 0.240080    
+## sexFemale                -4.828e-01  5.181e-02  -9.318  < 2e-16 ***
+## age                       2.903e-05  1.604e-03   0.018 0.985561    
+## eduyrs                   -2.537e-02  7.459e-03  -3.401 0.000688 ***
+## polintr_binHigh Interest -2.131e-01  5.345e-02  -3.986 6.99e-05 ***
+## lrscale                   9.359e-02  1.296e-02   7.223 7.65e-13 ***
 ## ---
 ## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
 ## 
-## Residual standard error: 1.074 on 1699 degrees of freedom
+## Residual standard error: 1.06 on 1699 degrees of freedom
 ##   (295 observations deleted due to missingness)
-## Multiple R-squared:  0.09368,    Adjusted R-squared:  0.09102 
-## F-statistic: 35.12 on 5 and 1699 DF,  p-value: < 2.2e-16
    +## Multiple R-squared: 0.09535, Adjusted R-squared: 0.09269 +## F-statistic: 35.81 on 5 and 1699 DF, p-value: < 2.2e-16

    That does it for our replication of the Kesilä (2006) analyses, but we still have @@ -1590,8 +1545,8 @@

    4.4.15

    the internal consistency (a type of reliability) of the hypothesized subscales.

    -
    -

    4.4.16

    +
    +

    4.4.17

    Estimate the internal consistency of the three Trust subscales and five Attitudes subscales implied by your EFA solutions from above.

      @@ -1609,17 +1564,17 @@

      4.4.16

      Click to show code -
      ## Run the reliability analysis on the subscale data:
-(
-  out <- select(ess, starts_with("stf")) %>%
-    psych::alpha()
-)
      +
      ## Run the reliability analysis on the subscale data:
+(
+  out <- select(ess, starts_with("stf")) %>%
+    psych::alpha()
+)
      ## 
 ## Reliability analysis   
 ## Call: psych::alpha(x = .)
 ## 
 ##   raw_alpha std.alpha G6(smc) average_r S/N    ase mean  sd median_r
-##       0.79      0.79    0.77      0.44 3.9 0.0024  6.4 1.7     0.41
+##       0.79      0.79    0.77      0.44 3.9 0.0023  5.4 1.7     0.41
 ## 
 ##     95% confidence boundaries 
 ##          lower alpha upper
@@ -1628,19 +1583,19 @@ 
      4.4.16 
      
 ## 
 ##  Reliability if an item is dropped:
 ##         raw_alpha std.alpha G6(smc) average_r S/N alpha se  var.r med.r
-## stfeco       0.74      0.74    0.70      0.42 2.9   0.0032 0.0067  0.39
-## stfgov       0.74      0.74    0.69      0.42 2.9   0.0031 0.0031  0.41
-## stfdem       0.75      0.75    0.71      0.43 3.0   0.0030 0.0074  0.41
-## stfedu       0.76      0.76    0.72      0.45 3.2   0.0029 0.0103  0.43
-## stfhlth      0.78      0.78    0.73      0.47 3.5   0.0027 0.0063  0.45
+## stfhlth      0.78      0.78    0.73      0.47 3.5   0.0026 0.0064  0.46
+## stfedu       0.76      0.76    0.72      0.45 3.2   0.0028 0.0109  0.44
+## stfeco       0.74      0.74    0.70      0.41 2.8   0.0031 0.0069  0.39
+## stfgov       0.74      0.74    0.69      0.42 2.9   0.0030 0.0035  0.41
+## stfdem       0.75      0.75    0.71      0.43 3.0   0.0029 0.0074  0.40
 ## 
 ##  Item statistics 
 ##             n raw.r std.r r.cor r.drop mean  sd
-## stfeco  17728  0.78  0.77  0.70   0.62  6.1 2.3
-## stfgov  17621  0.77  0.77  0.70   0.61  5.5 2.3
-## stfdem  17624  0.75  0.75  0.66   0.58  6.8 2.2
-## stfedu  17380  0.73  0.73  0.62   0.55  7.0 2.3
-## stfhlth 17981  0.70  0.69  0.57   0.50  6.8 2.3
      +## stfhlth 19481 0.69 0.69 0.56 0.50 5.8 2.3 +## stfedu 18844 0.73 0.73 0.62 0.55 5.9 2.3 +## stfeco 19211 0.78 0.78 0.70 0.63 5.0 2.4 +## stfgov 19106 0.77 0.76 0.69 0.61 4.5 2.3 +## stfdem 19106 0.75 0.75 0.67 0.59 5.7 2.3
      Click for explanation @@ -1660,7 +1615,7 @@

      4.4.16

      the internal consistency of this subscale, we simply provide a data frame containing only the subscale data to the alpha() function.

      The raw_alpha value is the estimate of Cronbach’s Alpha. In this case -\(\alpha = 0.795\), so the subscale is pretty reliable.

      +\(\alpha = 0.794\), so the subscale is pretty reliable.

      The table labeled “Reliability if an item is dropped” shows what Cronbach’s Alpha would be if each item were excluded from the scale. If this value is notably higher than the raw_alpha value, it could indicate a bad item.

      diff --git a/docs/in-class-exercises-4.html b/docs/in-class-exercises-4.html index 282827f6..68814973 100644 --- a/docs/in-class-exercises-4.html +++ b/docs/in-class-exercises-4.html @@ -333,28 +333,31 @@
    • 4.3.5
    • 4.3.6
    • 4.3.7
      • -
    • 4.3.8
      • -
    • 4.3.9
      • -
    • 4.3.10
      • +
    • 4.3.8
      • +
    • 4.3.9
      • +
    • 4.3.10
      • +
    • 4.3.11
      • +
    • 4.3.12
  • 4.4 In-Class Exercises
  • 5 CFA @@ -371,30 +374,30 @@
  • 5.3 At-Home Exercises
  • 5.4 In-Class Exercises
  • 6 Full SEM @@ -411,23 +414,22 @@
  • 6.3 At-Home Exercises
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models @@ -466,22 +468,17 @@

    5.4 In-Class Exercises +

    5.4.1

    Load the ESS data.
    Click to show code -
    ess <- readRDS("ess_round1_attitudes.rds")
    +
    ess <- readRDS("ess_round1.rds")

    We are going to conduct a CFA to evaluate the measurement model implied by the @@ -516,28 +513,28 @@

    5.4.2

    Click to show code -
    mod_5f <- '
-## Immigration Policy:
-ip =~ imrcntr + eimrcnt + eimpcnt + imsmetn + impcntr + imdfetn 
-
-## Social Threat:
-st =~ imbgeco + imbleco + imwbcnt + imwbcrm + imtcjob + imueclt
-
-## Refugee Policy:
-rp =~ gvrfgap + imrsprc + rfgbfml + rfggvfn + rfgawrk + rfgfrpc + shrrfg
-
-## Cultural Threat:
-ct =~ qfimchr + qfimwht + pplstrd + vrtrlg
-
-## Economic Threat:
-et =~ imwgdwn + imhecop 
-'
    +
    mod_5f <- '
+## Immigration Policy:
+ip =~ imrcntr + eimrcnt + eimpcnt + imsmetn + impcntr + imdfetn 
+
+## Social Threat:
+st =~ imbgeco + imbleco + imwbcnt + imwbcrm + imtcjob + imueclt
+
+## Refugee Policy:
+rp =~ gvrfgap + imrsprc + rfgbfml + rfggvfn + rfgawrk + rfgfrpc + shrrfg
+
+## Cultural Threat:
+ct =~ qfimchr + qfimwht + pplstrd + vrtrlg
+
+## Economic Threat:
+et =~ imwgdwn + imhecop 
+'

    Note: We don’t have to specify the latent covariances in the model syntax, we can tell lavaan to estimate all latent covariances when we fit the model.

    -
    +

    5.4.3

    Estimate the CFA model you defined above, and summarize the results.

      @@ -559,15 +556,15 @@

      5.4.3

      Click to show code -
      ## Load the lavaan package:
-library(lavaan)
-
-## Estimate the CFA model:
-fit_5f <- cfa(mod_5f, data = ess)
-
-## Summarize the fitted model:
-summary(fit_5f, fit.measures = TRUE, standardized = TRUE)
      -
      ## lavaan 0.6.16 ended normally after 71 iterations
+
      ## Load the lavaan package:
+library(lavaan)
+
+## Estimate the CFA model:
+fit_5f <- cfa(mod_5f, data = ess)
+
+## Summarize the fitted model:
+summary(fit_5f, fit.measures = TRUE, standardized = TRUE)
      
+
      ## lavaan 0.6.16 ended normally after 72 iterations
 ## 
 ##   Estimator                                         ML
 ##   Optimization method                           NLMINB
@@ -578,29 +575,29 @@ 
      5.4.3 
      
 ## 
 ## Model Test User Model:
 ##                                                        
-##   Test statistic                              18630.545
+##   Test statistic                              18631.556
 ##   Degrees of freedom                                265
 ##   P-value (Chi-square)                            0.000
 ## 
 ## Model Test Baseline Model:
 ## 
-##   Test statistic                            160506.963
+##   Test statistic                            159619.058
 ##   Degrees of freedom                               300
 ##   P-value                                        0.000
 ## 
 ## User Model versus Baseline Model:
 ## 
 ##   Comparative Fit Index (CFI)                    0.885
-##   Tucker-Lewis Index (TLI)                       0.870
+##   Tucker-Lewis Index (TLI)                       0.869
 ## 
 ## Loglikelihood and Information Criteria:
 ## 
-##   Loglikelihood user model (H0)            -520659.935
-##   Loglikelihood unrestricted model (H1)    -511344.662
+##   Loglikelihood user model (H0)            -520035.133
+##   Loglikelihood unrestricted model (H1)    -510719.354
 ##                                                       
-##   Akaike (AIC)                             1041439.869
-##   Bayesian (BIC)                           1041893.710
-##   Sample-size adjusted Bayesian (SABIC)    1041703.036
+##   Akaike (AIC)                             1040190.265
+##   Bayesian (BIC)                           1040644.106
+##   Sample-size adjusted Bayesian (SABIC)    1040453.432
 ## 
 ## Root Mean Square Error of Approximation:
 ## 
@@ -612,7 +609,7 @@ 
      5.4.3 
      
 ## 
 ## Standardized Root Mean Square Residual:
 ## 
-##   SRMR                                           0.049
+##   SRMR                                           0.048
 ## 
 ## Parameter Estimates:
 ## 
@@ -624,84 +621,84 @@ 
      5.4.3 
      
 ##                    Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
 ##   ip =~                                                                 
 ##     imrcntr           1.000                               0.617    0.748
-##     eimrcnt           0.942    0.011   84.956    0.000    0.582    0.696
-##     eimpcnt           1.126    0.010  113.425    0.000    0.695    0.898
-##     imsmetn           0.982    0.010   98.770    0.000    0.606    0.796
-##     impcntr           1.150    0.010  113.642    0.000    0.710    0.900
-##     imdfetn           1.131    0.010  111.811    0.000    0.698    0.887
+##     eimrcnt           0.942    0.011   84.943    0.000    0.582    0.696
+##     eimpcnt           1.127    0.010  113.413    0.000    0.695    0.898
+##     imsmetn           0.982    0.010   98.753    0.000    0.606    0.796
+##     impcntr           1.150    0.010  113.623    0.000    0.710    0.900
+##     imdfetn           1.132    0.010  111.802    0.000    0.698    0.887
 ##   st =~                                                                 
-##     imbgeco           1.000                               1.644    0.737
-##     imbleco           0.828    0.012   71.268    0.000    1.362    0.630
-##     imwbcnt           1.049    0.012   90.290    0.000    1.724    0.799
-##     imwbcrm           0.709    0.011   64.669    0.000    1.166    0.573
-##     imtcjob           0.760    0.011   68.999    0.000    1.248    0.611
-##     imueclt           1.022    0.013   80.247    0.000    1.680    0.708
+##     imbgeco           1.000                               1.608    0.728
+##     imbleco           0.826    0.012   69.222    0.000    1.327    0.619
+##     imwbcnt           1.046    0.012   88.056    0.000    1.682    0.792
+##     imwbcrm           0.713    0.011   63.102    0.000    1.146    0.564
+##     imtcjob           0.751    0.011   66.787    0.000    1.207    0.597
+##     imueclt           1.008    0.013   78.043    0.000    1.621    0.698
 ##   rp =~                                                                 
-##     gvrfgap           1.000                               0.649    0.601
-##     imrsprc           0.865    0.017   51.063    0.000    0.561    0.533
-##     rfgbfml           1.064    0.019   55.354    0.000    0.690    0.593
-##     rfggvfn           0.870    0.017   51.404    0.000    0.564    0.538
-##     rfgawrk           0.666    0.016   40.873    0.000    0.432    0.407
-##     rfgfrpc          -0.822    0.016  -50.427    0.000   -0.533   -0.525
-##     shrrfg           -1.021    0.018  -57.724    0.000   -0.662   -0.629
+##     gvrfgap           1.000                               0.659    0.610
+##     imrsprc           0.855    0.016   51.881    0.000    0.563    0.535
+##     rfgbfml           1.047    0.019   56.174    0.000    0.690    0.593
+##     rfggvfn           0.849    0.016   51.714    0.000    0.559    0.533
+##     rfgawrk           0.653    0.016   41.044    0.000    0.430    0.405
+##     rfgfrpc          -0.810    0.016  -51.095    0.000   -0.534   -0.525
+##     shrrfg           -0.999    0.017  -58.381    0.000   -0.658   -0.625
 ##   ct =~                                                                 
-##     qfimchr           1.000                               1.875    0.640
-##     qfimwht           0.940    0.017   55.019    0.000    1.762    0.670
-##     pplstrd          -0.349    0.007  -51.093    0.000   -0.655   -0.584
-##     vrtrlg            0.250    0.006   41.934    0.000    0.468    0.449
+##     qfimchr           1.000                               1.836    0.629
+##     qfimwht           0.941    0.017   54.250    0.000    1.728    0.659
+##     pplstrd          -0.366    0.007  -51.585    0.000   -0.673   -0.600
+##     vrtrlg            0.252    0.006   41.294    0.000    0.462    0.443
 ##   et =~                                                                 
-##     imwgdwn           1.000                               0.728    0.671
-##     imhecop           1.137    0.023   50.455    0.000    0.827    0.766
+##     imwgdwn           1.000                               0.723    0.667
+##     imhecop           1.151    0.023   49.736    0.000    0.832    0.771
 ## 
 ## Covariances:
 ##                    Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
 ##   ip ~~                                                                 
-##     st               -0.612    0.013  -48.650    0.000   -0.603   -0.603
-##     rp                0.260    0.006   45.201    0.000    0.649    0.649
-##     ct                0.634    0.016   40.791    0.000    0.548    0.548
-##     et               -0.208    0.006  -35.659    0.000   -0.463   -0.463
+##     st               -0.605    0.012  -48.693    0.000   -0.610   -0.610
+##     rp                0.264    0.006   45.566    0.000    0.648    0.648
+##     ct                0.634    0.015   41.007    0.000    0.560    0.560
+##     et               -0.206    0.006  -35.411    0.000   -0.462   -0.462
 ##   st ~~                                                                 
-##     rp               -0.815    0.017  -47.420    0.000   -0.764   -0.764
-##     ct               -1.543    0.041  -37.283    0.000   -0.501   -0.501
-##     et                0.704    0.018   39.849    0.000    0.588    0.588
+##     rp               -0.838    0.017  -48.329    0.000   -0.792   -0.792
+##     ct               -1.622    0.041  -39.091    0.000   -0.550   -0.550
+##     et                0.675    0.017   39.083    0.000    0.580    0.580
 ##   rp ~~                                                                 
-##     ct                0.625    0.018   34.830    0.000    0.514    0.514
-##     et               -0.232    0.007  -33.112    0.000   -0.492   -0.492
+##     ct                0.626    0.018   34.950    0.000    0.518    0.518
+##     et               -0.233    0.007  -33.007    0.000   -0.490   -0.490
 ##   ct ~~                                                                 
-##     et               -0.583    0.020  -29.517    0.000   -0.427   -0.427
+##     et               -0.592    0.020  -30.127    0.000   -0.446   -0.446
 ## 
 ## Variances:
 ##                    Estimate  Std.Err  z-value  P(>|z|)   Std.lv  Std.all
-##    .imrcntr           0.299    0.004   77.940    0.000    0.299    0.440
+##    .imrcntr           0.299    0.004   77.941    0.000    0.299    0.440
 ##    .eimrcnt           0.359    0.005   79.638    0.000    0.359    0.515
-##    .eimpcnt           0.116    0.002   62.833    0.000    0.116    0.193
-##    .imsmetn           0.212    0.003   75.579    0.000    0.212    0.366
-##    .impcntr           0.119    0.002   62.453    0.000    0.119    0.191
-##    .imdfetn           0.132    0.002   65.358    0.000    0.132    0.213
-##    .imbgeco           2.268    0.033   69.759    0.000    2.268    0.456
-##    .imbleco           2.814    0.037   76.364    0.000    2.814    0.603
-##    .imwbcnt           1.681    0.027   62.609    0.000    1.681    0.361
-##    .imwbcrm           2.783    0.035   78.457    0.000    2.783    0.672
-##    .imtcjob           2.622    0.034   77.160    0.000    2.622    0.627
-##    .imueclt           2.802    0.039   72.046    0.000    2.802    0.498
-##    .gvrfgap           0.746    0.010   73.853    0.000    0.746    0.639
-##    .imrsprc           0.793    0.010   77.015    0.000    0.793    0.716
-##    .rfgbfml           0.877    0.012   74.258    0.000    0.877    0.648
-##    .rfggvfn           0.783    0.010   76.834    0.000    0.783    0.711
-##    .rfgawrk           0.943    0.012   80.731    0.000    0.943    0.835
-##    .rfgfrpc           0.749    0.010   77.341    0.000    0.749    0.725
-##    .shrrfg            0.670    0.009   72.132    0.000    0.670    0.604
-##    .qfimchr           5.059    0.081   62.712    0.000    5.059    0.590
-##    .qfimwht           3.806    0.065   58.946    0.000    3.806    0.551
-##    .pplstrd           0.828    0.012   68.340    0.000    0.828    0.658
-##    .vrtrlg            0.867    0.011   76.703    0.000    0.867    0.798
-##    .imwgdwn           0.645    0.012   53.096    0.000    0.645    0.549
-##    .imhecop           0.481    0.013   35.709    0.000    0.481    0.413
-##     ip                0.381    0.007   51.585    0.000    1.000    1.000
-##     st                2.702    0.056   48.680    0.000    1.000    1.000
-##     rp                0.421    0.012   35.968    0.000    1.000    1.000
-##     ct                3.515    0.098   35.798    0.000    1.000    1.000
-##     et                0.529    0.015   35.342    0.000    1.000    1.000
      
+##    .eimpcnt           0.116    0.002   62.821    0.000    0.116    0.193
+##    .imsmetn           0.212    0.003   75.580    0.000    0.212    0.366
+##    .impcntr           0.119    0.002   62.454    0.000    0.119    0.191
+##    .imdfetn           0.132    0.002   65.344    0.000    0.132    0.213
+##    .imbgeco           2.288    0.033   70.261    0.000    2.288    0.470
+##    .imbleco           2.837    0.037   76.688    0.000    2.837    0.617
+##    .imwbcnt           1.677    0.027   63.198    0.000    1.677    0.372
+##    .imwbcrm           2.810    0.036   78.612    0.000    2.810    0.682
+##    .imtcjob           2.630    0.034   77.524    0.000    2.630    0.643
+##    .imueclt           2.761    0.038   72.515    0.000    2.761    0.512
+##    .gvrfgap           0.733    0.010   73.584    0.000    0.733    0.628
+##    .imrsprc           0.791    0.010   77.119    0.000    0.791    0.714
+##    .rfgbfml           0.877    0.012   74.508    0.000    0.877    0.648
+##    .rfggvfn           0.788    0.010   77.203    0.000    0.788    0.716
+##    .rfgawrk           0.945    0.012   80.870    0.000    0.945    0.836
+##    .rfgfrpc           0.749    0.010   77.501    0.000    0.749    0.724
+##    .shrrfg            0.676    0.009   72.682    0.000    0.676    0.609
+##    .qfimchr           5.142    0.080   64.113    0.000    5.142    0.604
+##    .qfimwht           3.891    0.064   60.623    0.000    3.891    0.566
+##    .pplstrd           0.804    0.012   67.054    0.000    0.804    0.640
+##    .vrtrlg            0.872    0.011   76.990    0.000    0.872    0.804
+##    .imwgdwn           0.652    0.012   53.300    0.000    0.652    0.555
+##    .imhecop           0.472    0.014   34.353    0.000    0.472    0.405
+##     ip                0.381    0.007   51.578    0.000    1.000    1.000
+##     st                2.584    0.054   47.795    0.000    1.000    1.000
+##     rp                0.434    0.012   36.748    0.000    1.000    1.000
+##     ct                3.371    0.096   35.174    0.000    1.000    1.000
+##     et                0.523    0.015   34.944    0.000    1.000    1.000
      Click for explanation @@ -744,12 +741,12 @@

      5.4.4

      Click to show code -
      mod_1f <- '
-ati =~ imrcntr + eimrcnt + eimpcnt + imsmetn + impcntr + imdfetn + imbgeco +
-       imbleco + imwbcnt + imwbcrm + imtcjob + imueclt + gvrfgap + imrsprc +
-       rfgbfml + rfggvfn + rfgawrk + rfgfrpc + shrrfg  + qfimchr + qfimwht +
-       pplstrd + vrtrlg  + imwgdwn + imhecop
-'
      +
      mod_1f <- '
+ati =~ imrcntr + eimrcnt + eimpcnt + imsmetn + impcntr + imdfetn + imbgeco +
+       imbleco + imwbcnt + imwbcrm + imtcjob + imueclt + gvrfgap + imrsprc +
+       rfgbfml + rfggvfn + rfgawrk + rfgfrpc + shrrfg  + qfimchr + qfimwht +
+       pplstrd + vrtrlg  + imwgdwn + imhecop
+'
    @@ -766,12 +763,12 @@

    5.4.5

    Click to show code -
    ## Estimate the one factor model:
-fit_1f <- cfa(mod_1f, data = ess)
-
-## Summarize the results:
-summary(fit_1f)
    -
    ## lavaan 0.6.16 ended normally after 48 iterations
+
    ## Estimate the one factor model:
+fit_1f <- cfa(mod_1f, data = ess)
+
+## Summarize the results:
+summary(fit_1f)
    
+
    ## lavaan 0.6.16 ended normally after 47 iterations
 ## 
 ##   Estimator                                         ML
 ##   Optimization method                           NLMINB
@@ -782,7 +779,7 @@ 
    5.4.5 
    
 ## 
 ## Model Test User Model:
 ##                                                        
-##   Test statistic                              50540.369
+##   Test statistic                              49510.917
 ##   Degrees of freedom                                275
 ##   P-value (Chi-square)                            0.000
 ## 
@@ -796,77 +793,77 @@ 
    5.4.5 
    
 ##                    Estimate  Std.Err  z-value  P(>|z|)
 ##   ati =~                                              
 ##     imrcntr           1.000                           
-##     eimrcnt           0.937    0.012   78.516    0.000
-##     eimpcnt           1.112    0.011  101.485    0.000
-##     imsmetn           0.987    0.011   91.214    0.000
-##     impcntr           1.146    0.011  102.640    0.000
-##     imdfetn           1.151    0.011  103.290    0.000
-##     imbgeco          -2.080    0.032  -65.037    0.000
-##     imbleco          -1.651    0.031  -53.067    0.000
-##     imwbcnt          -2.200    0.031  -71.199    0.000
-##     imwbcrm          -1.455    0.029  -49.601    0.000
-##     imtcjob          -1.559    0.029  -52.963    0.000
-##     imueclt          -2.230    0.034  -65.546    0.000
-##     gvrfgap           0.793    0.016   50.940    0.000
-##     imrsprc           0.752    0.015   49.544    0.000
-##     rfgbfml           0.857    0.017   51.125    0.000
-##     rfggvfn           0.724    0.015   47.876    0.000
-##     rfgawrk           0.530    0.015   34.483    0.000
-##     rfgfrpc          -0.750    0.015  -51.225    0.000
-##     shrrfg           -0.931    0.015  -61.532    0.000
-##     qfimchr           1.556    0.042   36.793    0.000
-##     qfimwht           1.730    0.038   45.639    0.000
-##     pplstrd          -0.872    0.016  -54.018    0.000
-##     vrtrlg            0.591    0.015   39.305    0.000
-##     imwgdwn          -0.686    0.016  -43.889    0.000
-##     imhecop          -0.774    0.016  -49.761    0.000
+##     eimrcnt           0.937    0.012   78.324    0.000
+##     eimpcnt           1.114    0.011  101.263    0.000
+##     imsmetn           0.987    0.011   90.990    0.000
+##     impcntr           1.147    0.011  102.371    0.000
+##     imdfetn           1.153    0.011  103.148    0.000
+##     imbgeco          -2.055    0.032  -64.749    0.000
+##     imbleco          -1.625    0.031  -52.533    0.000
+##     imwbcnt          -2.173    0.030  -71.324    0.000
+##     imwbcrm          -1.432    0.029  -48.849    0.000
+##     imtcjob          -1.532    0.029  -52.519    0.000
+##     imueclt          -2.198    0.033  -65.876    0.000
+##     gvrfgap           0.807    0.016   51.746    0.000
+##     imrsprc           0.757    0.015   49.790    0.000
+##     rfgbfml           0.861    0.017   51.272    0.000
+##     rfggvfn           0.722    0.015   47.671    0.000
+##     rfgawrk           0.530    0.015   34.448    0.000
+##     rfgfrpc          -0.755    0.015  -51.462    0.000
+##     shrrfg           -0.931    0.015  -61.438    0.000
+##     qfimchr           1.597    0.042   37.835    0.000
+##     qfimwht           1.769    0.038   46.697    0.000
+##     pplstrd          -0.873    0.016  -53.994    0.000
+##     vrtrlg            0.602    0.015   39.940    0.000
+##     imwgdwn          -0.682    0.016  -43.576    0.000
+##     imhecop          -0.773    0.016  -49.611    0.000
 ## 
 ## Variances:
 ##                    Estimate  Std.Err  z-value  P(>|z|)
-##    .imrcntr           0.326    0.004   78.988    0.000
-##    .eimrcnt           0.387    0.005   80.403    0.000
-##    .eimpcnt           0.161    0.002   70.807    0.000
-##    .imsmetn           0.235    0.003   77.068    0.000
-##    .impcntr           0.157    0.002   69.617    0.000
-##    .imdfetn           0.151    0.002   68.870    0.000
-##    .imbgeco           3.438    0.042   82.179    0.000
-##    .imbleco           3.704    0.045   83.099    0.000
-##    .imwbcnt           2.943    0.036   81.500    0.000
-##    .imwbcrm           3.395    0.041   83.296    0.000
-##    .imtcjob           3.321    0.040   83.105    0.000
-##    .imueclt           3.865    0.047   82.130    0.000
-##    .gvrfgap           0.945    0.011   83.222    0.000
-##    .imrsprc           0.909    0.011   83.299    0.000
-##    .rfgbfml           1.094    0.013   83.212    0.000
-##    .rfggvfn           0.916    0.011   83.385    0.000
-##    .rfgawrk           1.031    0.012   83.912    0.000
-##    .rfgfrpc           0.834    0.010   83.206    0.000
-##    .shrrfg            0.802    0.010   82.494    0.000
-##    .qfimchr           7.717    0.092   83.839    0.000
-##    .qfimwht           5.851    0.070   83.492    0.000
+##    .imrcntr           0.327    0.004   79.021    0.000
+##    .eimrcnt           0.388    0.005   80.422    0.000
+##    .eimpcnt           0.161    0.002   70.832    0.000
+##    .imsmetn           0.235    0.003   77.101    0.000
+##    .impcntr           0.158    0.002   69.688    0.000
+##    .imdfetn           0.150    0.002   68.791    0.000
+##    .imbgeco           3.381    0.041   82.203    0.000
+##    .imbleco           3.666    0.044   83.130    0.000
+##    .imwbcnt           2.839    0.035   81.477    0.000
+##    .imwbcrm           3.399    0.041   83.334    0.000
+##    .imtcjob           3.260    0.039   83.130    0.000
+##    .imueclt           3.683    0.045   82.092    0.000
+##    .gvrfgap           0.938    0.011   83.176    0.000
+##    .imrsprc           0.906    0.011   83.285    0.000
+##    .rfgbfml           1.092    0.013   83.203    0.000
+##    .rfggvfn           0.917    0.011   83.394    0.000
+##    .rfgawrk           1.031    0.012   83.913    0.000
+##    .rfgfrpc           0.832    0.010   83.192    0.000
+##    .shrrfg            0.803    0.010   82.499    0.000
+##    .qfimchr           7.613    0.091   83.803    0.000
+##    .qfimwht           5.772    0.069   83.442    0.000
 ##    .pplstrd           0.988    0.012   83.040    0.000
-##    .vrtrlg            0.962    0.011   83.752    0.000
-##    .imwgdwn           1.008    0.012   83.570    0.000
-##    .imhecop           0.953    0.011   83.287    0.000
-##     ati               0.354    0.007   49.032    0.000
    
-
    ## Compare fit statistics:
-fitMeasures(fit_5f,
-            fit.measures = c("npar",                  # Estimated parameters
-                             "chisq", "df", "pvalue", # Model fit vs. saturated
-                             "cfi", "tli",            # Model fit vs. baseline
-                             "rmsea", "srmr"),        # Model fit vs. saturated
-            output = "text")
    
+##    .vrtrlg            0.958    0.011   83.728    0.000
+##    .imwgdwn           1.010    0.012   83.583    0.000
+##    .imhecop           0.954    0.011   83.294    0.000
+##     ati               0.353    0.007   48.941    0.000
    +
    ## Compare fit statistics:
+fitMeasures(fit_5f,
+            fit.measures = c("npar",                  # Estimated parameters
+                             "chisq", "df", "pvalue", # Model fit vs. saturated
+                             "cfi", "tli",            # Model fit vs. baseline
+                             "rmsea", "srmr"),        # Model fit vs. saturated
+            output = "text")
    ## 
 ## Model Test User Model:
 ## 
-##   Test statistic                             18630.545
+##   Test statistic                             18631.556
 ##   Degrees of freedom                               265
 ##   P-value                                        0.000
 ## 
 ## User Model versus Baseline Model:
 ## 
 ##   Comparative Fit Index (CFI)                    0.885
-##   Tucker-Lewis Index (TLI)                       0.870
+##   Tucker-Lewis Index (TLI)                       0.869
 ## 
 ## Root Mean Square Error of Approximation:
 ## 
@@ -874,32 +871,32 @@ 
    5.4.5 
    
 ## 
 ## Standardized Root Mean Square Residual:
 ## 
-##   SRMR                                           0.049
    -
    fitMeasures(fit_1f,
-            fit.measures = c("npar",                  # Estimated parameters
-                             "chisq", "df", "pvalue", # Model fit vs. saturated
-                             "cfi", "tli",            # Model fit vs. baseline
-                             "rmsea", "srmr"),        # Model fit vs. saturated
-            output = "text")
    +## SRMR 0.048 +
    fitMeasures(fit_1f,
+            fit.measures = c("npar",                  # Estimated parameters
+                             "chisq", "df", "pvalue", # Model fit vs. saturated
+                             "cfi", "tli",            # Model fit vs. baseline
+                             "rmsea", "srmr"),        # Model fit vs. saturated
+            output = "text")
    ## 
 ## Model Test User Model:
 ## 
-##   Test statistic                             50540.369
+##   Test statistic                             49510.917
 ##   Degrees of freedom                               275
 ##   P-value                                        0.000
 ## 
 ## User Model versus Baseline Model:
 ## 
-##   Comparative Fit Index (CFI)                    0.686
-##   Tucker-Lewis Index (TLI)                       0.658
+##   Comparative Fit Index (CFI)                    0.691
+##   Tucker-Lewis Index (TLI)                       0.663
 ## 
 ## Root Mean Square Error of Approximation:
 ## 
-##   RMSEA                                          0.113
+##   RMSEA                                          0.112
 ## 
 ## Standardized Root Mean Square Residual:
 ## 
-##   SRMR                                           0.088
    +## SRMR 0.087
    Click for explanation @@ -909,7 +906,7 @@

    5.4.5

    -
    +

    5.4.6

    Given the CFA results from the five factor model, would a second-order CFA be appropriate for the Attitudes towards Immigration data? Why or why not?

    @@ -939,13 +936,13 @@

    5.4.7

    Click to show code -
    mod_2o <- paste(mod_5f,
-                'ati =~ ip + rp + st + ct + et',
-                sep = '\n')
-
-fit_2o <- cfa(mod_2o, data = ess)
-summary(fit_2o, fit.measures = TRUE)
    -
    ## lavaan 0.6.16 ended normally after 97 iterations
+
    mod_2o <- paste(mod_5f,
+                'ati =~ ip + rp + st + ct + et',
+                sep = '\n')
+
+fit_2o <- cfa(mod_2o, data = ess)
+summary(fit_2o, fit.measures = TRUE)
    
+
    ## lavaan 0.6.16 ended normally after 94 iterations
 ## 
 ##   Estimator                                         ML
 ##   Optimization method                           NLMINB
@@ -956,13 +953,13 @@ 
    5.4.7 
    
 ## 
 ## Model Test User Model:
 ##                                                        
-##   Test statistic                              19103.443
+##   Test statistic                              19121.111
 ##   Degrees of freedom                                270
 ##   P-value (Chi-square)                            0.000
 ## 
 ## Model Test Baseline Model:
 ## 
-##   Test statistic                            160506.963
+##   Test statistic                            159619.058
 ##   Degrees of freedom                               300
 ##   P-value                                        0.000
 ## 
@@ -973,12 +970,12 @@ 
    5.4.7 
    
 ## 
 ## Loglikelihood and Information Criteria:
 ## 
-##   Loglikelihood user model (H0)            -520896.384
-##   Loglikelihood unrestricted model (H1)    -511344.662
+##   Loglikelihood user model (H0)            -520279.910
+##   Loglikelihood unrestricted model (H1)    -510719.354
 ##                                                       
-##   Akaike (AIC)                             1041902.767
-##   Bayesian (BIC)                           1042318.788
-##   Sample-size adjusted Bayesian (SABIC)    1042144.003
+##   Akaike (AIC)                             1040669.820
+##   Bayesian (BIC)                           1041085.841
+##   Sample-size adjusted Bayesian (SABIC)    1040911.056
 ## 
 ## Root Mean Square Error of Approximation:
 ## 
@@ -990,7 +987,7 @@ 
    5.4.7 
    
 ## 
 ## Standardized Root Mean Square Residual:
 ## 
-##   SRMR                                           0.049
+##   SRMR                                           0.048
 ## 
 ## Parameter Estimates:
 ## 
@@ -1002,78 +999,78 @@ 
    5.4.7 
    
 ##                    Estimate  Std.Err  z-value  P(>|z|)
 ##   ip =~                                               
 ##     imrcntr           1.000                           
-##     eimrcnt           0.943    0.011   85.103    0.000
-##     eimpcnt           1.126    0.010  113.541    0.000
-##     imsmetn           0.982    0.010   98.923    0.000
-##     impcntr           1.149    0.010  113.687    0.000
-##     imdfetn           1.130    0.010  111.787    0.000
+##     eimrcnt           0.943    0.011   85.095    0.000
+##     eimpcnt           1.126    0.010  113.523    0.000
+##     imsmetn           0.982    0.010   98.910    0.000
+##     impcntr           1.149    0.010  113.651    0.000
+##     imdfetn           1.130    0.010  111.789    0.000
 ##   st =~                                               
 ##     imbgeco           1.000                           
-##     imbleco           0.824    0.012   70.853    0.000
-##     imwbcnt           1.051    0.012   90.406    0.000
-##     imwbcrm           0.707    0.011   64.414    0.000
-##     imtcjob           0.754    0.011   68.469    0.000
-##     imueclt           1.029    0.013   80.734    0.000
+##     imbleco           0.822    0.012   68.916    0.000
+##     imwbcnt           1.047    0.012   88.172    0.000
+##     imwbcrm           0.709    0.011   62.846    0.000
+##     imtcjob           0.747    0.011   66.424    0.000
+##     imueclt           1.013    0.013   78.434    0.000
 ##   rp =~                                               
 ##     gvrfgap           1.000                           
-##     imrsprc           0.862    0.017   50.603    0.000
-##     rfgbfml           1.059    0.019   54.809    0.000
-##     rfggvfn           0.868    0.017   51.004    0.000
-##     rfgawrk           0.667    0.016   40.676    0.000
-##     rfgfrpc          -0.835    0.016  -50.724    0.000
-##     shrrfg           -1.033    0.018  -57.819    0.000
+##     imrsprc           0.854    0.017   51.127    0.000
+##     rfgbfml           1.048    0.019   55.377    0.000
+##     rfggvfn           0.853    0.017   51.170    0.000
+##     rfgawrk           0.657    0.016   40.785    0.000
+##     rfgfrpc          -0.828    0.016  -51.249    0.000
+##     shrrfg           -1.020    0.017  -58.369    0.000
 ##   ct =~                                               
 ##     qfimchr           1.000                           
-##     qfimwht           0.937    0.018   52.705    0.000
-##     pplstrd          -0.373    0.007  -50.868    0.000
-##     vrtrlg            0.267    0.006   42.424    0.000
+##     qfimwht           0.939    0.018   51.902    0.000
+##     pplstrd          -0.389    0.008  -51.072    0.000
+##     vrtrlg            0.271    0.006   41.908    0.000
 ##   et =~                                               
 ##     imwgdwn           1.000                           
-##     imhecop           1.146    0.023   49.359    0.000
+##     imhecop           1.158    0.024   48.877    0.000
 ##   ati =~                                              
 ##     ip                1.000                           
-##     rp                1.229    0.023   53.071    0.000
-##     st               -3.072    0.051  -60.780    0.000
-##     ct                2.590    0.057   45.080    0.000
-##     et               -1.008    0.023  -43.069    0.000
+##     rp                1.264    0.024   53.732    0.000
+##     st               -3.123    0.051  -61.058    0.000
+##     ct                2.638    0.058   45.467    0.000
+##     et               -1.000    0.024  -42.490    0.000
 ## 
 ## Variances:
 ##                    Estimate  Std.Err  z-value  P(>|z|)
-##    .imrcntr           0.299    0.004   77.901    0.000
-##    .eimrcnt           0.359    0.005   79.599    0.000
-##    .eimpcnt           0.116    0.002   62.714    0.000
-##    .imsmetn           0.211    0.003   75.507    0.000
-##    .impcntr           0.119    0.002   62.458    0.000
-##    .imdfetn           0.133    0.002   65.444    0.000
-##    .imbgeco           2.267    0.033   69.679    0.000
-##    .imbleco           2.834    0.037   76.468    0.000
-##    .imwbcnt           1.668    0.027   62.244    0.000
-##    .imwbcrm           2.793    0.036   78.490    0.000
-##    .imtcjob           2.645    0.034   77.290    0.000
-##    .imueclt           2.765    0.039   71.650    0.000
-##    .gvrfgap           0.749    0.010   73.863    0.000
-##    .imrsprc           0.798    0.010   77.089    0.000
-##    .rfgbfml           0.885    0.012   74.412    0.000
-##    .rfggvfn           0.786    0.010   76.876    0.000
-##    .rfgawrk           0.944    0.012   80.716    0.000
-##    .rfgfrpc           0.742    0.010   77.025    0.000
-##    .shrrfg            0.663    0.009   71.643    0.000
-##    .qfimchr           5.269    0.082   64.348    0.000
-##    .qfimwht           4.010    0.065   61.244    0.000
-##    .pplstrd           0.797    0.012   65.994    0.000
-##    .vrtrlg            0.850    0.011   75.681    0.000
-##    .imwgdwn           0.649    0.012   52.727    0.000
-##    .imhecop           0.475    0.014   34.383    0.000
-##    .ip                0.173    0.004   43.678    0.000
-##    .st                0.734    0.025   29.357    0.000
-##    .rp                0.103    0.005   22.069    0.000
-##    .ct                1.905    0.063   30.027    0.000
-##    .et                0.313    0.010   31.643    0.000
-##     ati               0.209    0.006   37.582    0.000
    
+##    .imrcntr           0.299    0.004   77.900    0.000
+##    .eimrcnt           0.359    0.005   79.597    0.000
+##    .eimpcnt           0.116    0.002   62.698    0.000
+##    .imsmetn           0.211    0.003   75.502    0.000
+##    .impcntr           0.119    0.002   62.476    0.000
+##    .imdfetn           0.133    0.002   65.406    0.000
+##    .imbgeco           2.285    0.033   70.158    0.000
+##    .imbleco           2.852    0.037   76.762    0.000
+##    .imwbcnt           1.668    0.027   62.920    0.000
+##    .imwbcrm           2.821    0.036   78.653    0.000
+##    .imtcjob           2.646    0.034   77.607    0.000
+##    .imueclt           2.734    0.038   72.213    0.000
+##    .gvrfgap           0.740    0.010   73.738    0.000
+##    .imrsprc           0.797    0.010   77.211    0.000
+##    .rfgbfml           0.885    0.012   74.621    0.000
+##    .rfggvfn           0.791    0.010   77.189    0.000
+##    .rfgawrk           0.946    0.012   80.833    0.000
+##    .rfgfrpc           0.741    0.010   77.149    0.000
+##    .shrrfg            0.665    0.009   72.020    0.000
+##    .qfimchr           5.347    0.081   65.623    0.000
+##    .qfimwht           4.084    0.065   62.673    0.000
+##    .pplstrd           0.778    0.012   64.838    0.000
+##    .vrtrlg            0.854    0.011   75.931    0.000
+##    .imwgdwn           0.655    0.012   52.977    0.000
+##    .imhecop           0.468    0.014   33.353    0.000
+##    .ip                0.177    0.004   44.418    0.000
+##    .st                0.596    0.023   26.030    0.000
+##    .rp                0.101    0.005   21.784    0.000
+##    .ct                1.745    0.060   29.185    0.000
+##    .et                0.316    0.010   31.813    0.000
+##     ati               0.204    0.005   37.371    0.000
    -
    +

    5.4.8

    Compare the model fit of the first- and second-order five-factor models using the fitMeasures() function.

    @@ -1085,23 +1082,23 @@

    5.4.8

    Click to show code -
    fitMeasures(fit_5f,
-  fit.measures = c("npar",                  # Estimated parameters
-                   "chisq", "df", "pvalue", # Model fit vs. saturated
-                   "cfi", "tli",            # Model fit vs. baseline
-                   "rmsea", "srmr"),        # Model fit vs. saturated
-  output = "text")
    +
    fitMeasures(fit_5f,
+  fit.measures = c("npar",                  # Estimated parameters
+                   "chisq", "df", "pvalue", # Model fit vs. saturated
+                   "cfi", "tli",            # Model fit vs. baseline
+                   "rmsea", "srmr"),        # Model fit vs. saturated
+  output = "text")
    ## 
 ## Model Test User Model:
 ## 
-##   Test statistic                             18630.545
+##   Test statistic                             18631.556
 ##   Degrees of freedom                               265
 ##   P-value                                        0.000
 ## 
 ## User Model versus Baseline Model:
 ## 
 ##   Comparative Fit Index (CFI)                    0.885
-##   Tucker-Lewis Index (TLI)                       0.870
+##   Tucker-Lewis Index (TLI)                       0.869
 ## 
 ## Root Mean Square Error of Approximation:
 ## 
@@ -1109,17 +1106,17 @@ 
    5.4.8 
    
 ## 
 ## Standardized Root Mean Square Residual:
 ## 
-##   SRMR                                           0.049
    -
    fitMeasures(fit_2o,
-  fit.measures = c("npar",                  # Estimated parameters
-                   "chisq", "df", "pvalue", # Model fit vs. saturated
-                   "cfi", "tli",            # Model fit vs. baseline
-                   "rmsea", "srmr"),        # Model fit vs. saturated
-  output = "text")
    +## SRMR 0.048 +
    fitMeasures(fit_2o,
+  fit.measures = c("npar",                  # Estimated parameters
+                   "chisq", "df", "pvalue", # Model fit vs. saturated
+                   "cfi", "tli",            # Model fit vs. baseline
+                   "rmsea", "srmr"),        # Model fit vs. saturated
+  output = "text")
    ## 
 ## Model Test User Model:
 ## 
-##   Test statistic                             19103.443
+##   Test statistic                             19121.111
 ##   Degrees of freedom                               270
 ##   P-value                                        0.000
 ## 
@@ -1134,7 +1131,7 @@ 
    5.4.8 
    
 ## 
 ## Standardized Root Mean Square Residual:
 ## 
-##   SRMR                                           0.049
    +## SRMR 0.048
    Click for explanation @@ -1153,7 +1150,7 @@

    5.4.8

    be useful to start thinking about these concepts now. Two models are said to be nested if you can define one model by placing constraints on the other model.

    By way of example, consider the following two CFA models.

    -

    +

    The second model is nested within the first model, because we can define the second model by fixing the latent covariance to zero in the first model.

    Notice that the data contain \(6(6 + 1) / 2 = 21\) unique pieces of information. @@ -1175,7 +1172,7 @@

    5.4.8

    we may be considering, we can always convert that model to a saturated model by estimating all possible associations. Hence, all models are nested within the saturated model.

    -

    +

    Baseline Model

    Similarly, the baseline model (AKA, independence model) is nested within all @@ -1183,7 +1180,7 @@

    5.4.8

    observed items; all associations are constrained to zero. We can always convert our model to the baseline model by fixing all associations to zero. Hence, the baseline model is nested within all other models.

    -

    +

    When two models are nested, we can use a \(\Delta \chi^2\) test to check if the nested model fits significantly worse than its parent model. Whenever we place @@ -1194,7 +1191,7 @@

    5.4.8

    comparing models that we’ve estimated with cfa() or sem().

    -
    +

    5.4.9

    Use the anova() function to compare the five-factor model from 5.4.2 and one-factor model from 5.4.4.

    @@ -1208,10 +1205,10 @@

    5.4.9

    Click to show code -
    anova(fit_1f, fit_5f)
    +
    anova(fit_1f, fit_5f)
    @@ -1230,7 +1227,7 @@

    5.4.9

  • Larger \(\chi^2\) \(\Rightarrow\) Worse fit

    • Chisq diff is the difference between the two \(\chi^2\) values (i.e., -\(\Delta \chi^2\).

    +\(\Delta \chi^2\)).

    • How much better the more complex model fits the data
    • Larger \(\Delta \chi^2\) values indicate greater losses of fit induced by the @@ -1254,7 +1251,7 @@

      5.4.9

    -
    +

    5.4.10

    Use the anova() function to compare the first- and second-order five-factor models from 5.4.2 and 5.4.7.

    @@ -1266,10 +1263,10 @@

    5.4.10

    Click to show code -
    anova(fit_5f, fit_2o)
    +
    anova(fit_5f, fit_2o)
    @@ -1284,7 +1281,7 @@

    5.4.10

    -
    +

    5.4.11

    Based on the results above, would you say that you have successfully confirmed the five-factor structure implied by the EFA?

    @@ -1307,7 +1304,7 @@

    5.4.11

    -
    +

    5.4.12

    Modify the five-factor CFA from 5.4.2 by freeing the following parameters.

    @@ -1328,14 +1325,14 @@

    5.4.12

    Click to show code -
    fit_5f_cov <- paste(mod_5f, 
-                    'imrcntr ~~ eimrcnt', 
-                    'qfimchr ~~ qfimwht', 
-                    sep = '\n') %>%
-  cfa(data = ess)
-
-summary(fit_5f_cov, fit.measures = TRUE)
    -
    ## lavaan 0.6.16 ended normally after 74 iterations
+
    fit_5f_cov <- paste(mod_5f, 
+                    'imrcntr ~~ eimrcnt', 
+                    'qfimchr ~~ qfimwht', 
+                    sep = '\n') %>%
+  cfa(data = ess)
+
+summary(fit_5f_cov, fit.measures = TRUE)
    
+
    ## lavaan 0.6.16 ended normally after 77 iterations
 ## 
 ##   Estimator                                         ML
 ##   Optimization method                           NLMINB
@@ -1346,41 +1343,41 @@ 
    5.4.12 
    
 ## 
 ## Model Test User Model:
 ##                                                       
-##   Test statistic                              9725.604
+##   Test statistic                              9740.512
 ##   Degrees of freedom                               263
 ##   P-value (Chi-square)                           0.000
 ## 
 ## Model Test Baseline Model:
 ## 
-##   Test statistic                            160506.963
+##   Test statistic                            159619.058
 ##   Degrees of freedom                               300
 ##   P-value                                        0.000
 ## 
 ## User Model versus Baseline Model:
 ## 
 ##   Comparative Fit Index (CFI)                    0.941
-##   Tucker-Lewis Index (TLI)                       0.933
+##   Tucker-Lewis Index (TLI)                       0.932
 ## 
 ## Loglikelihood and Information Criteria:
 ## 
-##   Loglikelihood user model (H0)            -516207.464
-##   Loglikelihood unrestricted model (H1)    -511344.662
+##   Loglikelihood user model (H0)            -515589.611
+##   Loglikelihood unrestricted model (H1)    -510719.354
 ##                                                       
-##   Akaike (AIC)                             1032538.927
-##   Bayesian (BIC)                           1033007.897
-##   Sample-size adjusted Bayesian (SABIC)    1032810.866
+##   Akaike (AIC)                             1031303.221
+##   Bayesian (BIC)                           1031772.190
+##   Sample-size adjusted Bayesian (SABIC)    1031575.160
 ## 
 ## Root Mean Square Error of Approximation:
 ## 
 ##   RMSEA                                          0.050
 ##   90 Percent confidence interval - lower         0.049
 ##   90 Percent confidence interval - upper         0.051
-##   P-value H_0: RMSEA <= 0.050                    0.307
+##   P-value H_0: RMSEA <= 0.050                    0.280
 ##   P-value H_0: RMSEA >= 0.080                    0.000
 ## 
 ## Standardized Root Mean Square Residual:
 ## 
-##   SRMR                                           0.035
+##   SRMR                                           0.036
 ## 
 ## Parameter Estimates:
 ## 
@@ -1392,92 +1389,92 @@ 
    5.4.12 
    
 ##                    Estimate  Std.Err  z-value  P(>|z|)
 ##   ip =~                                               
 ##     imrcntr           1.000                           
-##     eimrcnt           0.928    0.007  126.266    0.000
-##     eimpcnt           1.183    0.011  106.527    0.000
-##     imsmetn           1.012    0.011   92.456    0.000
-##     impcntr           1.213    0.011  107.105    0.000
-##     imdfetn           1.181    0.011  104.575    0.000
+##     eimrcnt           0.928    0.007  126.255    0.000
+##     eimpcnt           1.184    0.011  106.508    0.000
+##     imsmetn           1.012    0.011   92.436    0.000
+##     impcntr           1.213    0.011  107.078    0.000
+##     imdfetn           1.181    0.011  104.566    0.000
 ##   st =~                                               
 ##     imbgeco           1.000                           
-##     imbleco           0.828    0.012   71.006    0.000
-##     imwbcnt           1.053    0.012   90.311    0.000
-##     imwbcrm           0.711    0.011   64.656    0.000
-##     imtcjob           0.759    0.011   68.719    0.000
-##     imueclt           1.029    0.013   80.517    0.000
+##     imbleco           0.826    0.012   69.006    0.000
+##     imwbcnt           1.050    0.012   88.051    0.000
+##     imwbcrm           0.715    0.011   63.128    0.000
+##     imtcjob           0.751    0.011   66.542    0.000
+##     imueclt           1.015    0.013   78.256    0.000
 ##   rp =~                                               
 ##     gvrfgap           1.000                           
-##     imrsprc           0.868    0.017   51.183    0.000
-##     rfgbfml           1.062    0.019   55.299    0.000
-##     rfggvfn           0.869    0.017   51.354    0.000
-##     rfgawrk           0.666    0.016   40.828    0.000
-##     rfgfrpc          -0.825    0.016  -50.578    0.000
-##     shrrfg           -1.024    0.018  -57.852    0.000
+##     imrsprc           0.858    0.017   51.965    0.000
+##     rfgbfml           1.046    0.019   56.104    0.000
+##     rfggvfn           0.848    0.016   51.644    0.000
+##     rfgawrk           0.652    0.016   40.998    0.000
+##     rfgfrpc          -0.813    0.016  -51.233    0.000
+##     shrrfg           -1.002    0.017  -58.499    0.000
 ##   ct =~                                               
 ##     qfimchr           1.000                           
-##     qfimwht           0.978    0.020   48.007    0.000
-##     pplstrd          -0.592    0.015  -40.195    0.000
-##     vrtrlg            0.399    0.011   36.002    0.000
+##     qfimwht           0.979    0.020   48.332    0.000
+##     pplstrd          -0.586    0.014  -40.685    0.000
+##     vrtrlg            0.397    0.011   36.273    0.000
 ##   et =~                                               
 ##     imwgdwn           1.000                           
-##     imhecop           1.141    0.023   50.347    0.000
+##     imhecop           1.157    0.023   49.549    0.000
 ## 
 ## Covariances:
 ##                    Estimate  Std.Err  z-value  P(>|z|)
 ##  .imrcntr ~~                                          
-##    .eimrcnt           0.230    0.004   59.902    0.000
+##    .eimrcnt           0.230    0.004   59.907    0.000
 ##  .qfimchr ~~                                          
-##    .qfimwht           2.617    0.064   40.671    0.000
+##    .qfimwht           2.558    0.064   40.233    0.000
 ##   ip ~~                                               
-##     st               -0.586    0.012  -47.994    0.000
-##     rp                0.251    0.006   44.837    0.000
-##     ct                0.461    0.014   34.074    0.000
-##     et               -0.199    0.006  -35.351    0.000
+##     st               -0.580    0.012  -48.041    0.000
+##     rp                0.255    0.006   45.185    0.000
+##     ct                0.467    0.014   34.425    0.000
+##     et               -0.197    0.006  -35.077    0.000
 ##   st ~~                                               
-##     rp               -0.812    0.017  -47.392    0.000
-##     ct               -1.349    0.039  -34.298    0.000
-##     et                0.700    0.018   39.737    0.000
+##     rp               -0.835    0.017  -48.285    0.000
+##     ct               -1.394    0.040  -35.128    0.000
+##     et                0.670    0.017   38.935    0.000
 ##   rp ~~                                               
-##     ct                0.530    0.016   32.121    0.000
-##     et               -0.232    0.007  -33.081    0.000
+##     ct                0.538    0.017   32.407    0.000
+##     et               -0.232    0.007  -32.949    0.000
 ##   ct ~~                                               
-##     et               -0.465    0.017  -27.756    0.000
+##     et               -0.469    0.017  -27.959    0.000
 ## 
 ## Variances:
 ##                    Estimate  Std.Err  z-value  P(>|z|)
-##    .imrcntr           0.330    0.004   78.901    0.000
+##    .imrcntr           0.330    0.004   78.903    0.000
 ##    .eimrcnt           0.396    0.005   80.392    0.000
-##    .eimpcnt           0.109    0.002   60.413    0.000
-##    .imsmetn           0.220    0.003   75.978    0.000
-##    .impcntr           0.107    0.002   58.862    0.000
-##    .imdfetn           0.131    0.002   64.653    0.000
-##    .imbgeco           2.280    0.033   70.047    0.000
-##    .imbleco           2.825    0.037   76.519    0.000
-##    .imwbcnt           1.673    0.027   62.636    0.000
-##    .imwbcrm           2.783    0.035   78.509    0.000
-##    .imtcjob           2.633    0.034   77.307    0.000
-##    .imueclt           2.777    0.039   71.947    0.000
-##    .gvrfgap           0.747    0.010   74.008    0.000
-##    .imrsprc           0.792    0.010   77.058    0.000
-##    .rfgbfml           0.880    0.012   74.449    0.000
-##    .rfggvfn           0.784    0.010   76.968    0.000
-##    .rfgawrk           0.944    0.012   80.793    0.000
-##    .rfgfrpc           0.747    0.010   77.367    0.000
-##    .shrrfg            0.669    0.009   72.185    0.000
-##    .qfimchr           6.887    0.091   75.331    0.000
-##    .qfimwht           5.297    0.072   73.350    0.000
-##    .pplstrd           0.667    0.013   51.552    0.000
-##    .vrtrlg            0.817    0.011   72.865    0.000
-##    .imwgdwn           0.647    0.012   53.266    0.000
-##    .imhecop           0.478    0.014   35.353    0.000
-##     ip                0.350    0.007   48.654    0.000
-##     st                2.689    0.055   48.557    0.000
-##     rp                0.420    0.012   35.958    0.000
-##     ct                1.687    0.073   23.045    0.000
-##     et                0.527    0.015   35.255    0.000
    
+##    .eimpcnt           0.109    0.002   60.401    0.000
+##    .imsmetn           0.220    0.003   75.979    0.000
+##    .impcntr           0.107    0.002   58.874    0.000
+##    .imdfetn           0.131    0.002   64.630    0.000
+##    .imbgeco           2.301    0.033   70.568    0.000
+##    .imbleco           2.845    0.037   76.832    0.000
+##    .imwbcnt           1.669    0.026   63.272    0.000
+##    .imwbcrm           2.808    0.036   78.659    0.000
+##    .imtcjob           2.639    0.034   77.663    0.000
+##    .imueclt           2.741    0.038   72.463    0.000
+##    .gvrfgap           0.734    0.010   73.743    0.000
+##    .imrsprc           0.790    0.010   77.164    0.000
+##    .rfgbfml           0.880    0.012   74.676    0.000
+##    .rfggvfn           0.790    0.010   77.322    0.000
+##    .rfgawrk           0.946    0.012   80.924    0.000
+##    .rfgfrpc           0.747    0.010   77.519    0.000
+##    .shrrfg            0.674    0.009   72.713    0.000
+##    .qfimchr           6.815    0.090   75.362    0.000
+##    .qfimwht           5.250    0.072   73.378    0.000
+##    .pplstrd           0.674    0.013   52.766    0.000
+##    .vrtrlg            0.818    0.011   73.191    0.000
+##    .imwgdwn           0.655    0.012   53.496    0.000
+##    .imhecop           0.468    0.014   33.845    0.000
+##     ip                0.350    0.007   48.646    0.000
+##     st                2.571    0.054   47.662    0.000
+##     rp                0.433    0.012   36.718    0.000
+##     ct                1.698    0.073   23.296    0.000
+##     et                0.520    0.015   34.814    0.000
    -
    +

    5.4.13

    Evaluate the model modifications.

      @@ -1488,45 +1485,45 @@

      5.4.13

      Click to show code -
      anova(fit_5f_cov, fit_5f)
      +
      anova(fit_5f_cov, fit_5f)
      -
      fitMeasures(fit_5f_cov)
      +
      fitMeasures(fit_5f_cov)
      ##                  npar                  fmin                 chisq 
-##                62.000                 0.341              9725.604 
+##                62.000                 0.342              9740.512 
 ##                    df                pvalue        baseline.chisq 
-##               263.000                 0.000            160506.963 
+##               263.000                 0.000            159619.058 
 ##           baseline.df       baseline.pvalue                   cfi 
 ##               300.000                 0.000                 0.941 
 ##                   tli                  nnfi                   rfi 
-##                 0.933                 0.933                 0.931 
+##                 0.932                 0.932                 0.930 
 ##                   nfi                  pnfi                   ifi 
-##                 0.939                 0.824                 0.941 
+##                 0.939                 0.823                 0.941 
 ##                   rni                  logl     unrestricted.logl 
-##                 0.941           -516207.464           -511344.662 
+##                 0.941           -515589.611           -510719.354 
 ##                   aic                   bic                ntotal 
-##           1032538.927           1033007.897             14243.000 
+##           1031303.221           1031772.190             14243.000 
 ##                  bic2                 rmsea        rmsea.ci.lower 
-##           1032810.866                 0.050                 0.049 
+##           1031575.160                 0.050                 0.049 
 ##        rmsea.ci.upper        rmsea.ci.level          rmsea.pvalue 
-##                 0.051                 0.900                 0.307 
+##                 0.051                 0.900                 0.280 
 ##        rmsea.close.h0 rmsea.notclose.pvalue     rmsea.notclose.h0 
 ##                 0.050                 0.000                 0.080 
 ##                   rmr            rmr_nomean                  srmr 
-##                 0.103                 0.103                 0.035 
+##                 0.103                 0.103                 0.036 
 ##          srmr_bentler   srmr_bentler_nomean                  crmr 
-##                 0.035                 0.035                 0.037 
+##                 0.036                 0.036                 0.037 
 ##           crmr_nomean            srmr_mplus     srmr_mplus_nomean 
-##                 0.037                 0.035                 0.035 
+##                 0.037                 0.036                 0.036 
 ##                 cn_05                 cn_01                   gfi 
-##               443.020               468.574                 0.944 
+##               442.344               467.858                 0.944 
 ##                  agfi                  pgfi                   mfi 
 ##                 0.931                 0.764                 0.717 
 ##                  ecvi 
-##                 0.692
      +## 0.693
      Click for explanation @@ -1577,7 +1574,7 @@

      5.4.13

      ``` ``` -## lavaan 0.6.16 ended normally after 74 iterations +## lavaan 0.6.16 ended normally after 77 iterations ## ## Estimator ML ## Optimization method NLMINB @@ -1588,7 +1585,7 @@

      5.4.13

      ## ## Model Test User Model: ## -## Test statistic 9725.604 +## Test statistic 9740.512 ## Degrees of freedom 263 ## P-value (Chi-square) 0.000 ## @@ -1602,88 +1599,88 @@

      5.4.13

      ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## ip =~ ## imrcntr 1.000 0.592 0.718 -## eimrcnt 0.928 0.007 126.266 0.000 0.549 0.658 -## eimpcnt 1.183 0.011 106.527 0.000 0.700 0.905 -## imsmetn 1.012 0.011 92.456 0.000 0.599 0.787 -## impcntr 1.213 0.011 107.105 0.000 0.718 0.910 -## imdfetn 1.181 0.011 104.575 0.000 0.699 0.888 +## eimrcnt 0.928 0.007 126.255 0.000 0.549 0.658 +## eimpcnt 1.184 0.011 106.508 0.000 0.700 0.905 +## imsmetn 1.012 0.011 92.436 0.000 0.599 0.787 +## impcntr 1.213 0.011 107.078 0.000 0.718 0.910 +## imdfetn 1.181 0.011 104.566 0.000 0.699 0.888 ## st =~ -## imbgeco 1.000 1.640 0.736 -## imbleco 0.828 0.012 71.006 0.000 1.358 0.628 -## imwbcnt 1.053 0.012 90.311 0.000 1.727 0.800 -## imwbcrm 0.711 0.011 64.656 0.000 1.167 0.573 -## imtcjob 0.759 0.011 68.719 0.000 1.244 0.608 -## imueclt 1.029 0.013 80.517 0.000 1.688 0.712 +## imbgeco 1.000 1.603 0.726 +## imbleco 0.826 0.012 69.006 0.000 1.324 0.617 +## imwbcnt 1.050 0.012 88.051 0.000 1.684 0.793 +## imwbcrm 0.715 0.011 63.128 0.000 1.147 0.565 +## imtcjob 0.751 0.011 66.542 0.000 1.204 0.595 +## imueclt 1.015 0.013 78.256 0.000 1.627 0.701 ## rp =~ -## gvrfgap 1.000 0.648 0.600 -## imrsprc 0.868 0.017 51.183 0.000 0.562 0.534 -## rfgbfml 1.062 0.019 55.299 0.000 0.688 0.592 -## rfggvfn 0.869 0.017 51.354 0.000 0.563 0.536 -## rfgawrk 0.666 0.016 40.828 0.000 0.431 0.406 -## rfgfrpc -0.825 0.016 -50.578 0.000 -0.535 -0.526 -## shrrfg -1.024 0.018 -57.852 0.000 -0.664 -0.630 +## gvrfgap 1.000 0.658 0.609 +## imrsprc 0.858 0.017 51.965 0.000 0.564 0.536 +## rfgbfml 1.046 0.019 56.104 0.000 0.689 0.592 +## rfggvfn 0.848 0.016 51.644 0.000 0.558 0.532 +## rfgawrk 0.652 0.016 40.998 0.000 0.429 0.404 +## rfgfrpc -0.813 0.016 -51.233 0.000 -0.535 -0.526 +## shrrfg -1.002 0.017 -58.499 0.000 -0.660 -0.626 ## ct =~ -## qfimchr 1.000 1.299 0.444 -## qfimwht 0.978 0.020 48.007 0.000 1.270 0.483 -## pplstrd -0.592 0.015 -40.195 0.000 -0.768 -0.685 -## vrtrlg 0.399 0.011 36.002 0.000 0.518 0.497 +## qfimchr 1.000 1.303 0.447 +## qfimwht 0.979 0.020 48.332 0.000 1.275 0.486 +## pplstrd -0.586 0.014 -40.685 0.000 -0.764 -0.681 +## vrtrlg 0.397 0.011 36.273 0.000 0.517 0.496 ## et =~ -## imwgdwn 1.000 0.726 0.670 -## imhecop 1.141 0.023 50.347 0.000 0.829 0.768 +## imwgdwn 1.000 0.721 0.665 +## imhecop 1.157 0.023 49.549 0.000 0.835 0.773 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## .imrcntr ~~ -## .eimrcnt 0.230 0.004 59.902 0.000 0.230 0.636 +## .eimrcnt 0.230 0.004 59.907 0.000 0.230 0.636 ## .qfimchr ~~ -## .qfimwht 2.617 0.064 40.671 0.000 2.617 0.433 +## .qfimwht 2.558 0.064 40.233 0.000 2.558 0.428 ## ip ~~ -## st -0.586 0.012 -47.994 0.000 -0.604 -0.604 -## rp 0.251 0.006 44.837 0.000 0.654 0.654 -## ct 0.461 0.014 34.074 0.000 0.600 0.600 -## et -0.199 0.006 -35.351 0.000 -0.463 -0.463 +## st -0.580 0.012 -48.041 0.000 -0.612 -0.612 +## rp 0.255 0.006 45.185 0.000 0.654 0.654 +## ct 0.467 0.014 34.425 0.000 0.605 0.605 +## et -0.197 0.006 -35.077 0.000 -0.462 -0.462 ## st ~~ -## rp -0.812 0.017 -47.392 0.000 -0.764 -0.764 -## ct -1.349 0.039 -34.298 0.000 -0.633 -0.633 -## et 0.700 0.018 39.737 0.000 0.588 0.588 +## rp -0.835 0.017 -48.285 0.000 -0.792 -0.792 +## ct -1.394 0.040 -35.128 0.000 -0.667 -0.667 +## et 0.670 0.017 38.935 0.000 0.580 0.580 ## rp ~~ -## ct 0.530 0.016 32.121 0.000 0.629 0.629 -## et -0.232 0.007 -33.081 0.000 -0.492 -0.492 +## ct 0.538 0.017 32.407 0.000 0.627 0.627 +## et -0.232 0.007 -32.949 0.000 -0.490 -0.490 ## ct ~~ -## et -0.465 0.017 -27.756 0.000 -0.493 -0.493 +## et -0.469 0.017 -27.959 0.000 -0.499 -0.499 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all -## .imrcntr 0.330 0.004 78.901 0.000 0.330 0.485 +## .imrcntr 0.330 0.004 78.903 0.000 0.330 0.485 ## .eimrcnt 0.396 0.005 80.392 0.000 0.396 0.567 -## .eimpcnt 0.109 0.002 60.413 0.000 0.109 0.181 -## .imsmetn 0.220 0.003 75.978 0.000 0.220 0.381 -## .impcntr 0.107 0.002 58.862 0.000 0.107 0.172 -## .imdfetn 0.131 0.002 64.653 0.000 0.131 0.211 -## .imbgeco 2.280 0.033 70.047 0.000 2.280 0.459 -## .imbleco 2.825 0.037 76.519 0.000 2.825 0.605 -## .imwbcnt 1.673 0.027 62.636 0.000 1.673 0.359 -## .imwbcrm 2.783 0.035 78.509 0.000 2.783 0.672 -## .imtcjob 2.633 0.034 77.307 0.000 2.633 0.630 -## .imueclt 2.777 0.039 71.947 0.000 2.777 0.494 -## .gvrfgap 0.747 0.010 74.008 0.000 0.747 0.640 -## .imrsprc 0.792 0.010 77.058 0.000 0.792 0.715 -## .rfgbfml 0.880 0.012 74.449 0.000 0.880 0.650 -## .rfggvfn 0.784 0.010 76.968 0.000 0.784 0.712 -## .rfgawrk 0.944 0.012 80.793 0.000 0.944 0.835 -## .rfgfrpc 0.747 0.010 77.367 0.000 0.747 0.723 -## .shrrfg 0.669 0.009 72.185 0.000 0.669 0.603 -## .qfimchr 6.887 0.091 75.331 0.000 6.887 0.803 -## .qfimwht 5.297 0.072 73.350 0.000 5.297 0.767 -## .pplstrd 0.667 0.013 51.552 0.000 0.667 0.530 -## .vrtrlg 0.817 0.011 72.865 0.000 0.817 0.753 -## .imwgdwn 0.647 0.012 53.266 0.000 0.647 0.551 -## .imhecop 0.478 0.014 35.353 0.000 0.478 0.410 -## ip 0.350 0.007 48.654 0.000 1.000 1.000 -## st 2.689 0.055 48.557 0.000 1.000 1.000 -## rp 0.420 0.012 35.958 0.000 1.000 1.000 -## ct 1.687 0.073 23.045 0.000 1.000 1.000 -## et 0.527 0.015 35.255 0.000 1.000 1.000 +## .eimpcnt 0.109 0.002 60.401 0.000 0.109 0.181 +## .imsmetn 0.220 0.003 75.979 0.000 0.220 0.381 +## .impcntr 0.107 0.002 58.874 0.000 0.107 0.172 +## .imdfetn 0.131 0.002 64.630 0.000 0.131 0.211 +## .imbgeco 2.301 0.033 70.568 0.000 2.301 0.472 +## .imbleco 2.845 0.037 76.832 0.000 2.845 0.619 +## .imwbcnt 1.669 0.026 63.272 0.000 1.669 0.371 +## .imwbcrm 2.808 0.036 78.659 0.000 2.808 0.681 +## .imtcjob 2.639 0.034 77.663 0.000 2.639 0.646 +## .imueclt 2.741 0.038 72.463 0.000 2.741 0.509 +## .gvrfgap 0.734 0.010 73.743 0.000 0.734 0.629 +## .imrsprc 0.790 0.010 77.164 0.000 0.790 0.713 +## .rfgbfml 0.880 0.012 74.676 0.000 0.880 0.650 +## .rfggvfn 0.790 0.010 77.322 0.000 0.790 0.717 +## .rfgawrk 0.946 0.012 80.924 0.000 0.946 0.837 +## .rfgfrpc 0.747 0.010 77.519 0.000 0.747 0.723 +## .shrrfg 0.674 0.009 72.713 0.000 0.674 0.608 +## .qfimchr 6.815 0.090 75.362 0.000 6.815 0.801 +## .qfimwht 5.250 0.072 73.378 0.000 5.250 0.764 +## .pplstrd 0.674 0.013 52.766 0.000 0.674 0.536 +## .vrtrlg 0.818 0.011 73.191 0.000 0.818 0.754 +## .imwgdwn 0.655 0.012 53.496 0.000 0.655 0.557 +## .imhecop 0.468 0.014 33.845 0.000 0.468 0.402 +## ip 0.350 0.007 48.646 0.000 1.000 1.000 +## st 2.571 0.054 47.662 0.000 1.000 1.000 +## rp 0.433 0.012 36.718 0.000 1.000 1.000 +## ct 1.698 0.073 23.296 0.000 1.000 1.000 +## et 0.520 0.015 34.814 0.000 1.000 1.000 ```
      diff --git a/docs/in-class-exercises-5.html b/docs/in-class-exercises-5.html index b983cab9..9b5fc2be 100644 --- a/docs/in-class-exercises-5.html +++ b/docs/in-class-exercises-5.html @@ -333,28 +333,31 @@
    • 4.3.5
    • 4.3.6
    • 4.3.7
      • -
    • 4.3.8
      • -
    • 4.3.9
      • -
    • 4.3.10
      • +
    • 4.3.8
      • +
    • 4.3.9
      • +
    • 4.3.10
      • +
    • 4.3.11
      • +
    • 4.3.12

  • 4.4 In-Class Exercises
  • 5 CFA @@ -371,30 +374,30 @@
  • 5.3 At-Home Exercises
  • 5.4 In-Class Exercises
  • 6 Full SEM @@ -411,23 +414,22 @@
  • 6.3 At-Home Exercises
  • 6.4 In-Class Exercises
  • 7 Multiple Group Models @@ -470,9 +472,10 @@

    6.4 In-Class Exercisestoradata.csv file. These data were synthesized according to the +results of Reinecke (1998)’s investigation of condom use by young people between +16 and 24 years old.

    The data contain the following variables:

    • respnr: Numeric participant ID
      • @@ -500,18 +503,18 @@

      6.4 In-Class Exercises +

      6.4.1

      Load the data contained in the toradata.csv file.
      Click for explanation -
      condom <- read.csv("toradata.csv", stringsAsFactors = TRUE)
      +
      condom <- read.csv("toradata.csv", stringsAsFactors = TRUE)
      -
      +

      6.4.2

      The data contain multiple indicators of attitudes, norms, and control. Run a CFA for these three latent variables.

      @@ -525,17 +528,17 @@

      6.4.2

      Click for explanation -
      library(lavaan)
-
-mod_cfa <- '
-attitudes =~ attit_1   + attit_2   + attit_3
-norms     =~ norm_1    + norm_2    + norm_3
-control   =~ control_1 + control_2 + control_3
-'
-
-fit <- cfa(mod_cfa, data = condom)
-
-summary(fit, fit.measures = TRUE)
      +
      library(lavaan)
+
+mod_cfa <- '
+attitudes =~ attit_1   + attit_2   + attit_3
+norms     =~ norm_1    + norm_2    + norm_3
+control   =~ control_1 + control_2 + control_3
+'
+
+fit <- cfa(mod_cfa, data = condom)
+
+summary(fit, fit.measures = TRUE)
      ## lavaan 0.6.16 ended normally after 29 iterations
 ## 
 ##   Estimator                                         ML
@@ -642,7 +645,7 @@ 
      6.4.2
      -
      +

      6.4.3

      Estimate the basic TORA model as an SEM.

        @@ -658,19 +661,19 @@

        6.4.3

        Click for explanation -
        mod <- '
-## Define the latent variables:
-attitudes =~ attit_1 + attit_2 + attit_3
-norms     =~ norm_1  + norm_2  + norm_3
-
-## Define the structural model:         
-intent   ~ attitudes + norms
-behavior ~ intent
-'
-
-fit <- sem(mod, data = condom)
-
-summary(fit, fit.measures = TRUE, rsquare = TRUE)
        +
        mod <- '
+## Define the latent variables:
+attitudes =~ attit_1 + attit_2 + attit_3
+norms     =~ norm_1  + norm_2  + norm_3
+
+## Define the structural model:         
+intent   ~ attitudes + norms
+behavior ~ intent
+'
+
+fit <- sem(mod, data = condom)
+
+summary(fit, fit.measures = TRUE, rsquare = TRUE)
        ## lavaan 0.6.16 ended normally after 24 iterations
 ## 
 ##   Estimator                                         ML
@@ -798,18 +801,18 @@ 
        6.4.4 
        
 
 Click for explanation
 
-
        mod_tora <- '
-attitudes =~ attit_1   + attit_2   + attit_3
-norms     =~ norm_1    + norm_2    + norm_3
-control   =~ control_1 + control_2 + control_3
-             
-intent   ~ attitudes + norms
-behavior ~ intent + control
-'
-
-fit_tora <- sem(mod_tora, data = condom)
-
-summary(fit_tora, fit.measures = TRUE, rsquare = TRUE)
        
+
        mod_tora <- '
+attitudes =~ attit_1   + attit_2   + attit_3
+norms     =~ norm_1    + norm_2    + norm_3
+control   =~ control_1 + control_2 + control_3
+             
+intent   ~ attitudes + norms
+behavior ~ intent + control
+'
+
+fit_tora <- sem(mod_tora, data = condom)
+
+summary(fit_tora, fit.measures = TRUE, rsquare = TRUE)
        
 
        ## lavaan 0.6.16 ended normally after 31 iterations
 ## 
 ##   Estimator                                         ML
@@ -942,7 +945,7 @@ 
        6.4.4 
        
 effect may be (partially) mediated by intention.
      -
      +

      6.4.5

      Evaluate the hypothesized indirect effects of attitudes and norms.

        @@ -959,25 +962,98 @@

        6.4.5

        Click for explanation -
        mod <- '
-attitudes =~ attit_1   + attit_2   + attit_3
-norms     =~ norm_1    + norm_2    + norm_3
-control   =~ control_1 + control_2 + control_3
-             
-intent   ~ a1 * attitudes + a2 * norms
-behavior ~ b * intent + control + attitudes + norms
-
-ie_att  := a1 * b
-ie_norm := a2 * b
-'
-
-set.seed(235711)
-
-fit <- sem(mod, data = condom, se = "bootstrap", bootstrap = 1000)
-
-summary(fit, ci = TRUE)
        -
        ## Length  Class   Mode 
-##      1 lavaan     S4
        +
        mod <- '
+attitudes =~ attit_1   + attit_2   + attit_3
+norms     =~ norm_1    + norm_2    + norm_3
+control   =~ control_1 + control_2 + control_3
+             
+intent   ~ a1 * attitudes + a2 * norms
+behavior ~ b * intent + control + attitudes + norms
+
+ie_att  := a1 * b
+ie_norm := a2 * b
+'
+
+set.seed(235711)
+
+fit <- sem(mod, data = condom, se = "bootstrap", bootstrap = 1000)
+
+summary(fit, ci = TRUE)
        +
        ## lavaan 0.6.16 ended normally after 36 iterations
+## 
+##   Estimator                                         ML
+##   Optimization method                           NLMINB
+##   Number of model parameters                        29
+## 
+##   Number of observations                           250
+## 
+## Model Test User Model:
+##                                                       
+##   Test statistic                                48.629
+##   Degrees of freedom                                37
+##   P-value (Chi-square)                           0.096
+## 
+## Parameter Estimates:
+## 
+##   Standard errors                            Bootstrap
+##   Number of requested bootstrap draws             1000
+##   Number of successful bootstrap draws            1000
+## 
+## Latent Variables:
+##                    Estimate  Std.Err  z-value  P(>|z|) ci.lower ci.upper
+##   attitudes =~                                                          
+##     attit_1           1.000                               1.000    1.000
+##     attit_2           1.033    0.060   17.261    0.000    0.925    1.165
+##     attit_3          -1.025    0.064  -15.894    0.000   -1.163   -0.902
+##   norms =~                                                              
+##     norm_1            1.000                               1.000    1.000
+##     norm_2            0.984    0.071   13.794    0.000    0.843    1.127
+##     norm_3            0.955    0.093   10.324    0.000    0.792    1.157
+##   control =~                                                            
+##     control_1         1.000                               1.000    1.000
+##     control_2         0.860    0.113    7.624    0.000    0.653    1.098
+##     control_3         0.996    0.147    6.790    0.000    0.748    1.320
+## 
+## Regressions:
+##                    Estimate  Std.Err  z-value  P(>|z|) ci.lower ci.upper
+##   intent ~                                                              
+##     attitudes (a1)    0.447    0.067    6.674    0.000    0.324    0.585
+##     norms     (a2)    0.706    0.078    9.094    0.000    0.569    0.878
+##   behavior ~                                                            
+##     intent     (b)    0.545    0.075    7.282    0.000    0.389    0.686
+##     control           0.428    0.232    1.847    0.065    0.046    0.934
+##     attitudes         0.010    0.122    0.084    0.933   -0.249    0.226
+##     norms             0.041    0.118    0.345    0.730   -0.194    0.266
+## 
+## Covariances:
+##                    Estimate  Std.Err  z-value  P(>|z|) ci.lower ci.upper
+##   attitudes ~~                                                          
+##     norms             0.342    0.070    4.883    0.000    0.208    0.480
+##     control           0.475    0.069    6.850    0.000    0.344    0.612
+##   norms ~~                                                              
+##     control           0.350    0.067    5.218    0.000    0.221    0.484
+## 
+## Variances:
+##                    Estimate  Std.Err  z-value  P(>|z|) ci.lower ci.upper
+##    .attit_1           0.432    0.050    8.720    0.000    0.331    0.526
+##    .attit_2           0.330    0.045    7.382    0.000    0.238    0.415
+##    .attit_3           0.343    0.049    6.992    0.000    0.244    0.444
+##    .norm_1            0.496    0.060    8.305    0.000    0.376    0.614
+##    .norm_2            0.533    0.077    6.951    0.000    0.390    0.687
+##    .norm_3            0.594    0.069    8.597    0.000    0.443    0.719
+##    .control_1         0.624    0.076    8.216    0.000    0.477    0.763
+##    .control_2         0.875    0.092    9.495    0.000    0.686    1.052
+##    .control_3         0.745    0.079    9.398    0.000    0.574    0.889
+##    .intent            0.409    0.050    8.169    0.000    0.309    0.507
+##    .behavior          0.544    0.058    9.379    0.000    0.415    0.639
+##     attitudes         0.872    0.104    8.387    0.000    0.675    1.077
+##     norms             0.751    0.099    7.557    0.000    0.556    0.941
+##     control           0.486    0.096    5.042    0.000    0.303    0.684
+## 
+## Defined Parameters:
+##                    Estimate  Std.Err  z-value  P(>|z|) ci.lower ci.upper
+##     ie_att            0.244    0.050    4.860    0.000    0.150    0.352
+##     ie_norm           0.385    0.066    5.835    0.000    0.268    0.527
        • Yes, both indirect effects are significant according to the 95% bootstrapped CIs.
        • Yes, both effects are completely moderated by behavioral intention. We can @@ -995,7 +1071,7 @@

          6.4.5

          accurately represent the data, or are they “dead weight”.

      -
      +

      6.4.6

      Use a \(\Delta \chi^2\) test to evaluate the necessity of including the direct effects of attitudes and norms on condom use in the model.

      @@ -1010,7 +1086,7 @@

      6.4.6

      to the fit of the model without the direct effects. We’ve already estimated both models, so we can simply submit the fitted lavaan objects to the anova() function.

      -
      anova(fit, fit_tora)
      +
      anova(fit, fit_tora)
      -
      anova(fit1, fit3)
      +
      anova(fit1, fit3)
      -
      BIC(fit1, fit2, fit3)
      +
      BIC(fit1, fit2, fit3)
      +
      + +Click for explanation +

      The above \(\Delta \chi^2\) tests tell us that the full mediation model fits significantly worse than the partial mediation model. Hence, forcing full mediation by fixing the direct effect to zero is an unreasonable restraint. The @@ -1403,19 +1434,17 @@

      6.4.7

      association represents the data just as well as a model that allows for both indirect and direct effects. Hence, we should prefer the more parsimonious total effects model.

      +
      +
      -

      While the two tests above lead us to prefer the non-mediated model, we cannot -directly say that the the complete mediation model fits significantly worse than -the non-mediated model. We have not directly compared those two models, and we -cannot do so with the \(\Delta \chi^2\). We cannot do such a test because these two -models are not nested: we must both add and remove a path to get from one model -specification to the other. Also, both models have the same degrees of freedom, -so we cannot define a sampling distribution against which we would compare the -\(\Delta \chi^2\), anyway.

      -

      We are not completely without options, though. We can use information criteria -to compare non-nested models. The two most popular information criteria are the -Akaike’s Information Criterion (AIC) and the Bayesian Information Criterion -(BIC).

      +

      Approach 2.2: Non-Nested Model Comparison

      +
      + +Click to show code + +

      We can also use information criteria to compare our models. The two most +popular information criteria are the Akaike’s Information Criterion (AIC) and +the Bayesian Information Criterion (BIC).

      ## Which model is the most parsimonious representation of the data?
 AIC(fit1, fit2, fit3)
      @@ -1429,11 +1458,33 @@

      6.4.7

      {"columns":[{"label":[""],"name":["_rn_"],"type":[""],"align":["left"]},{"label":["df"],"name":[1],"type":["dbl"],"align":["right"]},{"label":["BIC"],"name":[2],"type":["dbl"],"align":["right"]}],"data":[{"1":"28","2":"7255.553","_rn_":"fit1"},{"1":"27","2":"7265.439","_rn_":"fit2"},{"1":"27","2":"7251.400","_rn_":"fit3"}],"options":{"columns":{"min":{},"max":[10]},"rows":{"min":[10],"max":[10]},"pages":{}}}
      -

      When comparing models based on information criteria, a lower value indicates a +

      + +Click for explanation + +

      While the effect tests and the nested model comparisons both lead us to prefer +the non-mediated model, we cannot directly say that the complete mediation +model fits significantly worse than the non-mediated model. We have not directly +compared those two models, and we cannot do so with the \(\Delta \chi^2\). We +cannot do such a test because these two models are not nested: we must both add +and remove a path to get from one model specification to the other. Also, both +models have the same degrees of freedom, so we cannot define a sampling +distribution against which we would compare the \(\Delta \chi^2\), anyway.

      +

      We can use information criteria to get around this problem, though. Information +criteria can be used to compare both nested and non-nested models. These criteria +are designed to rank models by balancing their fit to the data and their complexity. +When comparing models based on information criteria, a lower value indicates a better model in the sense of a better balance of fit and parsimony. The above results show that both the AIC and the BIC agree that the no-mediation model is the best.

      +
      +
      +

      Conclusion

      +
      + +Click for explanation +

      So, in the end, regardless of how we approach the question, all of our results suggest modeling perceived behavioral control as a direct, non-mediated predictor of condom use.

      diff --git a/docs/index.html b/docs/index.html index 78f6f92d..b41a5810 100644 --- a/docs/index.html +++ b/docs/index.html @@ -464,7 +464,7 @@

      Course Information

      diff --git a/docs/search_index.json b/docs/search_index.json index dedc4190..e4e0aa50 100644 --- a/docs/search_index.json +++ b/docs/search_index.json @@ -1 +1 @@ -[["index.html", "Theory Construction and Statistical Modeling Course Information", " Theory Construction and Statistical Modeling Kyle M. Lang Last updated: 2023-10-14 Course Information In order to test a theory, we must express the theory as a statistical model and then test this model on quantitative (numeric) data. In this course we will use datasets from different disciplines within the social sciences (educational sciences, psychology, and sociology) to explain and illustrate theories and practices that are used in all social science disciplines to statistically model social science theories. This course uses existing tutorial datasets to practice the process of translating verbal theories into testable statistical models. If you are interested in the methods of acquiring high quality data to test your own theory, we recommend following the course Conducting a Survey which is taught from November to January. Most information about the course is available in this GitBook. Course-related communication will be through https://uu.blackboard.com (Log in with your student ID and password). "],["acknowledgement.html", "Acknowledgement", " Acknowledgement This course was originally developed by dr. Caspar van Lissa. I (dr. Kyle M. Lang) have modified Caspar’s original materials and take full responsibility for any errors or inaccuracies introduced through these modifications. Credit for any particularly effective piece of pedagogy should probably go to Caspar. You can view the original version of this course here on Caspar’s GitHub page. "],["instructors.html", "Instructors", " Instructors Coordinator: dr. Kyle M. Lang Lectures: dr. Kyle M. Lang Practicals: Rianne Kraakman Daniëlle Remmerswaal Danielle McCool "],["course-overview.html", "Course overview", " Course overview This course comprises three parts: Path analysis: You will learn how to estimate complex path models of observed variables (e.g., linked linear regressions) as structural equation models. Factor analysis: You will learn different ways of defining and estimating latent (unobserved) constructs. Full structural equation modeling: You will combine the first two topics to estimate path models describing the associations among latent constructs. Each of these three themes will be evaluated with a separate assignment. The first two assignments will be graded on a pass/fail basis. Your course grade will be based on your third assignment grade. "],["schedule.html", "Schedule", " Schedule Course Week Calendar Week Lecture/Practical Topic Workgroup Activity Assignment Deadline 0 36 Pre-course preparation 1 37 Introduction to R 2 38 Statistical modeling, Path analysis 3 39 Mediation, Moderation 4 40 Exploratory factor analysis (EFA) A1 Peer-Review A1: 2023-10-04 @ 23:59 5 41 Confirmatory factor analysis (CFA) 6 42 Structural equation modeling (SEM) A2 Peer-Review A2: 2023-10-18 @ 23:59 7 43 Multiple group models 8 44 Wrap-up A3 Peer-Review 9 45 Exam week: No class meetings A3: 2023-11-10 @ 23:59 NOTE: The schedule (including topics covered and assignment deadlines) is subject to change at the instructors’ discretion. "],["learning-goals.html", "Learning goals", " Learning goals In this course you will learn how to translate a social scientific theory into a statistical model, how to analyze your data with these models, and how to interpret and report your results following APA standards. After completing the course, you will be able to: Translate a verbal theory into a conceptual model, and translate a conceptual model into a statistical model. Independently analyze data using the free, open-source statistical software R. Apply a latent variable model to a real-life problem wherein the observed variables are only indirect indicators of an unobserved construct. Use a path model to represent the hypothesized causal relations among several variables, including relationships such as mediation and moderation. Explain to a fellow student how structural equation modeling combines latent variable models with path models and the benefits of doing so. Reflect critically on the decisions involved in defining and estimating structural equation models. "],["resources.html", "Resources", " Resources Literature You do not need a separate book for this course! Most of the information is contained within this GitBook and the course readings (which you will be able to access via links in this GitBook). All literature is freely available online, as long as you are logging in from within the UU-domain (i.e., from the UU campus or through an appropriate VPN). All readings are linked in this GitBook via either direct download links or DOIs. If you run into any trouble accessing a given article, searching for the title using Google Scholar or the University Library will probably due the trick. Software You will do all of your statistical analyses with the statistical programming language/environment R and the add-on package lavaan. If you want to expand your learning, you can follow this optional lavaan tutorial. "],["reading-questions.html", "Reading questions", " Reading questions Along with every article, we will provide reading questions. You will not be graded on the reading questions, but it is important to prepare the reading questions before every lecture. The reading questions serve several important purposes: Provide relevant background knowledge for the lecture Help you recognize and understand the key terms and concepts Make you aware of important publications that shaped the field Help you extract the relevant insights from the literature "],["weekly-preparation.html", "Weekly preparation", " Weekly preparation Before every class meeting (both lectures and practicals) you need to do the assigned homework (delineated in the GitBook chapter for that week). This course follows a flipped classroom procedure, so you must complete the weekly homework to meaningfully participate in, and benefit from, the class meetings. Background knowledge We assume you have basic knowledge about multivariate statistics before entering this course. You do not need any prior experience working with R. If you wish to refresh your knowledge, we recommend the chapters on ANOVA, multiple regression, and exploratory factor analysis from Field’s Discovering Statistics using R. If you cannot access the Field book, many other introductory statistics textbooks cover these topics equally well. So, use whatever you have lying around from past statistics courses. You could also try one of the following open-access options: Applied Statistics with R Introduction to Modern Statistics Introduction to Statistical Learning "],["grading.html", "Grading", " Grading Your grade for the course is based on a “portfolio” composed of the three take-home assignments: Path modeling Deadline: Wednesday 2023-10-04 at 23:59 Group assignment Pass/Fail Confirmatory factor analysis Deadline: Wednesday 2023-10-18 at 23:59 Group assignment Pass/Fail Full structural equation modeling Deadline: Friday 2023-11-10 at 23:59 Individual assignment Comprises your entire numeric course grade The specifics of the assignments will be explicated in the Assignments chapter of this GitBook "],["attendance.html", "Attendance", " Attendance Attendance is not mandatory, but we strongly encourage you to attend all lectures and practicals. In our experience, students who actively participate tend to pass the course, whereas those who do not participate tend to drop out or fail. The lectures and practicals build on each other, so, in the unfortunate event that you have to miss a class meeting, please make sure you have caught up with the material before the next session. "],["assignments.html", "Assignments", " Assignments This chapter contains the details and binding information about the three assignments that comprise the portfolio upon which your course grade is based. Below, you can find a brief idea of what each assignment will cover. For each assignment, you will use R to analyze some real-world data, and you will write up your results in a concise report (not a full research paper). Guidelines for these analyses/reports are delineated in the following three sections. You will submit your reports via Blackboard. You will complete the first two assignments in your Assignment Groups. You will complete the third assignment individually. The first two assignments are graded as pass/fail. You must pass both of these assignments to pass the course. The third assignment constitutes your course grade. "],["assignment-1-path-analysis.html", "Assignment 1: Path Analysis", " Assignment 1: Path Analysis For the first assignment, you will work in groups to apply a path model that describes how several variables could be causally related. The components of the first assignment are described below. Choose a suitable dataset, and describe the data. You can use any of the 8 datasets linked below. State the research question; define and explicate the theoretical path model. This model must include, at least, three variables. Use a path diagram to show your theoretical model. Translate your theoretical path model into lavaan syntax, and estimate the model. Include the code used to define and estimate your model as an appendix. Explain your rationale for important modeling decisions. Discuss the conceptual fit between your theory and your model. Evaluate the model assumptions. Discuss other important decisions that could have influence your results. Report the results in APA style. Provide relevant output in a suitable format. Include measures of explained variance for the dependent variables. Discuss the results. Use your results to answer the research question. Consider the strengths and limitations of your analysis. Evaluation See the Grading section below for more information on how Assignment 1 will be evaluated. You can access an evaluation matrix for Assignment 1 here. This matrix gives an indication of what level of work constitutes insufficient, sufficient, and excellent responses to the six components described above. Submission Assignment 1 is due at 23:59 on Wednesday 4 October 2023. Submit your report via the Assignment 1 portal on Blackboard. "],["assignment-2-confirmatory-factor-analysis.html", "Assignment 2: Confirmatory Factor Analysis", " Assignment 2: Confirmatory Factor Analysis In the second assignment, you will work in groups to run a CFA wherein the observed variables are indirect indicators of the unobserved constructs you want to analyze. The components of the second assignment are described below. Choose a suitable dataset, and describe the data. Ideally, you will work with the same data that you analyzed in Assignment 1. If you want to switch, you can use any of the 8 datasets linked below. State the research question; define and explicate the theoretical CFA model. This model must include, at least, two latent constructs. Use a path diagram to represent your model. Translate your theoretical model into lavaan syntax, and estimate the model. Include the code used to define and estimate your model as an appendix. Explain your rationale for important modeling decisions. Discuss the conceptual fit between your theory and your model. Evaluate the model assumptions. Discuss other important decisions that could have influence your results. Report the results in APA style. Provide relevant output in a suitable format. Include measures of model fit. Discuss the results. Use your results to answer the research question. Consider the strengths and limitations of your analysis. Evaluation See the Grading section below for more information on how Assignment 2 will be evaluated. You can access an evaluation matrix for Assignment 2 here. This matrix gives an indication of what level of work constitutes insufficient, sufficient, and excellent responses to the six components described above. Submission Assignment 2 is due at 23:59 on Wednesday 18 October 2023. Submit your report via the Assignment 2 portal on Blackboard. "],["a3_components.html", "Assignment 3: Full Structural Equation Model", " Assignment 3: Full Structural Equation Model In the third assignment, you will work individually to apply a full SEM that describes how several (latent) variables could be causally related. The components of the third assignment are described below. Choose a suitable dataset, and describe the data. Ideally, you will work with the same data that you analyzed in Assignments 1 & 2. If you want to switch, you can use any of the 8 datasets linked below. State the research question; define and explicate the theoretical SEM. The structural component of this model must include, at least, three variables. The model must include, at least, two latent variables. Use a path diagram to represent your model. Translate your theoretical SEM into lavaan syntax, and estimate the model. Include the code used to define and estimate your model as an appendix. Explain your rationale for important modeling decisions. Discuss the conceptual fit between your theory and your model. Evaluate the model assumptions. Discuss other important decisions that could have influence your results. Report the results in APA style. Provide relevant output in a suitable format. Include measures of model fit. Include measures of explained variance for the dependent variables. Discuss the results. Use your results to answer the research question. Consider the strengths and limitations of your analysis. Evaluation See the Grading section below for more information on how the component scores represented in the rubric are combined into an overall assignment grade. You can access an evaluation matrix for Assignment 3 here. This matrix gives an indication of what level of work constitutes insufficient, sufficient, and excellent responses to the six components described above. Submission Assignment 3 is due at 23:59 on Friday 10 November 2023. Submit your report via the Assignment 3 portal on Blackboard. "],["elaboration-tips.html", "Elaboration & Tips", " Elaboration & Tips Theoretical Model & Research Question You need to provide some justification for your model and research question, but only enough to demonstrate that you’ve actually conceptualized and estimated a theoretically plausible statistical model (as opposed to randomly combining variables until lavaan returns a pretty picture). You have several ways to show that your model is plausible. Use common-sense arguments. Reference (a small number of) published papers. Replicate an existing model/research question. Don’t provide a rigorous literature-supported theoretical motivation. You don’t have the time to conduct a thorough literature review, and we don’t have the time to read such reviews when grading. Literature review is not one of the learning goals for this course, so you cannot get “bonus points” for an extensive literature review. You are free to test any plausible model that meets the size requirements. You can derive your own model/research question or you can replicate a published analysis. Model Specifications We will not cover methods for modeling categorical outcome variables. So, use only continuous variables as outcomes. DVs in path models and the structural parts of SEMs Observed indicators of latent factors in CFA/SEM NOTE: You may treat ordinal items as continuous, for the purposes of these assignments. We will not cover methods for latent variable interactions. Don’t specify a theoretical model that requires an interaction involving a latent construct. There is one exception to the above prohibition. If the moderator is an observed grouping variable, you can estimate the model as a multiple-group model. We’ll cover these methods in Week 7. Assumptions You need to show that you’re thinking about the assumptions and their impact on your results, but you don’t need to run thorough model diagnostics. Indeed, the task of checking assumptions isn’t nearly as straight forward in path analysis, CFA, and SEM as it is in linear regression modeling. You won’t be able to directly apply the methods you have learned for regression diagnostics, for example. Since all of our models are estimated with normal-theory maximum likelihood, the fundamental assumption of all the models we’ll consider in this course boils down to the following. All random variables in my model are i.i.d. multivariate normally distributed. So, you can get by with basic data screening and checking the observed random variables in your model (i.e., all variables other than fixed predictors) for normality. Since checking for multivariate normality is a bit tricky, we’ll only ask you to evaluate univariate normality. You should do these evaluations via graphical means. To summarize, we’re looking for the following. Data Consider whether the measurement level of your data matches the assumptions of your model. Check your variables for univariate outliers. If you find any outliers, either treat them in some way or explain why you are retaining them for the analysis. Check for missing data. For the purposes of the assignment, you can use complete case analysis to work around the missing data. If you’re up for more of a challenge, feel free to try multiple imputation or full information maximum likelihood. Model Evaluate the univariate normality of any random, observed variables in your model. E.g., DVs in path models, observed IVs modeled as random variables, indicators of latent factors If you fit a multiple-group model for Assignment 3, do this evaluation within groups. Use graphical tools to evaluate the normality assumption. Normal QQ-Plots Histograms Results What do we mean by reporting your results “in a suitable format”? Basically, put some effort into making your results readable, and don’t include a bunch of superfluous information. Part of demonstrating that you understand the analysis is showing that you know which pieces of output convey the important information. Tabulate your results; don’t directly copy the R output. Don’t include everything lavaan gives you. Include only the output needed to understand your results and support your conclusions. "],["data_options.html", "Data", " Data Below, you can find links to a few suitable datasets that you can use for the assignments. You must use one of the following datasets. You may not choose your own data from the wild. Coping with Covid Dataset Codebook Pre-Registration Feminist Perspectives Scale Dataset Article Hypersensitive Narcissism Scale & Dirty Dozen Dataset HSNS Article DD Article Kentucky Inventory of Mindfulness Skills Dataset Article Depression Anxiety Stress Scale Dataset DASS Information Nomophobia Dataset Recylced Water Acceptance Dataset Article "],["procedures.html", "Procedures", " Procedures Formatting You must submit your assignment reports in PDF format. Each report should include a title page. The title page should include the following information: The name of the assignment. The names of all assignment authors (i.e., all group members for Assignments 1 & 2, your name for Assignment 3). The Assignment Group number (only for Assignments 1 & 2). You must include the code used to define and run your model(s) as an appendix. Try to format the text in this appendix clearly. Use a monospace font. Length You may use as many words as necessary to adequately explain yourself; though, concision and parsimony are encouraged. Note that the assignments are not intended to be full-blown papers! The focus should be on the definition of your model, how this model relates to theory (introduction), and what you have learned from your estimated model (discussion). For each of the assignments, you should be able to get the job done in fewer than 10 pages of text (excluding title page, figures, appendices, and references). Submission You will submit your reports through Blackboard. Each assignment has a corresponding item in the “Assignments” section of the BB page through which you will submit your reports. For Assignments 1 & 2, you may only submit one report per group. Designate one group member to submit the report. The grade for this submission will apply to all group members. If something goes wrong with the submission, or you notice a mistake (before the deadline) that you want to correct, you may upload a new version of your report. We will grade the final submitted version. The submissions will be screened with Ouriginal. "],["grading-1.html", "Grading", " Grading Group Assignments Assignments 1 & 2 are simply graded as pass/fail. To pass, your submission must: Do a reasonable job of addressing the relevant components listed above Be submitted before the deadline Otherwise, you will fail the assignment. Individual Assignment Assignment 3 will be fully graded on the usual 10-point scale. Points will be allocated according to the extent to which your submission addresses the six components listed above. The evaluation matrix gives an indication of how these points will be apportioned. Further details over the grading procedures for Assignment 3 (e.g., exactly how your 10-point grade will be defined) will be provided at a later date. Assuming your group passes the first two assignments, your final course grade will simply be your Assignment 3 grade. Resits You must get a “pass” for Assignments 1 & 2 and score at least 5.5 on Assignment 3 to pass the course. If you fail any of the assignments, you will have the opportunity to resit the failed assignment(s). If you resit Assignment 3, your revised graded cannot be higher than 6. Further details on the resit procedure will be provided at a later date. Example Assignment You can find an example of a good submission (for an older version of Assignment 2) here. This example is not perfect (no paper ever is), and several points could be improved. That being said, this submission exemplifies what we’re looking for in your project reports. So, following the spirit of this example would earn you a high grade. "],["rules.html", "Rules", " Rules Resources For all three assignments, you may use any reference materials you like, including: All course materials The course GitBook Additional books and papers The internet Collaboration You will complete the first two assignments in groups. Although you will work in groups, your group may not work together with other groups. You will complete the final assignment individually. For this assignment, you may not work with anyone else. For all three assignments, you are obligated to submit original work (i.e., work conducted for this course by you or your group). Submitting an assignment that violates this condition constitutes fraud. Such cases of fraud will be addressed according to the University’s standard policy. Academic integrity Hopefully, you also feel a moral obligation to obey the rules. For this course, we have implemented an examination that allows you to showcase what you have learned in a more realistic way than a written exam would allow. This assessment format spares you the stress of long exams (the two exams for this course used to be 4 hours each) and the attendant studying/cramming. The assignments will also help you assess your ability to independently analyse data, which is important to know for your future courses and/or career. However, this format also assumes that you complete the assignments in good faith. So, I simply ask that you hold up your end of the bargain, and submit your original work to show us what you’ve learned. Strict stuff By submitting your assignments (both group and individual), you confirm the following: You have completed the assignment yourself (or with your group) You are submitting work that you have written yourself (or with your group) You are using your own UU credentials to submit the assignment You have not had outside help that violates the conditions delineated above while completing the assignment All assignments will be submitted via Ouriginal in Blackboard and, thereby, checked for plagiarism. If fraud or plagiarism is detected or suspected, we will inform the Board of Examiners in the usual manner. In the event of demonstrable fraud, the sanctions delineated in Article 5.15 of the Education and Examination Regulations (EER) will apply. "],["software-setup.html", "Software Setup", " Software Setup This chapter will help you prepare for the course by showing how to install R and RStudio on your computer. If you’re already using R, there may be nothing new for you here. That being said, you should look over this chapter to ensure that your current setup will be compatible with the course requirements. If you have never used R before, this chapter is essential! The information is this chapter will be crucial for getting your computer ready for the course. "],["typographic-conventions.html", "Typographic Conventions", " Typographic Conventions Throughout this GitBook, we (try to) use a consistent set of typographic conventions: Functions are typeset in a code font, and the name of the function is always followed by parentheses E.g., sum(), mean() Other R objects (e.g., data objects, function arguments) are in also typeset in a code font but without parentheses E.g., seTE, method.tau Sometimes, we’ll use the package name followed by two colons (::, the so-called *scope-resolution operator), like lavaan::sem(). This command is valid R code and will run if you copy it into your R console. The lavaan:: part of the command tells R that we want to use the sem() from the lavaan package. "],["installing-software.html", "Installing software", " Installing software Before we start the course, we have to install three things: R: A free program for statistical programming RStudio: An integrated development environment (IDE) which makes it easier to work with R. Several packages: Separate pieces of ‘add-on’ software for R with functions to do specific analyses. Packages also include documentation describing how to use their functions and sample data. Installing R The latest version of R is available here. Click the appropriate link for your operating system and follow the instructions for installing the latest stable release. Depending on which OS you select, you may be given an option to install different components (e.g., base, contrib, Rtools). For this course, you will only need the base package. Installing RStudio Download the Free Desktop version of RStudio from the download page of the RStudio website. Installing packages To participate in this course, you will need a few essential R packages. Here’s an overview of the packages and why we need them: Package Description lavaan A sophisticated and user-friendly package for structural equation modeling dplyr A powerful suite of data-processing tools ggplot2 A flexible and user-friendly package for making graphs tidySEM Plotting and tabulating the output of SEM-models semTools Comparing models, establishing measurement invariance across groups psych Descriptive statistics and EFA rockchalk Probing interactions foreign Loading data from SPSS ‘.sav’ files readxl Loading data from Excel ‘.xslx’ files To install these packages, we use the install.packages() function in R. Open RStudio Inside RStudio, find the window named Console on left side of the screen. Copy the following code into the console and hit Enter/Return to run the command. install.packages(c("lavaan", "dplyr", "ggplot2", "tidySEM", "semTools", "psych", "rockchalk", "foreign", "readxl"), dependencies = TRUE) "],["course-data.html", "Course Data", " Course Data All of the data files you will need for the course are available in this SurfDrive directory. Follow the link to download a ZIP archive containing the data you will need to complete the practical exercises and assignments. Extract these data files to a convenient location on your computer. "],["note-on-data-updates.html", "Note on Data Updates", " Note on Data Updates During the course, we may need to update some of these datasets and/or add some new datasets to the SurfDrive directory. If so, you will need to download the updated data. We will let you know if and when any datasets are modified. In such situations, you are responsible for updating your data. Working with outdated data will probably produce incorrect results. Your answer won’t match the solutions we expect. Your answer will be marked as incorrect, even if the code used to produce the answer is correct. Points lost on an assignment due to using outdated datasets will not be returned. "],["introduction-to-r.html", "1 Introduction to R", " 1 Introduction to R This week is all about getting up-and-running with R and RStudio. Homework before the lecture Complete the preparatory material: Read over the Course Information chapter Work through the Software Setup chapter Watch the Lecture Recording for this week. Homework before the practical Complete the At-Home Exercises. Practical content During the practical you will work on the In-Class Exercises. "],["lecture.html", "1.1 Lecture", " 1.1 Lecture This week, you will learn the basics of R and RStudio. Rather than re-inventing the proverbial wheel, we’re linked to existing resources developed by R-Ladies Sydney. 1.1.1 Recordings Tour of RStudio \\[\\\\[6pt]\\] R Packages \\[\\\\[6pt]\\] Data I/0 1.1.2 Slides You can access the accompanying resources on the R-Ladies Sydney website here. "],["reading.html", "1.2 Reading", " 1.2 Reading There is no official reading this week. If you’d like to deepen your dive into R, feel free to check out Hadley Wickham’s excellent book R for Data Science. Otherwise, you may want to get a jump-start on the At-Home Exercises for this week. \\[\\\\[12pt]\\] "],["at-home-exercises.html", "1.3 At-Home Exercises", " 1.3 At-Home Exercises This week is all about gaining familiarity with R and RStudio. We’ll be using the primers available on Posit Cloud to work through some basic elements of data visualization and statistical programming in R. Although you should already have R working, this week’s at-home and in-class exercises don’t require that you have R installed on your system. If following along within this GitBook doesn’t work for you, you can also find the tutorials online on the Posit Primers page. 1.3.1 Visualizations with R 1.3.2 Programming with R End of At-Home Exercises "],["in-class-exercises.html", "1.4 In-Class Exercises", " 1.4 In-Class Exercises In the practical this week, we’ll go a little further into what it’s possible with R. Don’t worry if you cannot remember everything in these primers—they’re only meant to familiarize you with what is possible and to get you some experience interacting with R and RStudio. The following primers come from Posit Cloud and were created with the learnr package. 1.4.1 Viewing Data This first primer introduces a special data format called a tibble, as well as some functions for viewing your data. 1.4.2 Dissecting Data In the next primer, we’ll explore tools to subset and rearrange you data: select(), filter(), and arrange(). 1.4.3 Grouping and Manipulating Data Advanced If you made it through the previous two sections with ease and want to challenge yourself, go ahead with this next section. If you’re running short on time, you can skip ahead to Exploratory Data Analysis. \\[\\\\[3pt]\\] 1.4.4 Exploratory Data Analysis 1.4.5 Visualizing Data Visualizing data is a great way to start understanding a data set. In this section, we’ll highlight a few examples of how you can use the ggplot2 libarary to visualize your data. Primers on many other visualizations are available on Posit Cloud. Bar Charts for Categorical Variables Scatterplots for Continuous Variables 1.4.6 Tidying Data This primer will provide an overview of what’s meant by “tidy data”. You only need to complete the Tidy Data section—the sections on Gathering and Spreading columns are useful, but we won’t ask you to apply those techniques in this course. Recap Hopefully, you now feel more comfortable using some of R’s basic functionality and packages to work with data. Here’s a brief description of the functions covered above: install.packages() for installing packages Remember to put the package names in quotes library() for loading packages View() for viewing your dataset select() for picking only certain columns filter() for picking only certain rows arrange() for changing the rows order %>% aka “the pipe” for chaining commands together In RStudio, you can hit ctrl+shift+m as a handy key combination ? for help files Logical tests and Boolean operators == equal to != not equal to < less than <= less than or equal to > greater than >= greater than or equal to is.na() is the value NA (not available) !is.na is the value not NA & and (true only if the left and right are both true) | or (true if either the left or right are true) ! not (invert true/false) %in% in (is left in the larger set of right values) any() any (true if any in the set are true) all() all (true if all in the set are true) xor() xor (true if one and only one of the set are true) ggplot2 ggplot() create the basic object from which to building a plot aes() contains the aesthetic mappings (like x and y) geom_bar() bar plots for distributions of categorical variables geom_point() scatterplots for plotting two continuous variables geom_label_repel() for plotting text facet_wrap() for creating sets of conditional plots End of In-Class Exercises "],["statistical-modeling-path-analysis.html", "2 Statistical Modeling & Path Analysis", " 2 Statistical Modeling & Path Analysis This week, we will cover statistical modeling and path analysis. Homework before the lecture Watch the Lecture Recording for this week. Complete the Reading for this week, and answer the associated reading questions. Homework before the practical Complete the At-Home Exercises. Practical content During the practical you will work on the In-Class Exercises. "],["lecture-1.html", "2.1 Lecture", " 2.1 Lecture In this lecture, we will begin by discussing the paradigm and contextualizing statistical modeling relative to other ways that we can conduct statistical analyses. We will conclude with an introduction to . 2.1.1 Recordings Statistical Reasoning Statistical Modeling Path Analysis 2.1.2 Slides You can download the lectures slides here "],["reading-1.html", "2.2 Reading", " 2.2 Reading Reference Smaldino, P. E. (2017). Models are stupid, and we need more of them. In R.R. Vallacher, S.J. Read, & A. Nowakt (Eds.), Computational Social Psychology (pp. 311–331). New York: Routledge. SKIP PAGES 322 - 327 Questions What are the differences between a “verbal model” and a “formal model”? As explained in the paragraph “A Brief Note on Statistical Models”, formal models are not the same as statistical models. Still, we can learn a lot from Smaldino’s approach. Write down three insights from this paper that you would like to apply to your statistical modeling during this course. Suggested Reading (Optional) The following paper is not required, but it’s definitely worth a read. Breiman provides a very interesting perspective on different ways to approach a modeling-based analysis. Breiman, L. (2001). Statistical Modeling: The Two Cultures (with comments and a rejoinder by the author). Statistical Science, 16(3) 199–231. https://doi.org/10.1214/ss/1009213726 "],["at-home-exercises-1.html", "2.3 At-Home Exercises", " 2.3 At-Home Exercises Load the LifeSat.sav data. library(dplyr) library(haven) LifeSat <- read_spss("LifeSat.sav") 2.3.1 Make a table of descriptive statistics for the variables: LifSat, educ, ChildSup, SpouSup, and age. What is the average age in the sample? What is the range (youngest and oldest child)? Hint: Use the tidySEM::descriptives() function.` Click for explanation The package tidySEM contains the descriptives() function for computing descriptive statistics. The describe() function in the psych package is a good alternative. library(tidySEM) descriptives(LifeSat[ , c("LifSat", "educ", "ChildSup", "SpouSup", "age")]) 2.3.2 Run a simple linear regression with LifSat as the dependent variable and educ as the independent variable. Hints: The lm() function (short for linear model) does linear regression. The summary() function provides relevant summary statistics for the model. It can be helpful to store the results of your analysis in an object. Click for explanation results <- lm(LifSat ~ educ, data = LifeSat) summary(results) ## ## Call: ## lm(formula = LifSat ~ educ, data = LifeSat) ## ## Residuals: ## Min 1Q Median 3Q Max ## -43.781 -11.866 2.018 12.418 43.018 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) 35.184 7.874 4.469 2.15e-05 *** ## educ 3.466 1.173 2.956 0.00392 ** ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 17.64 on 96 degrees of freedom ## Multiple R-squared: 0.08344, Adjusted R-squared: 0.0739 ## F-statistic: 8.74 on 1 and 96 DF, p-value: 0.003918 2.3.3 Repeat the analysis from 2.3.2 with age as the independent variable. Click for explanation results <- lm(LifSat ~ age, data = LifeSat) summary(results) ## ## Call: ## lm(formula = LifSat ~ age, data = LifeSat) ## ## Residuals: ## Min 1Q Median 3Q Max ## -35.321 -14.184 3.192 13.593 40.626 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) 200.2302 52.1385 3.840 0.00022 *** ## age -2.0265 0.7417 -2.732 0.00749 ** ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 17.75 on 96 degrees of freedom ## Multiple R-squared: 0.07215, Adjusted R-squared: 0.06249 ## F-statistic: 7.465 on 1 and 96 DF, p-value: 0.007487 2.3.4 Repeat the analysis from 2.3.2 and 2.3.3 with ChildSup as the independent variable. Click for explanation results <- lm(LifSat ~ ChildSup, data = LifeSat) summary(results) ## ## Call: ## lm(formula = LifSat ~ ChildSup, data = LifeSat) ## ## Residuals: ## Min 1Q Median 3Q Max ## -37.32 -12.14 0.66 12.41 44.68 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) 37.559 8.342 4.502 1.89e-05 *** ## ChildSup 2.960 1.188 2.492 0.0144 * ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 17.86 on 96 degrees of freedom ## Multiple R-squared: 0.06076, Adjusted R-squared: 0.05098 ## F-statistic: 6.211 on 1 and 96 DF, p-value: 0.01441 2.3.5 Run a multiple linear regression with LifSat as the dependent variable and educ, age, and ChildSup as the independent variables. Hint: You can use the + sign to add multiple variables to the RHS of your model formula. Click for explanation results <- lm(LifSat ~ educ + age + ChildSup, data = LifeSat) summary(results) ## ## Call: ## lm(formula = LifSat ~ educ + age + ChildSup, data = LifeSat) ## ## Residuals: ## Min 1Q Median 3Q Max ## -32.98 -12.56 2.68 11.03 41.91 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) 134.9801 53.2798 2.533 0.0130 * ## educ 2.8171 1.1436 2.463 0.0156 * ## age -1.5952 0.7188 -2.219 0.0289 * ## ChildSup 2.4092 1.1361 2.121 0.0366 * ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 16.92 on 94 degrees of freedom ## Multiple R-squared: 0.1741, Adjusted R-squared: 0.1477 ## F-statistic: 6.603 on 3 and 94 DF, p-value: 0.0004254 2.3.6 Compare the results from 2.3.5 with those from 2.3.2, 2.3.3, and 2.3.4. What do you notice when you compare the estimated slopes for each of the three predictors in the multiple regression model with the corresponding estimates from the simple regression models? "],["in-class-exercises-1.html", "2.4 In-Class Exercises", " 2.4 In-Class Exercises During this practical, you will work through some exercises meant to expand your statistical reasoning skills and improve your understanding of linear models. For this exercise, having some familiarity with regression will be helpful. If you feel like you need to refresh your knowledge in this area, consider the resources listed in the Background knowledge section. Data: You will use the following dataset for these exercises. Sesam.sav 2.4.1 Data Exploration Open the file “Sesam.sav” # Load `dplyr` for data processing: library(dplyr) # Load the `haven` library for reading in SPSS files: library(haven) ## Load the 'Sesam.sav' data ## Use haven::zap_formats() to remove SPSS attributes sesam <- read_sav(file = "Sesam.sav") %>% zap_formats() This file is part of a larger dataset that evaluates the impact of the first year of the Sesame Street television series. Sesame Street is mainly concerned with teaching preschool related skills to children in the 3–5 year age range. The following variables will be used in this exercise: age: measured in months prelet: knowledge of letters before watching Sesame Street (range 0–58) prenumb: knowledge of numbers before watching Sesame Street (range 0–54) prerelat: knowledge of size/amount/position relationships before watching Sesame Street (range 0–17) peabody: vocabulary maturity before watching Sesame Street (range 20–120) postnumb: knowledge of numbers after a year of Sesame Street (range 0–54) Note: Unless stated otherwise, the following questions refer to the sesam data and the above variables. 2.4.1.1 What is the type of each variable? Hint: The output of the str() function should be helpful here. Click to show code ## Examine the data structure: str(sesam) ## tibble [240 × 8] (S3: tbl_df/tbl/data.frame) ## $ id : num [1:240] 1 2 3 4 5 6 7 8 9 10 ... ## $ age : num [1:240] 66 67 56 49 69 54 47 51 69 53 ... ## $ prelet : num [1:240] 23 26 14 11 47 26 12 48 44 38 ... ## $ prenumb : num [1:240] 40 39 9 14 51 33 13 52 42 31 ... ## $ prerelat: num [1:240] 14 16 9 9 17 14 11 15 15 10 ... ## $ peabody : num [1:240] 62 80 32 27 71 32 28 38 49 32 ... ## $ postnumb: num [1:240] 44 39 40 19 54 39 44 51 48 52 ... ## $ gain : num [1:240] 4 0 31 5 3 6 31 -1 6 21 ... ## ..- attr(*, "display_width")= int 10 Click for explanation All variables are numeric. str() uses the abbreviation “num” to indicate a numeric vector. 2.4.1.2 What is the average age in the sample? What is the age range (youngest and oldest child)? Hint: Use tidySEM::descriptives() Click to show code As in the take home exercises, you can use the descriptives() function from the tidySEM package to describe the data: library(tidySEM) descriptives(sesam) Click for explanation We can get the average age from the “mean” column in the table ( 51.5), and the age range from the columns “min” and “max”, (34 and 69 respectively.) 2.4.1.3 What is the average gain in knowledge of numbers? What is the standard deviation of this gain? Hints: You will need to compute the gain and save the change score as a new object. You can then use the base-R functions mean() and sd() to do the calculations. Click to show code Create a new variable that represents the difference between pre- and post-test scores on knowledge of numbers: sesam <- mutate(sesam, ndif = postnumb - prenumb) Compute the mean and SD of the change score: sesam %>% summarise(mean(ndif), sd(ndif)) 2.4.1.4 Create an appropriate visualization of the gain scores you computed in 2.4.1.3. Justify your choice of visualization. Hint: Some applicable visualizations are explained in the Visualizations with R section. Click to show code library(ggplot2) ## Create an empty baseline plot object: p <- ggplot(sesam, aes(x = ndif)) ## Add some appropriate geoms: p + geom_histogram() p + geom_density() p + geom_boxplot() Click for explanation Because the gain score is numeric, we should use something appropriate for showing the distribution of a continuous variable. In this case, we can use either a density plot, or a histogram (remember from the lecture, this is like a density plot, but binned). We can also use a box plot, which can be a concise way to display a lot of information about a variable in a little less space. 2.4.1.5 Create a visualization that provides information about the bivariate relationship between the pre- and post-test number knowledge. Justify your choice of visualization. Describe the relationship based on what you see in your visualization. Hint: Again, the Visualizations with R section may provide some useful insights. Click to show code ## Create a scatterplot of the pre- and post-test number knowledge ggplot(sesam, aes(x = prenumb, y = postnumb)) + geom_point() Click for explanation A scatterplot is a good tool for showing patterns in the way that two continuous variables relate to each other. From it, we can quickly gather information about whether a relationship exists, its direction, its strength, how much variation there is, and whether or not a relationship might be non-linear. Based on this scatterplot, we see a positive relationship between the prior knowledge of numbers and the knowledge of numbers at the end of the study. Children who started with a higher level of numeracy also ended with a higher level of numeracy. There is a considerable amount of variance in the relationship. Not every child increases their numeracy between pre-test and post-test. Children show differing amounts of increase. 2.4.2 Linear Modeling 2.4.2.1 Are there significant, bivariate associations between postnumb and the following variables? age prelet prenumb prerelat peabody Use Pearson correlations to answer this question. You do not need to check the assumptions here (though you would in real life). Hint: The base-R cor.test() function and the corr.test() function from the psych package will both conduct hypothesis tests for a correlation coefficients (the base-R cor() function only computes the coefficients). Click to show code library(psych) ## Test the correlations using psych::corr.test(): sesam %>% select(postnumb, age, prelet, prenumb, prerelat, peabody) %>% corr.test() ## Call:corr.test(x = .) ## Correlation matrix ## postnumb age prelet prenumb prerelat peabody ## postnumb 1.00 0.34 0.50 0.68 0.54 0.52 ## age 0.34 1.00 0.33 0.43 0.44 0.29 ## prelet 0.50 0.33 1.00 0.72 0.47 0.40 ## prenumb 0.68 0.43 0.72 1.00 0.72 0.61 ## prerelat 0.54 0.44 0.47 0.72 1.00 0.56 ## peabody 0.52 0.29 0.40 0.61 0.56 1.00 ## Sample Size ## [1] 240 ## Probability values (Entries above the diagonal are adjusted for multiple tests.) ## postnumb age prelet prenumb prerelat peabody ## postnumb 0 0 0 0 0 0 ## age 0 0 0 0 0 0 ## prelet 0 0 0 0 0 0 ## prenumb 0 0 0 0 0 0 ## prerelat 0 0 0 0 0 0 ## peabody 0 0 0 0 0 0 ## ## To see confidence intervals of the correlations, print with the short=FALSE option ## OR ## library(magrittr) ## Test the correlations using multiple cor.test() calls: sesam %$% cor.test(postnumb, age) ## ## Pearson's product-moment correlation ## ## data: postnumb and age ## t = 5.5972, df = 238, p-value = 5.979e-08 ## alternative hypothesis: true correlation is not equal to 0 ## 95 percent confidence interval: ## 0.2241066 0.4483253 ## sample estimates: ## cor ## 0.3410578 sesam %$% cor.test(postnumb, prelet) ## ## Pearson's product-moment correlation ## ## data: postnumb and prelet ## t = 8.9986, df = 238, p-value < 2.2e-16 ## alternative hypothesis: true correlation is not equal to 0 ## 95 percent confidence interval: ## 0.4029239 0.5926632 ## sample estimates: ## cor ## 0.5038464 sesam %$% cor.test(postnumb, prenumb) ## ## Pearson's product-moment correlation ## ## data: postnumb and prenumb ## t = 14.133, df = 238, p-value < 2.2e-16 ## alternative hypothesis: true correlation is not equal to 0 ## 95 percent confidence interval: ## 0.6002172 0.7389277 ## sample estimates: ## cor ## 0.6755051 sesam %$% cor.test(postnumb, prerelat) ## ## Pearson's product-moment correlation ## ## data: postnumb and prerelat ## t = 9.9857, df = 238, p-value < 2.2e-16 ## alternative hypothesis: true correlation is not equal to 0 ## 95 percent confidence interval: ## 0.4475469 0.6268773 ## sample estimates: ## cor ## 0.5433818 sesam %$% cor.test(postnumb, peabody) ## ## Pearson's product-moment correlation ## ## data: postnumb and peabody ## t = 9.395, df = 238, p-value < 2.2e-16 ## alternative hypothesis: true correlation is not equal to 0 ## 95 percent confidence interval: ## 0.4212427 0.6067923 ## sample estimates: ## cor ## 0.520128 Click for explanation Yes, based on the p-values (remember that 0 here really means very small, making it less than .05), we would say that there are significant correlations between postnumb and all other variables in the data. (In fact, all variables in the data are significantly correlated with one another.) 2.4.2.2 Do age and prenumb explain a significant proportion of the variance in postnumb? What statistic did you use to justify your conclusion? Interpret the model fit. Use the lm() function to fit your model. Click to show code lmOut <- lm(postnumb ~ age + prenumb, data = sesam) summary(lmOut) ## ## Call: ## lm(formula = postnumb ~ age + prenumb, data = sesam) ## ## Residuals: ## Min 1Q Median 3Q Max ## -38.130 -6.456 -0.456 5.435 22.568 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) 7.4242 5.1854 1.432 0.154 ## age 0.1225 0.1084 1.131 0.259 ## prenumb 0.7809 0.0637 12.259 <2e-16 *** ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 9.486 on 237 degrees of freedom ## Multiple R-squared: 0.4592, Adjusted R-squared: 0.4547 ## F-statistic: 100.6 on 2 and 237 DF, p-value: < 2.2e-16 Click for explanation Yes, age and prenumb explain a significant amount of variability in postnumb (\\(R^2 = 0.459\\), \\(F[2, 237] = 100.629\\), \\(p < 0.001\\)). We use the F statistic for the overall test of model fit to support this conclusion. The variables age and prenumb together explain 45.9% of the variability in postnumb. 2.4.2.3 Write the null and alternative hypotheses tested for in 2.4.2.2. Click for explanation Since we are testing for explained variance, our hypotheses concern the \\(R^2\\). \\[ \\begin{align*} H_0: R^2 = 0\\\\ H_1: R^2 > 0 \\end{align*} \\] Note that this is a directional hypotheses because the \\(R^2\\) cannot be negative. 2.4.2.4 Define the model syntax to estimate the model from 2.4.2.2 as a path analysis using lavaan. Click to show code mod <- 'postnumb ~ 1 + age + prenumb' 2.4.2.5 Estimate the path analytic model you defined above. Use the lavaan::sem() function to estimate the model. Click to show code library(lavaan) lavOut1 <- sem(mod, data = sesam) 2.4.2.6 Summarize the fitted model you estimated above. Use the summary() function to summarize the model. Click to show code summary(lavOut1) ## lavaan 0.6.16 ended normally after 1 iteration ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 4 ## ## Number of observations 240 ## ## Model Test User Model: ## ## Test statistic 0.000 ## Degrees of freedom 0 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## postnumb ~ ## age 0.123 0.108 1.138 0.255 ## prenumb 0.781 0.063 12.336 0.000 ## ## Intercepts: ## Estimate Std.Err z-value P(>|z|) ## .postnumb 7.424 5.153 1.441 0.150 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .postnumb 88.864 8.112 10.954 0.000 In OLS regression, the predictor variables are usually treated as fixed and do not covary. We can easily relax this assumption in path analysis. 2.4.2.7 Re-estimate the path analytic model you defined in 2.4.2.4. Specify the predictors as random, correlated variables. Hint: You can make the predictors random in, at least, two ways: Modify the model syntax to specify the correlation between age and prenumb. Add fixed.x = FALSE to your sem() call. Click to show code lavOut2 <- sem(mod, data = sesam, fixed.x = FALSE) ## OR ## mod <- ' postnumb ~ 1 + age + prenumb age ~~ prenumb ' lavOut2 <- sem(mod, data = sesam) 2.4.2.8 Summarize the fitted model you estimated above. Compare the results to those from the OLS regression in 2.4.2.2 and the path model in 2.4.2.5. Click to show code summary(lavOut2) ## lavaan 0.6.16 ended normally after 26 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 9 ## ## Number of observations 240 ## ## Model Test User Model: ## ## Test statistic 0.000 ## Degrees of freedom 0 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## postnumb ~ ## age 0.123 0.108 1.138 0.255 ## prenumb 0.781 0.063 12.336 0.000 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ## age ~~ ## prenumb 28.930 4.701 6.154 0.000 ## ## Intercepts: ## Estimate Std.Err z-value P(>|z|) ## .postnumb 7.424 5.153 1.441 0.150 ## age 51.525 0.405 127.344 0.000 ## prenumb 20.896 0.688 30.359 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .postnumb 88.864 8.112 10.954 0.000 ## age 39.291 3.587 10.954 0.000 ## prenumb 113.702 10.379 10.954 0.000 summary(lavOut1) ## lavaan 0.6.16 ended normally after 1 iteration ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 4 ## ## Number of observations 240 ## ## Model Test User Model: ## ## Test statistic 0.000 ## Degrees of freedom 0 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## postnumb ~ ## age 0.123 0.108 1.138 0.255 ## prenumb 0.781 0.063 12.336 0.000 ## ## Intercepts: ## Estimate Std.Err z-value P(>|z|) ## .postnumb 7.424 5.153 1.441 0.150 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .postnumb 88.864 8.112 10.954 0.000 summary(lmOut) ## ## Call: ## lm(formula = postnumb ~ age + prenumb, data = sesam) ## ## Residuals: ## Min 1Q Median 3Q Max ## -38.130 -6.456 -0.456 5.435 22.568 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) 7.4242 5.1854 1.432 0.154 ## age 0.1225 0.1084 1.131 0.259 ## prenumb 0.7809 0.0637 12.259 <2e-16 *** ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 9.486 on 237 degrees of freedom ## Multiple R-squared: 0.4592, Adjusted R-squared: 0.4547 ## F-statistic: 100.6 on 2 and 237 DF, p-value: < 2.2e-16 2.4.2.9 Consider the path model below. How many regression coefficients are estimated in this model? How many variances are estimated? How many covariances are estimated? Click for explanation Six regression coefficients (red) Four (residual) variances (blue) No covariances 2.4.2.10 Consider a multiple regression analysis with three continuous independent variables: scores on tests of language, history, and logic, and one continuous dependent variable: score on a math test. We want to know if scores on the language, history, and logic tests can predict the math test score. Sketch a path model that you could use to answer this question How many regression parameters are there? How many variances could you estimate? How many covariances could you estimate? 2.4.3 Categorical IVs Load the Drivers.sav data. # Read the data into a data frame named 'drivers': drivers <- read_sav("Drivers.sav") %>% as_factor() # This preserves the SPSS labels for nominal variables In this section, we will evaluate the following research question: Does talking on the phone interfere with people's driving skills? These data come from an experiment. The condition variable represents the three experimental conditions: Hand-held phone Hands-free phone Control (no phone) We will use condition as the IV in our models. The DV, RT, represents the participant’s reaction time (in milliseconds) during a driving simulation. 2.4.3.1 Use the package ggplot2 to create a density plot for the variable RT. What concept are we representing with this plot? Hint: Consider the lap times example from the statistical modeling section of Lecture 2. Click to show code ggplot(drivers, aes(x = RT)) + geom_density() Click for explanation This shows the distribution of all the combined reaction times from drivers in all three categories. 2.4.3.2 Modify this density plot by mapping the variable condition from your data to the fill aesthetic in ggplot. What is the difference between this plot and the previous plot? Do you think there is evidence for differences between the groups? How might we test this by fitting a model to our sample? Click to show code Hint: To modify the transparency of the densities, use the aesthetic alpha. ggplot(drivers, aes(x = RT, fill = condition)) + geom_density(alpha = .5) Click for explanation This figure models the conditional distribution of reaction time, where the type of cell phone usage is the grouping factor. Things you can look at to visually assess whether the three groups differ are the amount of overlap of the distributions, how much distance there is between the individual means, and whether the combined distribution is much different than the conditional distributions. If we are willing to assume that these conditional distributions are normally distributed and have equivalent variances, we could use a linear model with dummy-coded predictors. Aside: ANOVA vs. Linear Regression As you may know, the mathematical model underlying ANOVA is just a linear regression model with nominal IVs. So, in terms of the underlying statistical models, there is no difference between ANOVA and regression; the differences lie in the focus of the analysis. ANOVA is really a type of statistical test wherein we are testing hypotheses about the effects of some set of nominal grouping factors on some continuous outcome. When doing an ANOVA, we usually don’t interact directly with the parameter estimates from the underlying model. Regression is a type of statistical model (i.e., a way to represent a univariate distribution with a conditional mean and fixed variance). When we do a regression analysis, we primarily focus on the estimated parameters of the underling linear model. When doing ANOVA in R, we estimate the model exactly as we would for linear regression; we simply summarize the results differently. If you want to summarize your model in terms of the sums of squares table you usually see when running an ANOVA, you can supply your fitted lm object to the anova() function. This is a statistical modeling course, not a statistical testing course, so we will not consider ANOVA any further. 2.4.3.3 Estimate a linear model that will answer the research question stated in the beginning of this section. Use lm() to estimate the model. Summarize the fitted model and use the results to answer the research question. Click to show code library(magrittr) lmOut <- drivers %>% mutate(condition = relevel(condition, ref = "control")) %$% lm(RT ~ condition) summary(lmOut) ## ## Call: ## lm(formula = RT ~ condition) ## ## Residuals: ## Min 1Q Median 3Q Max ## -317.50 -71.25 2.98 89.55 243.45 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) 553.75 29.08 19.042 <2e-16 *** ## conditionhand-held 100.75 41.13 2.450 0.0174 * ## conditionhands-free 63.80 41.13 1.551 0.1264 ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 130.1 on 57 degrees of freedom ## Multiple R-squared: 0.09729, Adjusted R-squared: 0.06562 ## F-statistic: 3.072 on 2 and 57 DF, p-value: 0.05408 anova(lmOut) Click for explanation The effect of condition on RT is nonsignificant (\\(F[2, 57] = 3.07\\), \\(p = 0.054\\)). Therefore, based on these results, we do not have evidence for an effect of mobile phone usage on driving performance. 2.4.3.4 Use lavaan to estimate the model from 2.4.3.3 as a path model. Hint: lavaan won’t let us use factors for our categorical predictors. So, you will need to create your own dummy codes. Click to show code mod <- 'RT ~ 1 + HH + HF' lavOut <- drivers %>% mutate(HH = ifelse(condition == "hand-held", 1, 0), # Create dummy code for "hand-held" condition HF = ifelse(condition == "hands-free", 1, 0) # Create dummy code for "hands-free" condition ) %>% sem(mod, data = .) # Estimate the model summary(lavOut) ## lavaan 0.6.16 ended normally after 1 iteration ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 4 ## ## Number of observations 60 ## ## Model Test User Model: ## ## Test statistic 0.000 ## Degrees of freedom 0 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## RT ~ ## HH 100.750 40.085 2.513 0.012 ## HF 63.800 40.085 1.592 0.111 ## ## Intercepts: ## Estimate Std.Err z-value P(>|z|) ## .RT 553.750 28.344 19.537 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .RT 16068.028 2933.607 5.477 0.000 At this point, we haven’t covered the tools you need to conduct the ANOVA-style tests with path models. So, you can’t yet answer the research question with the above model. When we discuss model comparisons, you’ll get the missing tools. End of In-Class Exercises 2 "],["mediation-moderation.html", "3 Mediation & Moderation", " 3 Mediation & Moderation In this lecture, we will discuss two particular types of processes that we can model using path analysis: mediation and moderation. Homework before the lecture Watch the Lecture Recording for this week. Complete the Reading for this week, and answer the associated reading questions. Homework before the practical Complete the At-Home Exercises. Practical content During the practical you will work on the In-Class Exercises. "],["lecture-2.html", "3.1 Lecture", " 3.1 Lecture Researchers often have theories about possible causal processes linking multiple variables. Mediation is a particularly important example of such a process in which in an input variable, X, influences the outcome, Y, through an intermediary variable, M (the mediator). For instance, psychotherapy (X), may affect thoughts (M), which in turn affects mood (Y). We can investigate mediation via a specific sequence of linear regression equations, but path modeling will make our lives much easier. We can use path models to simultaneously estimate multiple related regression equations. So, mediation analysis is an ideal application of path modeling. In this lecture, we consider both approaches and discuss their relative strengths and weaknesses. As with mediation, researchers often posit theories involving moderation. Moderation implies that the effect of X on Y depends on another variable, Z. For instance, the effect of feedback (X) on performance (Y) may depend on age (Z). Older children might process feedback more effectively than younger children. Hence, the feedback is more effective for older children than for younger children, and the effect of feedback on performance is stronger for older children than for younger children. In such a case, we would say that age moderates the effect of feedback on performance. 3.1.1 Recordings Note: In the following recordings, the slide numbers are a bit of a mess, because I made these videos by cutting together recordings that used different slide decks. My apologies to those who are particularly distracted by continuity errors. Mediation Basics Mediation Testing Bootstrapping Moderation Basics Moderation Probing 3.1.2 Slides You can download the lecture slides here "],["reading-2.html", "3.2 Reading", " 3.2 Reading Reference Baron, R. M. & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical Considerations. Journal of Personality and Individual Differences, 51(6), 1173–1182 Questions What is mediation? Give an example of mediation. According to the authors, we must satisfy four criteria to infer mediation. What are these criteria? What is “moderation”, and how is it different from “mediation”? Give an example of moderation. What are the four methods given by Baron and Kenny as suitable ways to to study interaction effects? The authors suggest that one of the most common ways to address unreliability is to use multiple indicators. Thinking back to what you’ve learned about factor analysis, briefly explain why multiple indicators can improve reliability. How can you determine whether a variable is a mediator or moderator? Reference Hayes, A. F. (2009). Beyond Baron and Kenny: Statistical mediation analysis in the new millennium. Communication Monographs, 76(4), 408–420. Questions What is an indirect or mediated effect? What is the difference between the total and direct effect? What is the main problem with the Barron & Kenny “Causal Steps Approach”? What is bootstrapping, and why is it a better way to test mediation than Sobel’s test? Explain how it is possible that “effects that don’t exist can be mediated”. "],["at-home-exercises-2.html", "3.3 At-Home Exercises", " 3.3 At-Home Exercises 3.3.1 Mediation In the first part of this practical, we will analyze the data contained in SelfEsteem.sav. These data comprise 143 observations of the following variables.1 case: Participant ID number ParAtt: Parental Attachment PeerAtt: Peer Attachment Emp: Empathy ProSoc: Prosocial behavior Aggr: Aggression SelfEst: Self-esteem 3.3.1.1 Load the SelfEsteem.sav data. Note: Unless otherwise specified, all analyses in Section 3.3.1 apply to these data. Click to show code library(haven) seData <- read_sav("SelfEsteem.sav") Suppose we are interested in the (indirect) effect of peer attachment on self-esteem, and whether empathy has a mediating effect on this relationship. We might generate the following hypotheses: Better peer relationships promote higher self esteem This effect is mediated by a student’s empathy levels, where better peer relationships increase empathy, and higher levels of empathy lead to higher self-esteem. To evaluate these hypotheses, we will use lavaan to estimate a path model. 3.3.1.2 Draw a path model (on paper) that can be used to test the above hypotheses. Label the input (X), outcome (Y), and mediator/intermediary (M). Label the paths a, b, and c’. Hint: Refer back to the Mediation Basics lecture if you need help here. Click for explanation 3.3.1.3 Specify the lavaan model syntax implied by the path diagram shown above. Save the resulting character string as an object in your environment. Hint: Refer back to the example in which opinions of systematic racism mediate the relationship between political affiliation and support for affirmative action policies from the Mediation Testing lecture this week. Click to show code mod <- ' ## Equation for outcome: SelfEst ~ Emp + PeerAtt ## Equation for the mediator: Emp ~ PeerAtt ' 3.3.1.4 Use the lavaan::sem() function to estimate the model defined in 3.3.1.3. Use the default settings in sem(). Click to show code library(lavaan) out <- sem(mod, data = seData) 3.3.1.5 Explore the summary of the fitted model. Which numbers correspond to the a, b, and c’ paths? Interpret these paths. Do the direction of the effects seem to align with our hypothesis? Click to show code summary(out) ## lavaan 0.6.16 ended normally after 1 iteration ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 5 ## ## Number of observations 143 ## ## Model Test User Model: ## ## Test statistic 0.000 ## Degrees of freedom 0 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## SelfEst ~ ## Emp 0.234 0.091 2.568 0.010 ## PeerAtt 0.174 0.088 1.968 0.049 ## Emp ~ ## PeerAtt 0.349 0.076 4.628 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .SelfEst 0.934 0.110 8.456 0.000 ## .Emp 0.785 0.093 8.456 0.000 Click for explanation The results show estimates of the a path (Emp ~ PeerAtt), the b path (SelfEst ~ Emp), and the c’ path (SelfEst ~ PeerAtt). All three of these effects are positive and significant, including the direct effect of PeerAtt on SelfEst (\\(\\beta = 0.174\\), \\(Z = 1.97\\), \\(p = 0.025\\)), and the parts of the indirect effect made up by the effect of PeerAtt on Emp (\\(\\beta = 0.349\\), \\(Z = 4.63\\), \\(p = 0\\)), and Emp on SelfEst (\\(\\beta = 0.234\\), \\(Z = 2.57\\), \\(p = 0.005\\)). We can see that the direction of the effects seems to support of our hypotheses, but without taking the next steps to investigate the indirect effect, we should be hesitant to say more. Remember that an indirect effect (IE) is the product of multiple regression slopes. Therefore, to estimate an IE, we must define this product in our model syntax. In lavaan, we define the new IE parameter in two steps. Label the relevant regression paths. Use the labels to define a new parameter that represent the desired IE. We can define new parameters in lavaan model syntax via the := operator. The lavaan website contains a tutorial on this procedure: http://lavaan.ugent.be/tutorial/mediation.html 3.3.1.6 Use the procedure described above to modify the model syntax from 3.3.1.3 by adding the definition of the hypothesized IE from PeerAtt to SelfEst. Click to show code mod <- ' ## Equation for outcome: SelfEst ~ b * Emp + PeerAtt ## Equation for mediator: Emp ~ a * PeerAtt ## Indirect effect: ie := a * b ' Click for explanation Notice that I only label the parameters that I will use to define the IE. You are free to label any parameter that you like, but I choose the to label only the minimally sufficient set to avoid cluttering the code/output. 3.3.1.7 Use lavaan::sem() to estimate the model with the IEs defined. Use the default settings for sem(). Is the hypothesized IE significant according to the default tests? Hint: Refer to the Mediation Testing lecture Click to show code out <- sem(mod, data = seData) summary(out) ## lavaan 0.6.16 ended normally after 1 iteration ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 5 ## ## Number of observations 143 ## ## Model Test User Model: ## ## Test statistic 0.000 ## Degrees of freedom 0 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## SelfEst ~ ## Emp (b) 0.234 0.091 2.568 0.010 ## PeerAtt 0.174 0.088 1.968 0.049 ## Emp ~ ## PeerAtt (a) 0.349 0.076 4.628 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .SelfEst 0.934 0.110 8.456 0.000 ## .Emp 0.785 0.093 8.456 0.000 ## ## Defined Parameters: ## Estimate Std.Err z-value P(>|z|) ## ie 0.082 0.036 2.245 0.025 Click for explanation The IE of Peer Attachment on Self Esteem through Empathy is statistically significant (\\(\\hat{\\textit{IE}} = 0.082\\), \\(Z = 2.25\\), \\(p = 0.012\\)). Note: The p-value above doesn’t match the output because we’re testing a directional hypothesis, but lavaan conducts two-tailed tests for the model parameters. As we learned in the lecture, the above test of the indirect effect is equivalent to Sobel’s Z test (which we don’t really want). An appropriate, robust test of the indirect effect requires bootstrapping, which we will do later this week as part of the in-class exercises. For now, we’ll add another input variable to our model: parental attachment. We will use this model to evaluate the following research questions: Is there a direct effect of parental attachment on self-esteem, after controlling for peer attachment and empathy? Is there a direct effect of peer attachment on self-esteem, after controlling for parental attachment and empathy? Is the effect of parental attachment on self-esteem mediated by empathy, after controlling for peer attachment? Is the effect of peer attachment on self-esteem mediated by empathy, after controlling for parental attachment? 3.3.1.8 Run the path model needed to test the research questions listed above. Specify the lavaan model syntax implied by the research questions. Allow peer attachment and parental attachment to covary. Define two new parameters to represent the hypothesized indirect effects. Estimate the model using lavaan::sem(). Use the default settings in sem(). Investigate the model summary. Click to show code mod <- ' ## Equation for outcome: SelfEst ~ b * Emp + ParAtt + PeerAtt ## Equation for mediator: Emp ~ a1 * ParAtt + a2 * PeerAtt ## Covariance: ParAtt ~~ PeerAtt ie_ParAtt := a1 * b ie_PeerAtt := a2 * b ' out <- sem(mod, data = seData) summary(out) ## lavaan 0.6.16 ended normally after 16 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 10 ## ## Number of observations 143 ## ## Model Test User Model: ## ## Test statistic 0.000 ## Degrees of freedom 0 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## SelfEst ~ ## Emp (b) 0.206 0.088 2.357 0.018 ## ParAtt 0.287 0.078 3.650 0.000 ## PeerAtt 0.024 0.094 0.252 0.801 ## Emp ~ ## ParAtt (a1) 0.078 0.075 1.045 0.296 ## PeerAtt (a2) 0.306 0.086 3.557 0.000 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ## ParAtt ~~ ## PeerAtt 0.537 0.103 5.215 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .SelfEst 0.854 0.101 8.456 0.000 ## .Emp 0.779 0.092 8.456 0.000 ## ParAtt 1.277 0.151 8.456 0.000 ## PeerAtt 0.963 0.114 8.456 0.000 ## ## Defined Parameters: ## Estimate Std.Err z-value P(>|z|) ## ie_ParAtt 0.016 0.017 0.956 0.339 ## ie_PeerAtt 0.063 0.032 1.965 0.049 3.3.1.9 What can we say about the two indirect effects? Can we say that empathy mediates both paths? Click to show explanation According to the Sobel-style test, after controlling for parental attachment, the indirect effect of peer attachment on self-esteem was statistically significant (\\(\\hat{IE} = 0.063\\), \\(Z = 1.96\\), \\(p = 0.049\\)), as was the analogous direct effect (\\(\\hat{\\beta} = 0.306\\), \\(Z = 3.56\\), \\(p < 0.001\\)). After controlling for peer attachment, neither the indirect effect (\\(\\hat{IE} = 0.016\\), \\(Z = 0.96\\), \\(p = 0.339\\)) nor the direct effect (\\(\\hat{\\beta} = 0.078\\), \\(Z = 1.05\\), \\(p = 0.296\\)) of parental attachment on self-esteem was significant, though. 3.3.2 Moderation Remember that moderation attempts to describe when one variable influences another. For the home exercise, we’ll go back to the Sesame Street data we worked with for the in-class exercises last week. 3.3.2.1 Load the Sesam2.sav data.2 NOTE: Unless otherwise specified, all analyses in Section 3.3.2 use these data. Click to show code # Read the data into an object called 'sesam2': sesam2 <- read_sav("Sesam2.sav") VIEWCAT is a nominal grouping variable, but it is represented as a numeric variable in the sesam2 data. The levels represent the following frequencies of Sesame Street viewership of the children in the data: VIEWCAT = 1: Rarely/Never VIEWCAT = 2: 2–3 times a week VIEWCAT = 3: 4–5 times a week VIEWCAT = 4: > 5 times a week 3.3.2.2 Convert VIEWCAT into a factor. Make sure that VIEWCAT = 1 is the reference group. Hints: You can identify the reference group with the levels() or contrasts() functions. The reference group is the group labelled with the first level printed by levels(). When you run contrasts(), you will see a pattern matrix that defines a certain dummy coding scheme. The reference group is the group that has zeros in each column of this matrix. If you need to change the reference group, you can use the relevel() function. Click to show code library(forcats) ## Convert 'VIEWCAT' to a factor: sesam2 <- sesam2 %>% mutate(VIEWCAT = factor(VIEWCAT)) ## Optionally specify the labels # sesam2 <- # sesam2 %>% # mutate(VIEWCAT = factor(VIEWCAT, # levels = c(1, 2, 3, 4), # labels = c("Rarely/never", # "2-3 times per week", # "4-5 times per week", # "> 5 times per week"))) ## Check the reference group: levels(sesam2$VIEWCAT) ## [1] "1" "2" "3" "4" contrasts(sesam2$VIEWCAT) ## 2 3 4 ## 1 0 0 0 ## 2 1 0 0 ## 3 0 1 0 ## 4 0 0 1 ## If necessary, relevel # sesam <- # sesam2 %>% # mutate(VIEWCAT = relevel(VIEWCAT, 1)) 3.3.2.3 Use lm() to estimate a multiple regression model wherein VIEWCAT predicts POSTNUMB. Summarize the model. Interpret the estimates. Click to show code lmOut <- lm(POSTNUMB ~ VIEWCAT, data = sesam2) summary(lmOut) ## ## Call: ## lm(formula = POSTNUMB ~ VIEWCAT, data = sesam2) ## ## Residuals: ## Min 1Q Median 3Q Max ## -25.474 -7.942 0.240 8.526 25.240 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) 18.760 2.316 8.102 8.95e-14 *** ## VIEWCAT2 9.331 2.900 3.218 0.00154 ** ## VIEWCAT3 14.714 2.777 5.298 3.49e-07 *** ## VIEWCAT4 18.032 2.809 6.419 1.24e-09 *** ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 11.58 on 175 degrees of freedom ## Multiple R-squared: 0.2102, Adjusted R-squared: 0.1967 ## F-statistic: 15.53 on 3 and 175 DF, p-value: 5.337e-09 Click for explanation Viewing category explains a statistically significant proportion of the variance in the post-test score of numbers learned (\\(R^2 = 0.21\\), \\(F(3, 175) = 15.53\\), \\(p < 0.001\\)). Kids who never or rarely watched Sesame Street had an average score of 18.76 on the post-test. Kids with weekly viewing habits of 2–3, 4–5, or 5+ times per week all had significantly higher scores on the post-test than kids who never or rarely watched Sesame Street (2–3: \\(\\hat{\\beta} = 9.33\\), \\(t = 3.22\\), \\(p = 0.002\\); 4–5: \\(\\hat{\\beta} = 14.71\\), \\(t = 5.3\\), \\(p < 0.001\\); 5+: \\(\\hat{\\beta} = 18.03\\), \\(t = 6.42\\), \\(p < 0.001\\)). If we compare the box plot, kernel density plot, and model output below, the relationships between the regression coefficient estimates for the viewing categories and the group means should be evident. 3.3.2.4 Use ggplot() to make a scatterplot with AGE on the x-axis and POSTNUMB on the y-axis. Color the points according to the their VIEWCAT level. Save the plot object to a variable in your environment. Hint: You can map color to the levels of a variable on your dataset by assigning the variable names to the color argument of the aes() function in ggplot(). Click to show code library(ggplot2) ## Add aes(..., color = VIEWCAT) to get different colors for each group: p <- ggplot(sesam2, aes(x = AGE, y = POSTNUMB, color = VIEWCAT)) + geom_point() # Add points for scatterplot ## Print the plot stored as 'p': p We assigned the global color aesthetic to the VIEWCAT variable, so the points are colored based on their group. 3.3.2.5 Add linear regression lines for each group to the above scatterplot. Hints: You can add regression lines with ggplot2::geom_smooth() To get linear regression lines, set the argument method = \"lm\" To omit error envelopes, set the argument se = FALSE Click to show code ## Add OLS best-fit lines: p + geom_smooth(method = "lm", se = FALSE) The global color aesthetic assignment from above carries through to any additional plot elements that we add, including the regression lines. So, we also get a separate regression line for each VIEWCAT group. 3.3.2.6 How would you interpret the pattern of regression lines above? Click for explanation All the lines show a positive slope, so post-test number recognition appears to increase along with increasing age. The lines are not parallel, though. So VIEWCAT may be moderating the effect of AGE on POSTNUMB. Based on the figure we just created, we may want to test for moderation in our regression model. To do so, we need to add an interaction between AGE and VIEWCAT. The VIEWCAT factor is represented by 3 dummy codes in our model, though. So when we interact AGE and VIEWCAT, we will create 3 interaction terms. To test the overall moderating influence of VIEWCAT, we need to conduct a multiparameter hypothesis test of all 3 interaction terms. One way that we can go about implementing such a test is through a hierarchical regression analysis entailing three steps: Estimate the additive model wherein we regress POSTNUMB onto AGE and VIEWCAT without any interaction. Estimate the moderated model by adding the interaction between AGE and VIEWCAT into the additive model. Conduct a \\(\\Delta R^2\\) test to compare the fit of the two models. 3.3.2.7 Conduct the hierarchical regression analysis described above. Does VIEWCAT significantly moderate the effect of AGE on POSTNUMB? Provide statistical justification for your conclusion. Click to show code ## Estimate the additive model a view the results: results_add <- lm(POSTNUMB ~ VIEWCAT + AGE, data = sesam2) summary(results_add) ## ## Call: ## lm(formula = POSTNUMB ~ VIEWCAT + AGE, data = sesam2) ## ## Residuals: ## Min 1Q Median 3Q Max ## -23.680 -8.003 -0.070 8.464 22.635 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) -10.1056 6.5091 -1.553 0.12235 ## VIEWCAT2 9.1453 2.7390 3.339 0.00103 ** ## VIEWCAT3 13.8602 2.6294 5.271 3.98e-07 *** ## VIEWCAT4 16.9215 2.6636 6.353 1.79e-09 *** ## AGE 0.5750 0.1221 4.708 5.08e-06 *** ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 10.94 on 174 degrees of freedom ## Multiple R-squared: 0.2995, Adjusted R-squared: 0.2834 ## F-statistic: 18.6 on 4 and 174 DF, p-value: 9.642e-13 ## Estimate the moderated model and view the results: results_mod <- lm(POSTNUMB ~ VIEWCAT * AGE, data = sesam2) summary(results_mod) ## ## Call: ## lm(formula = POSTNUMB ~ VIEWCAT * AGE, data = sesam2) ## ## Residuals: ## Min 1Q Median 3Q Max ## -23.8371 -8.2387 0.6158 8.7988 22.5611 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) -18.7211 15.5883 -1.201 0.2314 ## VIEWCAT2 9.9741 20.6227 0.484 0.6293 ## VIEWCAT3 23.5825 19.3591 1.218 0.2248 ## VIEWCAT4 34.3969 19.3600 1.777 0.0774 . ## AGE 0.7466 0.3074 2.429 0.0162 * ## VIEWCAT2:AGE -0.0175 0.4060 -0.043 0.9657 ## VIEWCAT3:AGE -0.1930 0.3782 -0.510 0.6104 ## VIEWCAT4:AGE -0.3416 0.3770 -0.906 0.3663 ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 10.99 on 171 degrees of freedom ## Multiple R-squared: 0.3046, Adjusted R-squared: 0.2762 ## F-statistic: 10.7 on 7 and 171 DF, p-value: 3.79e-11 ## Test for moderation: anova(results_add, results_mod) Click for explanation VIEWCAT does not significantly moderate the effect of AGE on POSTNUMB (\\(F[3, 171] = 0.422\\), \\(p = 0.738\\)). 3.3.2.8 Sketch the analytic path diagrams for the additive and moderated models you estimated in 3.3.2.7 (on paper). Click for explanation Additive Model Moderated Model End of At-Home Exercises 3 These data were simulated from the covariance matrix provided in Laible, D. J., Carlo, G., & Roesch, S. C. (2004). Pathways to self-esteem in late adolescence: The role of parent and peer attachment, empathy, and social behaviours. Journal of adolescence, 27(6), 703-716.↩︎ These data are from the very interesting study: Ball, S., & Bogatz, G. A. (1970). A Summary of the Major Findings in” The First Year of Sesame Street: An Evaluation”.↩︎ "],["in-class-exercises-2.html", "3.4 In-Class Exercises", " 3.4 In-Class Exercises 3.4.1 Mediation In this practical, we’ll go back to the data from the at-home exercises, SelfEsteem.sav. Recall that these data comprise 143 observations of the following variables. case: Participant ID number ParAtt: Parental Attachment PeerAtt: Peer Attachment Emp: Empathy ProSoc: Prosocial behavior Aggr: Aggression SelfEst: Self-esteem When we last worked with the data, we built a model with one mediator (Emp), creating indirect effects between our predictors ParAtt and PeerAtt, and our outcome variable SelfEst. Below, you will estimate a more complex, multiple-mediator model. 3.4.1.1 Load the data into the object seData using haven::read_sav() Click to show code library(haven) seData <- read_sav("SelfEsteem.sav") For this analysis, we are interested in the (indirect) effects of parental and peer attachment on self-esteem. Furthermore, we want to evaluate the mediating roles of empathy and social behavior (i.e., prosocial behavior and aggression). Specifically, we have the following hypotheses. Better peer relationships will promote higher self-esteem via a three-step indirect process. Better peer relationships will increase empathy levels. Higher empathy will increase prosocial behavior and decrease aggressive behavior. More prosocial behaviors and less aggressive behavior will both produce higher self-esteem. Better relationships with parents directly increase self-esteem. To evaluate these hypotheses, we will use lavaan to estimate the following multiple mediator model as a path model. 3.4.1.2 Specify the lavaan model syntax implied by the path diagram shown above. Save the resulting character string as an object in your environment. Click to show code mod0 <- ' ## Equation for outcome: SelfEst ~ ProSoc + Aggr + Emp + ParAtt + PeerAtt ## Equations for stage 2 mediators: ProSoc ~ PeerAtt + ParAtt + Emp Aggr ~ PeerAtt + ParAtt + Emp ## Equation for stage 1 mediator: Emp ~ ParAtt + PeerAtt ## Covariances: ProSoc ~~ Aggr ParAtt ~~ PeerAtt ' 3.4.1.3 Use the lavaan::sem() function to estimate the model defined in 3.4.1.2. Use the default settings in sem(). Summarize the fitted model. Click to show code library(lavaan) out <- sem(mod0, data = seData) summary(out) ## lavaan 0.6.16 ended normally after 16 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 21 ## ## Number of observations 143 ## ## Model Test User Model: ## ## Test statistic 0.000 ## Degrees of freedom 0 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## SelfEst ~ ## ProSoc 0.252 0.096 2.634 0.008 ## Aggr 0.185 0.085 2.172 0.030 ## Emp 0.143 0.098 1.460 0.144 ## ParAtt 0.244 0.078 3.133 0.002 ## PeerAtt 0.051 0.091 0.555 0.579 ## ProSoc ~ ## PeerAtt -0.037 0.080 -0.469 0.639 ## ParAtt 0.193 0.067 2.886 0.004 ## Emp 0.477 0.074 6.411 0.000 ## Aggr ~ ## PeerAtt -0.095 0.090 -1.055 0.291 ## ParAtt -0.034 0.075 -0.454 0.650 ## Emp -0.309 0.084 -3.697 0.000 ## Emp ~ ## ParAtt 0.078 0.075 1.045 0.296 ## PeerAtt 0.306 0.086 3.557 0.000 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ## .ProSoc ~~ ## .Aggr -0.086 0.058 -1.476 0.140 ## ParAtt ~~ ## PeerAtt 0.537 0.103 5.215 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .SelfEst 0.796 0.094 8.456 0.000 ## .ProSoc 0.618 0.073 8.456 0.000 ## .Aggr 0.777 0.092 8.456 0.000 ## .Emp 0.779 0.092 8.456 0.000 ## ParAtt 1.277 0.151 8.456 0.000 ## PeerAtt 0.963 0.114 8.456 0.000 3.4.1.4 Considering the parameter estimates from 3.4.1.3, what can you say about the hypotheses? Click for explanation Notice that all of the hypotheses stated above are explicitly directional. Hence, when evaluating the significance of the structural paths that speak to these hypotheses, we should use one-tailed tests. We cannot ask lavaan to return one-tailed p-values, but we have no need to do so. We can simply divide the two-tailed p-values in half. The significant direct effect of ParAtt on SelfEst (\\(\\beta = 0.244\\), \\(Z = 3.13\\), \\(p = 0.001\\)) and the lack of a significant direct effect of PeerAtt on SelfEst (\\(\\beta = 0.051\\), \\(Z = 0.555\\), \\(p = 0.29\\)) align with our hypotheses. The remaining patterns of individual estimates also seem to conform to the hypotheses (e.g., all of the individual paths comprising the indirect effects of PeerAtt on SelfEst are significant). We cannot make any firm conclusions until we actually estimate and test the indirect effects, though. 3.4.1.5 Modify the model syntax from 3.4.1.2 by adding definitions of the two hypothesized IEs from PeerAtt to SelfEst. Click to show code You can use any labeling scheme that makes sense to you, but I recommend adopting some kind of systematic rule. Here, I will label the individual estimates in terms of the short variable names used in the path diagram above. mod <- ' ## Equation for outcome: SelfEst ~ y_m21 * ProSoc + y_m22 * Aggr + Emp + ParAtt + PeerAtt ## Equations for stage 2 mediators: ProSoc ~ m21_x2 * PeerAtt + ParAtt + m21_m1 * Emp Aggr ~ m22_x2 * PeerAtt + ParAtt + m22_m1 * Emp ## Equation for stage 1 mediator: Emp ~ ParAtt + m1_x2 * PeerAtt ## Covariances: ProSoc ~~ Aggr ParAtt ~~ PeerAtt ## Indirect effects: ie_pro := m1_x2 * m21_m1 * y_m21 ie_agg := m1_x2 * m22_m1 * y_m22 ' 3.4.1.6 Use lavaan::sem() to estimate the model with the IEs defined. Use the default settings for sem(). Are the hypothesized IEs significant according to the default tests? Click to show code out <- sem(mod, data = seData) summary(out) ## lavaan 0.6.16 ended normally after 16 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 21 ## ## Number of observations 143 ## ## Model Test User Model: ## ## Test statistic 0.000 ## Degrees of freedom 0 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## SelfEst ~ ## ProSoc (y_21) 0.252 0.096 2.634 0.008 ## Aggr (y_22) 0.185 0.085 2.172 0.030 ## Emp 0.143 0.098 1.460 0.144 ## ParAtt 0.244 0.078 3.133 0.002 ## PerAtt 0.051 0.091 0.555 0.579 ## ProSoc ~ ## PerAtt (m21_2) -0.037 0.080 -0.469 0.639 ## ParAtt 0.193 0.067 2.886 0.004 ## Emp (m21_1) 0.477 0.074 6.411 0.000 ## Aggr ~ ## PerAtt (m22_2) -0.095 0.090 -1.055 0.291 ## ParAtt -0.034 0.075 -0.454 0.650 ## Emp (m22_1) -0.309 0.084 -3.697 0.000 ## Emp ~ ## ParAtt 0.078 0.075 1.045 0.296 ## PerAtt (m1_2) 0.306 0.086 3.557 0.000 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ## .ProSoc ~~ ## .Aggr -0.086 0.058 -1.476 0.140 ## ParAtt ~~ ## PeerAtt 0.537 0.103 5.215 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .SelfEst 0.796 0.094 8.456 0.000 ## .ProSoc 0.618 0.073 8.456 0.000 ## .Aggr 0.777 0.092 8.456 0.000 ## .Emp 0.779 0.092 8.456 0.000 ## ParAtt 1.277 0.151 8.456 0.000 ## PeerAtt 0.963 0.114 8.456 0.000 ## ## Defined Parameters: ## Estimate Std.Err z-value P(>|z|) ## ie_pro 0.037 0.018 2.010 0.044 ## ie_agg -0.017 0.011 -1.657 0.098 Click for explanation The IE of Peer Attachment on Self Esteem through Empathy and Prosocial Behavior is significant (\\(\\hat{\\textit{IE}} = 0.037\\), \\(Z = 2.01\\), \\(p = 0.022\\)), as is the analogous IE through Aggressive Behavior (\\(\\hat{\\textit{IE}} = -0.017\\), \\(Z = -1.66\\), \\(p = 0.049\\)). Though, this latter effect is just barely significant at the \\(\\alpha = 0.05\\) level. The tests we used to evaluate the significance of the IEs in 3.4.1.6 are flawed because they assume normal sampling distributions for the IEs. However the IEs are defined as products of multiple, normally distributed, regression slopes. So the IEs themselves cannot be normally distributed (at least in finite samples), and the results of the normal-theory significance tests may be misleading. To get an accurate test of the IEs, we should use bootstrapping to generate an empirical sampling distribution for each IE. In lavaan, we implement bootstrapping by specifying the se = \"bootstrap\" option in the fitting function (i.e., the cfa() or sem() function) and specifying the number of bootstrap samples via the bootstrap option. Workflow Tip To draw reliable conclusions from bootstrapped results, we need many bootstrap samples (i.e., B > 1000), but we must estimate the full model for each of these samples, so the estimation can take a long time. To avoid too much frustration, you should first estimate the model without bootstrapping to make sure everything is specified correctly. Only after you are certain that your code is correct do you want to run the full bootstrapped version. 3.4.1.7 Re-estimate the model from 3.4.1.6 using 1000 bootstrap samples. Other than the se and bootstrap options, use the defaults. Are the hypothesized IEs significant according to the bootstrap-based test statistics? Click to show code ## Set a seed to get replicable bootstrap samples: set.seed(235711) ## Estimate the model with bootstrapping: out_boot <- sem(mod, data = seData, se = "bootstrap", bootstrap = 1000) ## Summarize the model: summary(out_boot) ## lavaan 0.6.16 ended normally after 16 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 21 ## ## Number of observations 143 ## ## Model Test User Model: ## ## Test statistic 0.000 ## Degrees of freedom 0 ## ## Parameter Estimates: ## ## Standard errors Bootstrap ## Number of requested bootstrap draws 1000 ## Number of successful bootstrap draws 1000 ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## SelfEst ~ ## ProSoc (y_21) 0.252 0.100 2.529 0.011 ## Aggr (y_22) 0.185 0.085 2.174 0.030 ## Emp 0.143 0.095 1.507 0.132 ## ParAtt 0.244 0.079 3.089 0.002 ## PerAtt 0.051 0.095 0.530 0.596 ## ProSoc ~ ## PerAtt (m21_2) -0.037 0.082 -0.456 0.648 ## ParAtt 0.193 0.068 2.831 0.005 ## Emp (m21_1) 0.477 0.078 6.092 0.000 ## Aggr ~ ## PerAtt (m22_2) -0.095 0.087 -1.093 0.275 ## ParAtt -0.034 0.076 -0.448 0.654 ## Emp (m22_1) -0.309 0.092 -3.356 0.001 ## Emp ~ ## ParAtt 0.078 0.072 1.092 0.275 ## PerAtt (m1_2) 0.306 0.079 3.896 0.000 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ## .ProSoc ~~ ## .Aggr -0.086 0.058 -1.493 0.135 ## ParAtt ~~ ## PeerAtt 0.537 0.128 4.195 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .SelfEst 0.796 0.082 9.698 0.000 ## .ProSoc 0.618 0.068 9.114 0.000 ## .Aggr 0.777 0.104 7.476 0.000 ## .Emp 0.779 0.090 8.651 0.000 ## ParAtt 1.277 0.197 6.473 0.000 ## PeerAtt 0.963 0.105 9.203 0.000 ## ## Defined Parameters: ## Estimate Std.Err z-value P(>|z|) ## ie_pro 0.037 0.019 1.891 0.059 ## ie_agg -0.017 0.011 -1.638 0.101 Click for explanation As with the normal-theory tests, the hypothesized IE of Peer Attachment on Self Esteem was significant (\\(\\hat{\\textit{IE}} = 0.037\\), \\(Z = 1.89\\), \\(p = 0.029\\)), but the IE of Aggressive Behavior has crossed into nonsignificant territory (\\(\\hat{\\textit{IE}} = -0.017\\), \\(Z = -1.64\\), \\(p = 0.051\\)). Note: Bootstrapping is a stochastic method, so each run can provide different results. Since the indirect effect of aggressive behavior is so close to the critical value, you may come to a different conclusions vis-á-vis statistical significance if you run this analysis with a different random number seed or a different number of bootstrap samples. When you use the summary() function to summarize the bootstrapped model from 3.4.1.7, the output will probably look pretty much the same as it did in 3.4.1.6, but it’s not. The standard errors and test statistics in the bootstrapped summary are derived from empirical sampling distributions, whereas these values are based on an assumed normal sampling distribution in 3.4.1.6. The standard method of testing IEs with bootstrapping is to compute confidence intervals (CIs) from the empirical sampling distribution of the IEs. In lavaan, we can compute basic (percentile, 95%) CIs by adding the ci = TRUE option to the summary() function. To evaluate our directional hypotheses at an \\(\\alpha = 0.05\\) level, however, we need to compute 90% CIs. We can get more control over the summary statistics (include the CIs) with the parameterEstimates() function. 3.4.1.8 Check the documentation for lavaan::parameterEstimates(). Click to show code ?parameterEstimates 3.4.1.9 Use the parameterEstimates() function to compute bootstrapped CIs for the hypothesized IEs. Compute percentile CIs. Are the IEs significant according to the bootstrapped CIs? Click to show code parameterEstimates(out_boot, ci = TRUE, level = 0.9) Click for explanation When evaluating a directional hypothesis with a CI, we only consider one of the interval’s boundaries. For a hypothesized positive effect, we check only if the lower boundary is greater than zero. For a hypothesized negative effect, we check if the upper boundary is less than zero. As with the previous tests, the IE of Peer Attachment on Self Esteem through Empathy and Prosocial Behavior is significant (\\(\\hat{\\textit{IE}} = 0.037\\), \\(95\\% ~ CI = [0.009; \\infty]\\)), but the analogous IE through Aggressive Behavior is not quite significant (\\(\\hat{\\textit{IE}} = -0.017\\), \\(95\\% ~ CI = [-\\infty; -0.003]\\)). 3.4.1.10 Based on the analyses you’ve conducted here, what do you conclude vis-à-vis the original hypotheses? Click for explanation When using normal-theory tests, both hypothesized indirect effects between Peer Attachment and Self Esteem were supported in that the IE through Empathy and Prosocial Behavior as well as the IE through Empathy and Aggressive Behavior were both significant. The hypothesized direct effect of Parent Attachment on Self Esteem was also born out via a significant direct effect in the model. When testing the indirect effects with bootstrapping, however, the effect through Aggressive Behavior was nonsignificant. Since bootstrapping gives a more accurate test of the indirect effect, we should probably trust these results more than the normal-theory results. We should not infer a significant indirect effect of Peer Attachment on Self Esteem transmitted through Empathy and Aggressive Behavior. These results may not tell the whole story, though. We have not tested for indirect effects between Parent Attachment and Self Esteem, and we have not evaluated simpler indirect effects between Peer Attachment and Self Esteem (e.g., PeerAtt \\(\\rightarrow\\) Emp \\(\\rightarrow\\) SelfEst). 3.4.2 Moderation We will first analyze a synthetic version of the Outlook on Life Survey data. The original data were collected in the United States in 2012 to measure, among other things, attitudes about racial issues, opinions of the Federal government, and beliefs about the future. We will work with a synthesized subset of the original data. You can access these synthetic data as outlook.rds. This dataset comprises 2288 observations of the following 13 variables. d1:d3: Three observed indicators of a construct measuring disillusionment with the US Federal government. Higher scores indicate more disillusionment s1:s4: Four observed indicators of a construct measuring the perceived achievability of material success. Higher scores indicate greater perceived achievability progress: A single item assessing perceived progress toward achieving the “American Dream” Higher scores indicate greater perceived progress merit: A single item assessing endorsement of the meritocratic ideal that hard work leads to success. Higher scores indicate stronger endorsement of the meritocratic ideal lib2Con: A single item assessing liberal-to-conservative orientation Lower scores are more liberal, higher scores are more conservative party: A four-level factor indicating self-reported political party affiliation disillusion: A scale score representing disillusionment with the US Federal government Created as the mean of d1:d3 success: A scale score representing the perceived achievability of material success Created as the mean of s1:s4 To satisfy the access and licensing conditions under which the original data are distributed, the data contained in outlook.rds were synthesized from the original variables using the methods described by Volker and Vink (2021). You can access the original data here, and you can access the code used to process the data here. 3.4.2.1 Read in the outlook.rds dataset. Hint: An RDS file is an R object that’s been saved to a file. To read in this type of file, we use readRDS() from base R. Click to show code outlook <- readRDS("outlook.rds") 3.4.2.2 Summarize the outlook data to get a sense of their characteristics. Click to show code head(outlook) summary(outlook) ## d1 d2 d3 s1 ## Min. :1.000 Min. :1.000 Min. :1.000 Min. :1.000 ## 1st Qu.:3.000 1st Qu.:2.000 1st Qu.:3.000 1st Qu.:2.000 ## Median :4.000 Median :3.000 Median :4.000 Median :2.000 ## Mean :3.642 Mean :3.218 Mean :3.629 Mean :2.288 ## 3rd Qu.:4.000 3rd Qu.:4.000 3rd Qu.:4.000 3rd Qu.:3.000 ## Max. :5.000 Max. :5.000 Max. :5.000 Max. :4.000 ## s2 s3 s4 progress ## Min. :1.000 Min. :1.000 Min. :1.000 Min. : 1.000 ## 1st Qu.:1.000 1st Qu.:1.000 1st Qu.:2.000 1st Qu.: 5.000 ## Median :2.000 Median :2.000 Median :2.000 Median : 7.000 ## Mean :1.922 Mean :2.012 Mean :2.469 Mean : 6.432 ## 3rd Qu.:2.000 3rd Qu.:3.000 3rd Qu.:3.000 3rd Qu.: 8.000 ## Max. :4.000 Max. :4.000 Max. :4.000 Max. :10.000 ## merit lib2Con party disillusion ## Min. :1.000 Min. :1.000 republican : 332 Min. :1.000 ## 1st Qu.:4.000 1st Qu.:3.000 democrat :1264 1st Qu.:3.000 ## Median :5.000 Median :4.000 independent: 576 Median :3.667 ## Mean :4.826 Mean :3.998 other : 116 Mean :3.497 ## 3rd Qu.:6.000 3rd Qu.:5.000 3rd Qu.:4.000 ## Max. :7.000 Max. :7.000 Max. :5.000 ## success ## Min. :1.000 ## 1st Qu.:1.750 ## Median :2.000 ## Mean :2.173 ## 3rd Qu.:2.500 ## Max. :4.000 str(outlook) ## 'data.frame': 2288 obs. of 13 variables: ## $ d1 : num 4 4 4 5 5 4 5 4 4 4 ... ## $ d2 : num 4 2 4 4 3 5 4 2 4 5 ... ## $ d3 : num 4 4 4 5 4 4 4 3 3 4 ... ## $ s1 : num 3 3 4 2 2 2 2 1 3 3 ... ## $ s2 : num 2 2 2 1 1 2 1 1 2 2 ... ## $ s3 : num 3 2 4 1 2 1 1 1 3 2 ... ## $ s4 : num 3 3 3 1 2 3 3 2 2 2 ... ## $ progress : num 8 4 6 1 6 5 7 6 9 7 ... ## $ merit : num 6 5 5 4 3 4 2 5 5 5 ... ## $ lib2Con : num 5 6 4 1 4 4 4 4 4 5 ... ## $ party : Factor w/ 4 levels "republican","democrat",..: 1 3 3 2 2 2 2 2 4 1 ... ## $ disillusion: num 4 3.33 4 4.67 4 ... ## $ success : num 2.75 2.5 3.25 1.25 1.75 2 1.75 1.25 2.5 2.25 ... We will first use OLS regression to estimate a model encoding the following relations: Belief in the achievability of success, success, predicts perceived progress toward the American Dream, progress, as the focal effect. Disillusionment with the US Federal government, disillusion moderates the success \\(\\rightarrow\\) progress effect. Placement on the liberal-to-conservative continuum, lib2Con is partialed out as a covariate. 3.4.2.3 Draw the conceptual path diagram for the model described above. Click for explanation 3.4.2.4 Write out the regression equation necessary to evaluate the moderation hypothesis described above. Click for explanation \\[ Y_{progress} = \\beta_0 + \\beta_1 W_{lib2Con} + \\beta_2 X_{success} + \\beta_3 Z_{disillusion} + \\beta_4 XZ + \\varepsilon \\] 3.4.2.5 Use lm() to estimate the moderated regression model via OLS regression. Click to show code olsFit <- lm(progress ~ lib2Con + success * disillusion, data = outlook) 3.4.2.6 Summarize the fitted model and interpret the results. Is the moderation hypothesis supported? How does disillusionment level affect the focal effect? Click to show code summary(olsFit) ## ## Call: ## lm(formula = progress ~ lib2Con + success * disillusion, data = outlook) ## ## Residuals: ## Min 1Q Median 3Q Max ## -7.4315 -1.2525 0.1307 1.4369 5.6717 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) 6.81128 0.62073 10.973 < 2e-16 *** ## lib2Con 0.03052 0.03040 1.004 0.3155 ## success 0.42360 0.25853 1.638 0.1015 ## disillusion -0.78002 0.16864 -4.625 3.95e-06 *** ## success:disillusion 0.17429 0.07273 2.396 0.0166 * ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 2.041 on 2283 degrees of freedom ## Multiple R-squared: 0.1385, Adjusted R-squared: 0.137 ## F-statistic: 91.74 on 4 and 2283 DF, p-value: < 2.2e-16 Click for explanation Yes, disillusion significantly moderates the relation between success and progress (\\(\\beta = 0.174\\), \\(t[2283] = 2.396\\), \\(p = 0.017\\)) such that the effect of success on progress increases as levels of disillusion increase, after controlling for lib2Con. The rockchalk package contains some useful routines for probing interactions estimated via lm(). Specifically, the plotslopes() function will estimate and plot simple slopes, and the testSlopes() function tests the simple slopes estimated by plotSlopes(). 3.4.2.7 Probe the interaction. Use the plotSlopes() and testSlopes() functions from the rockchalk package to conduct a simple slopes analysis for the model from 3.4.2.5. Click to show code library(rockchalk) ## Estimate and plot simple slopes: psOut <- plotSlopes(olsFit, plotx = "success", modx = "disillusion", modxVals = "std.dev") ## Test the simple slopes: tsOut <- testSlopes(psOut) ## Values of disillusion OUTSIDE this interval: ## lo hi ## -28.9332857 0.2672244 ## cause the slope of (b1 + b2*disillusion)success to be statistically significant ## View the results: tsOut$hypotests Note: The message printed by testSlopes() gives the boundaries of the Johnson-Neyman Region of Significance (Johnson & Neyman, 1936). Johnson-Neyman analysis is an alternative method of probing interactions that we have not covered in this course. For more information, check out Preacher, et al. (2006). We will now use lavaan to estimate the moderated regression model from above as a path analysis. 3.4.2.8 Define the model syntax for the path analytic version of the model described above. Parameterize the model as in the OLS regression. Use only observed items and scale scores. Click to show code pathMod <- ' progress ~ 1 + lib2Con + success + disillusion + success:disillusion ' 3.4.2.9 Estimate the path model on the outlook data. Click to show code pathFit <- sem(pathMod, data = outlook) 3.4.2.10 Summarize the fitted path model and interpret the results. Do the results match the OLS regression results? What proportion of the variability in progress is explained by this model? Hint: the function lavInspect() can be used to extract information from models Click to show code summary(pathFit) ## lavaan 0.6.16 ended normally after 1 iteration ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 6 ## ## Number of observations 2288 ## ## Model Test User Model: ## ## Test statistic 0.000 ## Degrees of freedom 0 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## progress ~ ## lib2Con 0.031 0.030 1.005 0.315 ## success 0.424 0.258 1.640 0.101 ## disillusion -0.780 0.168 -4.630 0.000 ## success:dsllsn 0.174 0.073 2.399 0.016 ## ## Intercepts: ## Estimate Std.Err z-value P(>|z|) ## .progress 6.811 0.620 10.985 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .progress 4.157 0.123 33.823 0.000 lavInspect(pathFit, "r2") ## progress ## 0.138 Click for explanation Yes, the estimates and inferential conclusions are all the same as in the OLS regression model. The model explains 13.85% of the variability in progress. The semTools package contains some helpful routines for probing interactions estimated via the lavaan() function (or one of it’s wrappers). Specifically, the probe2WayMC() and plotProbe() functions will estimate/test simple slopes and plot the estimated simple slopes, respectively. 3.4.2.11 Probe the interaction from 3.4.2.9 using semTools utilities. Use probe2WayMC() to estimate and test the simple slopes. Use plotProbe() to visualize the simple slopes. Define the simple slopes with the same conditional values of disillusion that you used in 3.4.2.7. Which simple slopes are significant? Do these results match the results from 3.4.2.7? Click to show code library(semTools) ## Define the conditional values at which to calculate simple slopes: condVals <- summarise(outlook, "m-sd" = mean(disillusion) - sd(disillusion), mean = mean(disillusion), "m+sd" = mean(disillusion) + sd(disillusion) ) %>% unlist() ## Compute simple slopes and intercepts: ssOut <- probe2WayMC(pathFit, nameX = c("success", "disillusion", "success:disillusion"), nameY = "progress", modVar = "disillusion", valProbe = condVals) ## Check the results: ssOut ## $SimpleIntcept ## disillusion est se z pvalue ## m-sd 2.719 4.690 0.231 20.271 0 ## mean 3.497 4.084 0.190 21.508 0 ## m+sd 4.274 3.477 0.230 15.122 0 ## ## $SimpleSlope ## disillusion est se z pvalue ## m-sd 2.719 0.897 0.083 10.792 0 ## mean 3.497 1.033 0.065 15.994 0 ## m+sd 4.274 1.169 0.088 13.223 0 ## Visualize the simple slopes: plotProbe(ssOut, xlim = range(outlook$success), xlab = "Ease of Personal Success", ylab = "Progress toward American Dream", legendArgs = list(legend = names(condVals)) ) Click for explanation Each of the simple slopes is significant. As level of disillusionment increases, the effect of success on progress also increases, and this effect is significant for all levels of disillusion considered here. These results match the simple slopes from the OLS regression analysis. End of In-Class Exercises 3 "],["efa.html", "4 EFA", " 4 EFA This week will be a general introduction to latent variables and scaling procedures. We will discuss several different aspects of exploratory factor analysis (EFA). Most notably: The differences between Principal Component Analyses (PCA) and Factor Analysis Model estimation and factor extraction methods Factor rotations You will have to make decisions regarding each of these aspects when conducting a factor analysis. We will also discuss reliability and factor scores as means of evaluating the properties of a scale. Homework before the lecture Watch the Lecture Recording for this week. Complete the Reading for this week, and answer the associated reading questions. Homework before the practical Complete the At-Home Exercises. Practical content During the practical you will work on the In-Class Exercises. "],["lecture-3.html", "4.1 Lecture", " 4.1 Lecture How do you know if you have measured the putative hypothetical construct that you intend to measure? The methods introduced in this lecture (namely, latent variables, factor analysis, and reliability analysis) can shed empirical light on this issue. In the social and behavioral sciences we’re often forced to measure key concepts indirectly. For example, we have no way of directly quantifying a person’s current level of depression, or their innate motivation, or their risk-aversion, or any of the other myriad psychological features that comprise the human mental state. In truth, we cannot really measure these hypothetical constructs at all, we must estimate latent representations thereof (though, psychometricians still use the language of physical measurement to describe this process). Furthermore, we can rarely estimate an adequate representation with only a single observed variable (e.g., question on a survey, score on a test, reading from a sensor). We generally need several observed variables to reliably represent a single hypothetical construct. For example, we cannot accurately determine someone’s IQ or socio-economic status based on their response to a single question; we need several questions that each tap into slightly different aspects of IQ or SES. Given multiple items measuring the same construct, we can use the methods discussed in this lecture (i.e., factor analysis and reliability analysis) to evaluate the quality of our measurement (i.e., how well we have estimated the underlying hypothetical construct). If we do well enough in this estimation task, we will be able to combine these estimated latent variables with the path analysis methods discussed in previous two weeks to produce the full structural equation models that we will cover at the end of this course. 4.1.1 Recording Notes: This week (and next), we’ll be re-using Caspar van Lissa’s old slides and lecture recording. So, you’ll see Caspar in the following video, and the slides will have a notably different flavor than our usual materials. Don’t be confused by any mention of “model fit” in the lecture. We haven’t covered model fit yet, but we will do so next week. 4.1.2 Slides You can download the lecture slides here. "],["reading-3.html", "4.2 Reading", " 4.2 Reading This week, you will read two papers. Reference 1 Preacher, K. J., & MacCullum, R. C. (2003). Repairing Tom Swift’s electric factor analysis machine, Understanding Statistics 2(1) 13–43. Questions 1 What is a latent variable? Give an example of a latent variable. What is factor analysis, and what can you investigate using this method? In the introduction, Preacher and Maccallum describe a “little jiffy” method of doing factor analysis. Briefly describe this little jiffy—or bad practice—method. Briefly explain the key differences between Principal Component Analyses (PCA) and Exploratory Factor Analyses (EFA). What is the purpose of factor rotation? Reference 2 Kestilä, E. (2006). Is there demand for radical right populism in the Finnish electorate? Scandinavian Political Studies 29(3), 169–191. Questions 2 What is the research question that the author tries to answer? Briefly describe the characteristics of the Radical Right Parties (RRP) in Europe. What are the two main explanations of support for RRP upon which this paper focuses? Does the empirical part of the paper reflect the theoretical framework well? Why or why not? According to the author, is Finland very different from other European countries on the main dependent variables? What is the author’s conclusion (i.e., how does the author answer the research question)? "],["at-home-exercises-3.html", "4.3 At-Home Exercises", " 4.3 At-Home Exercises In these exercises, you will attempt to replicate some of the analyses from the second reading for this week: Kestilä, E. (2006). Is there demand for radical right populism in the Finnish electorate? Scandinavian Political Studies 29(3), 169–191. The data for this practical were collected during the first round of the European Social Survey (ESS). The ESS is a repeated cross-sectional survey administered in 32 European countries. The first wave was collected in 2002, and two new waves have been collected each year since. You can find more info and access the data at https://www.europeansocialsurvey.org. The data we will analyze for this practical are contained in the file named ESSround1-a.sav. This file contains data for all respondents, but only includes those variables that you will need to complete the following exercises. 4.3.1 Load the ESSround1-a.sav dataset into R. Inspect the data after loading to make sure everything went well. Click to show code ## Load the 'haven' package: library(haven) library(tidySEM) ## Read the 'ESSround1-a.sav' data into a data frame called 'ess': ess <- read_spss("ESSround1-a.sav") ## Inspect the result: dim(ess) head(ess) descriptives(ess) ## [1] 42359 50 Click here for a description of the variables. Variable Description name Title of dataset essround ESS round edition Edition proddate Production date cntry Country idno Respondent’s identification number trstlgl Trust in the legal system trstplc Trust in the police trstun Trust in the United Nations trstep Trust in the European Parliament trstprl Trust in country’s parliament stfhlth State of health services in country nowadays stfedu State of education in country nowadays stfeco How satisfied with present state of economy in country stfgov How satisfied with the national government stfdem How satisfied with the way democracy works in country pltinvt Politicians interested in votes rather than peoples opinions pltcare Politicians in general care what people like respondent think trstplt Trust in politicians imsmetn Allow many/few immigrants of same race/ethnic group as majority imdfetn Allow many/few immigrants of different race/ethnic group from majority eimrcnt Allow many/few immigrants from richer countries in Europe eimpcnt Allow many/few immigrants from poorer countries in Europe imrcntr Allow many/few immigrants from richer countries outside Europe impcntr Allow many/few immigrants from poorer countries outside Europe qfimchr Qualification for immigration: christian background qfimwht Qualification for immigration: be white imwgdwn Average wages/salaries generally brought down by immigrants imhecop Immigrants harm economic prospects of the poor more than the rich imtcjob Immigrants take jobs away in country or create new jobs imbleco Taxes and services: immigrants take out more than they put in or less imbgeco Immigration bad or good for country’s economy imueclt Country’s cultural life undermined or enriched by immigrants imwbcnt Immigrants make country worse or better place to live imwbcrm Immigrants make country’s crime problems worse or better imrsprc Richer countries should be responsible for accepting people from poorer countries pplstrd Better for a country if almost everyone share customs and traditions vrtrlg Better for a country if a variety of different religions shrrfg Country has more than its fair share of people applying refugee status rfgawrk People applying refugee status allowed to work while cases considered gvrfgap Government should be generous judging applications for refugee status rfgfrpc Most refugee applicants not in real fear of persecution own countries rfggvfn Financial support to refugee applicants while cases considered rfgbfml Granted refugees should be entitled to bring close family members gndr Gender yrbrn Year of birth edulvl Highest level of education eduyrs Years of full-time education completed polintr How interested in politics lrscale Placement on left right scale One thing you might notice when inspecting the ess data is that most of the variables are stored as labelled vectors. When loading SPSS data, haven will use these labelled vectors to preserve the metadata associated with SPSS scale variables (i.e., variable labels and value labels). While it’s good to have this metadata available, we want to analyze these items as numeric variables and factors, so the value labels are only going to make our lives harder. Thankfully, the labelled package contains many routines for manipulating labelled vectors. We’ll deal with the numeric variables in just a bit, but our first task will be to covert grouping variables to factors. 4.3.2 Convert the cntry, gndr, edulvl, and polintr variables into factors. Use the as_factor() function to do the conversion. Convert edulvl and polintr to ordered factors. Click to see code library(dplyr) ess <- mutate(ess, country = as_factor(cntry), sex = as_factor(gndr), edulvl = as_factor(edulvl, ordered = TRUE), polintr = as_factor(polintr, ordered = TRUE) ) The ess dataset contains much more information than Kestilä (2006) used. Kestilä only analyzed data from the following ten countries: Austria Belgium Denmark Finland France Germany Italy Netherlands Norway Sweden So, our next task is to subset the data to only the relevant population. When we apply logical subsetting, we can select rows from a dataset based on logical conditions. In this case, we want to select only rows from the 10 countries listed above. 4.3.3 Subset the data to include only the 10 countries analyzed by Kestilä (2006). Inspect the subsetted data to check that everything went well. Hints: Use the %in% operator to create a logical vector that indicates which elements of the cntry variable are in the set of target counties. Use the droplevels() levels function to clean up empty factor levels. Click to show code ## Create a character vector naming the target countries: targets <- c("Austria", "Belgium", "Denmark", "Finland", "France", "Germany", "Italy", "Netherlands", "Norway", "Sweden") ## Select only those rows that come from a target country: ess <- filter(ess, country %in% targets) %>% # Subset rows droplevels() # Drop empty factor levels ## Inspect the result: dim(ess) ## [1] 19690 52 table(ess$country) ## ## Austria Belgium Germany Denmark Finland France ## 2257 1899 2919 1506 2000 1503 ## Italy Netherlands Norway Sweden ## 1207 2364 2036 1999 In keeping with common practice, we will treat ordinal Likert-type rating scales with five or more levels as continuous. Since some R routines will treat labelled vectors as discrete variables, we can make things easier for ourselves by converting all the labelled vectors in our data to numeric vectors. We can use the labelled::remove_val_labels() function to strip the value labels and convert all of the labelled vectors to numeric vectors. 4.3.4 Convert the remaining labelled vectors to numeric vectors. Click to see code ## If necessary, install the labelled package: # install.packages("labelled", repos = "https://cloud.r-project.org") ## Load the labelled package: library(labelled) ## Strip the value labels: ess <- remove_val_labels(ess) ## Check the effects: str(ess) ## tibble [19,690 × 52] (S3: tbl_df/tbl/data.frame) ## $ name : chr [1:19690] "ESS1e06_1" "ESS1e06_1" "ESS1e06_1" "ESS1e06_1" ... ## ..- attr(*, "label")= chr "Title of dataset" ## ..- attr(*, "format.spss")= chr "A9" ## ..- attr(*, "display_width")= int 14 ## $ essround: num [1:19690] 1 1 1 1 1 1 1 1 1 1 ... ## ..- attr(*, "label")= chr "ESS round" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 10 ## $ edition : chr [1:19690] "6.1" "6.1" "6.1" "6.1" ... ## ..- attr(*, "label")= chr "Edition" ## ..- attr(*, "format.spss")= chr "A3" ## ..- attr(*, "display_width")= int 9 ## $ proddate: chr [1:19690] "03.10.2008" "03.10.2008" "03.10.2008" "03.10.2008" ... ## ..- attr(*, "label")= chr "Production date" ## ..- attr(*, "format.spss")= chr "A10" ## ..- attr(*, "display_width")= int 12 ## $ cntry : num [1:19690] 1 18 1 1 18 1 2 18 1 18 ... ## ..- attr(*, "label")= chr "Country" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 7 ## $ idno : num [1:19690] 1 1 2 3 3 4 4 4 6 6 ... ## ..- attr(*, "label")= chr "Respondent's identification number" ## ..- attr(*, "format.spss")= chr "F9.0" ## ..- attr(*, "display_width")= int 11 ## $ trstlgl : num [1:19690] 10 6 8 4 8 10 9 7 7 7 ... ## ..- attr(*, "label")= chr "Trust in the legal system" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ trstplc : num [1:19690] 10 8 5 8 8 9 8 9 4 9 ... ## ..- attr(*, "label")= chr "Trust in the police" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ trstun : num [1:19690] 9 8 6 NA 5 8 NA 7 5 7 ... ## ..- attr(*, "label")= chr "Trust in the United Nations" ## ..- attr(*, "format.spss")= chr "F2.0" ## $ trstep : num [1:19690] NA 3 0 7 3 7 0 3 4 6 ... ## ..- attr(*, "label")= chr "Trust in the European Parliament" ## ..- attr(*, "format.spss")= chr "F2.0" ## $ trstprl : num [1:19690] 9 7 0 6 8 8 10 2 6 8 ... ## ..- attr(*, "label")= chr "Trust in country's parliament" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ stfhlth : num [1:19690] 10 4 0 7 6 8 NA 6 3 5 ... ## ..- attr(*, "label")= chr "State of health services in country nowadays" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ stfedu : num [1:19690] 8 7 7 5 8 7 NA 7 6 7 ... ## ..- attr(*, "label")= chr "State of education in country nowadays" ## ..- attr(*, "format.spss")= chr "F2.0" ## $ stfeco : num [1:19690] 7 6 0 7 8 6 NA 9 8 9 ... ## ..- attr(*, "label")= chr "How satisfied with present state of economy in country" ## ..- attr(*, "format.spss")= chr "F2.0" ## $ stfgov : num [1:19690] 7 7 0 7 6 3 NA 5 5 7 ... ## ..- attr(*, "label")= chr "How satisfied with the national government" ## ..- attr(*, "format.spss")= chr "F2.0" ## $ stfdem : num [1:19690] 8 5 5 5 7 7 NA 7 7 9 ... ## ..- attr(*, "label")= chr "How satisfied with the way democracy works in country" ## ..- attr(*, "format.spss")= chr "F2.0" ## $ pltinvt : num [1:19690] 1 3 1 1 4 1 1 3 2 3 ... ## ..- attr(*, "label")= chr "Politicians interested in votes rather than peoples opinions" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ pltcare : num [1:19690] 1 4 1 1 4 3 2 5 2 3 ... ## ..- attr(*, "label")= chr "Politicians in general care what people like respondent think" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ trstplt : num [1:19690] 0 5 0 2 5 4 8 2 4 6 ... ## ..- attr(*, "label")= chr "Trust in politicians" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ imsmetn : num [1:19690] 4 3 2 3 2 1 NA 2 NA 1 ... ## ..- attr(*, "label")= chr "Allow many/few immigrants of same race/ethnic group as majority" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ imdfetn : num [1:19690] 3 3 2 3 2 2 NA 2 NA 1 ... ## ..- attr(*, "label")= chr "Allow many/few immigrants of different race/ethnic group from majority" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ eimrcnt : num [1:19690] 4 2 2 2 3 1 NA 2 NA 1 ... ## ..- attr(*, "label")= chr "Allow many/few immigrants from richer countries in Europe" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ eimpcnt : num [1:19690] 3 2 2 2 2 2 NA 2 NA 1 ... ## ..- attr(*, "label")= chr "Allow many/few immigrants from poorer countries in Europe" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ imrcntr : num [1:19690] 3 3 2 2 2 1 NA 2 NA 2 ... ## ..- attr(*, "label")= chr "Allow many/few immigrants from richer countries outside Europe" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ impcntr : num [1:19690] 3 2 2 3 2 1 NA 2 NA 2 ... ## ..- attr(*, "label")= chr "Allow many/few immigrants from poorer countries outside Europe" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ qfimchr : num [1:19690] 4 2 0 6 2 0 99 0 1 2 ... ## ..- attr(*, "label")= chr "Qualification for immigration: christian background" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ qfimwht : num [1:19690] 1 0 0 0 0 0 99 0 0 1 ... ## ..- attr(*, "label")= chr "Qualification for immigration: be white" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ imwgdwn : num [1:19690] 3 4 2 2 3 3 NA 4 NA 4 ... ## ..- attr(*, "label")= chr "Average wages/salaries generally brought down by immigrants" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ imhecop : num [1:19690] 2 2 1 4 3 2 NA 3 NA 2 ... ## ..- attr(*, "label")= chr "Immigrants harm economic prospects of the poor more than the rich" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ imtcjob : num [1:19690] 7 5 6 5 7 10 NA 8 NA 4 ... ## ..- attr(*, "label")= chr "Immigrants take jobs away in country or create new jobs" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ imbleco : num [1:19690] 9 4 2 NA 3 10 NA 9 NA 6 ... ## ..- attr(*, "label")= chr "Taxes and services: immigrants take out more than they put in or less" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ imbgeco : num [1:19690] 4 3 10 7 5 10 NA 8 NA 5 ... ## ..- attr(*, "label")= chr "Immigration bad or good for country's economy" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ imueclt : num [1:19690] 9 4 10 5 4 10 NA 9 NA 3 ... ## ..- attr(*, "label")= chr "Country's cultural life undermined or enriched by immigrants" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ imwbcnt : num [1:19690] 7 3 5 5 5 10 NA 8 NA 5 ... ## ..- attr(*, "label")= chr "Immigrants make country worse or better place to live" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ imwbcrm : num [1:19690] 3 3 5 2 3 5 NA 5 NA 3 ... ## ..- attr(*, "label")= chr "Immigrants make country's crime problems worse or better" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ imrsprc : num [1:19690] 2 2 1 4 1 2 NA 1 1 3 ... ## ..- attr(*, "label")= chr "Richer countries should be responsible for accepting people from poorer countries" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ pplstrd : num [1:19690] 2 4 2 2 3 4 NA 4 4 2 ... ## ..- attr(*, "label")= chr "Better for a country if almost everyone share customs and traditions" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ vrtrlg : num [1:19690] 3 5 3 2 4 1 NA 4 2 3 ... ## ..- attr(*, "label")= chr "Better for a country if a variety of different religions" ## ..- attr(*, "format.spss")= chr "F1.0" ## $ shrrfg : num [1:19690] 3 2 1 1 3 3 NA 3 4 3 ... ## ..- attr(*, "label")= chr "Country has more than its fair share of people applying refugee status" ## ..- attr(*, "format.spss")= chr "F1.0" ## $ rfgawrk : num [1:19690] 2 2 1 2 2 2 NA 2 1 2 ... ## ..- attr(*, "label")= chr "People applying refugee status allowed to work while cases considered" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ gvrfgap : num [1:19690] 4 3 2 4 2 2 NA 3 2 4 ... ## ..- attr(*, "label")= chr "Government should be generous judging applications for refugee status" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ rfgfrpc : num [1:19690] 4 3 2 4 4 4 NA 4 3 4 ... ## ..- attr(*, "label")= chr "Most refugee applicants not in real fear of persecution own countries" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ rfggvfn : num [1:19690] 2 3 2 4 3 2 NA 2 2 2 ... ## ..- attr(*, "label")= chr "Financial support to refugee applicants while cases considered" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ rfgbfml : num [1:19690] 2 3 1 2 2 1 NA 4 2 3 ... ## ..- attr(*, "label")= chr "Granted refugees should be entitled to bring close family members" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ gndr : num [1:19690] 1 2 1 2 2 1 NA 2 2 1 ... ## ..- attr(*, "label")= chr "Gender" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 6 ## $ yrbrn : num [1:19690] 1949 1978 1953 1940 1964 ... ## ..- attr(*, "label")= chr "Year of birth" ## ..- attr(*, "format.spss")= chr "F4.0" ## ..- attr(*, "display_width")= int 7 ## $ edulvl : Ord.factor w/ 7 levels "Not completed primary education"<..: NA 4 NA NA 4 NA NA 7 NA 6 ... ## $ eduyrs : num [1:19690] 11 16 14 9 12 18 NA 17 15 17 ... ## ..- attr(*, "label")= chr "Years of full-time education completed" ## ..- attr(*, "format.spss")= chr "F2.0" ## $ polintr : Ord.factor w/ 4 levels "Very interested"<..: 3 3 1 2 3 2 1 4 3 3 ... ## $ lrscale : num [1:19690] 6 7 6 5 8 5 NA 8 5 7 ... ## ..- attr(*, "label")= chr "Placement on left right scale" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ country : Factor w/ 10 levels "Austria","Belgium",..: 1 9 1 1 9 1 2 9 1 9 ... ## $ sex : Factor w/ 2 levels "Male","Female": 1 2 1 2 2 1 NA 2 2 1 ... descriptives(ess) Click for explanation Note that the numeric variables are now simple numeric vectors, but the variable labels have been retained as column attributes (which is probably useful). If we want to completely nuke the labelling information, we can use the labelled::remove_labels() function to do so. In addition to screening with summary statistics, we can also visualize the variables’ distributions. You have already created a few such visualizations for single variables. Now, we will use a few tricks to efficiently plot each of our target variables. The first step in this process will be to convert the interesting part of our data from “wide format” (one column per variable) into “long format” (one column of variable names, one column of data values). The pivot_longer() function from the tidyr package provides a convenient way to execute this conversion. 4.3.5 Use tidyr::pivot_longer() to create a long-formatted data frame from the target variables in ess. The target variables are all columns from trstlgl to rfgbfml. Click to show code ## Load the tidyr package: library(tidyr) ## Convert the target variables into a long-formatted data frame: ess_plot <- pivot_longer(ess, cols = trstlgl:rfgbfml, # Which columns to convert names_to = "variable", # Name for the new grouping variable values_to = "value") # Name for the column of stacked values The next step in the process will be to plot the variables using ggplot(). In the above code, I’ve named the new grouping variable variable and the new stacked data variable value. So, to create one plot for each (original, wide-format) variable, we will use the facet_wrap() function to facet the plots of value on the variable column (i.e., create a separate conditional plot of value for each unique value in variable). 4.3.6 Use ggplot() with an appropriate geom (e.g., geom_histogram(), geom_density(), geom_boxplot()) and facet_wrap() to visualize each of the target variables. Hint: To implement the faceting, simply add facet_wrap(~ variable, scales = \"free_x\") to the end of your ggplot() call (obviously, replacing “variable” with whatever you named the grouping variable in your pivot_longer() call). Click to show code library(ggplot2) ggplot(ess_plot, aes(x = value)) + geom_histogram() + # Create a histogram facet_wrap(~ variable, scales = "free_x") # Facet on 'variable' Click for explanation Notice that the variables are actually discrete (i.e., each variable takes only a few integer values). However, most variables look relatively normal despite being categorical. So, we’ll bend the rules a bit and analyze these variables as continuous. It also looks like there’s something weird going on with qfimchr and qfimwht. More on that below. 4.3.7 Check the descriptives for the target variables again. Do you see any remaining issues? Click to show code select(ess, trstlgl:rfgbfml) %>% descriptives() Click for explanation The variables qfimchr and qfimwht both contain values that fall outside the expected range for our survey responses: 77, 88, and 99. In SPSS, these were labeled as “Refusal” “Don’t know” and “No answer” respectively, and would not have contributed to the analysis. 4.3.8 Correct any remaining issues you found above. Click to show code ess <- ess %>% mutate(across(c(qfimchr, qfimwht), na_if, 77)) %>% mutate(across(c(qfimchr, qfimwht), na_if, 88)) %>% mutate(across(c(qfimchr, qfimwht), na_if, 99)) ## Check the results: select(ess, trstlgl:rfgbfml) %>% descriptives() Click to show explanation Here, we need to tell R that these values should be considered missing, or NA. Otherwise they will contribute the numeric value to the analysis, as though someone had provided an answer of 77 on a 10-point scale. We’ve done quite a bit of data processing, and we’ll continue to use these data for several future practicals, so it would be a good idea to save the processed dataset for later use. When saving data that you plan to analyze in R, you will usually want to use the R Data Set (RDS) format. Datasets saved in RDS format retain all of their attributes and formatting (e.g., factor are still factors, missing values are coded as NA, etc.). So, you don’t have to redo any data processing before future analyses. 4.3.9 Use the saveRDS() function to save the processed dataset. Click to show code ## Save the processed data: saveRDS(ess, "ess_round1.rds") Now, we’re ready to run the analyses and see if we can replicate the Kestilä (2006) results. 4.3.10 Run two principal component analyses (PCA): one for trust in politics, one for attitudes towards immigration. Use the principal() function from the psych package. Use exactly the same specifications as Kestilä (2006) concerning the estimation method, rotation, number of components extracted, etc. Hints: Remember that you can view the help file for psych::principal() by running ?psych::principal or, if the psych package already loaded, simply running ?principal. When you print the output from psych::principal(), you can use the cut option to hide any factor loadings smaller than a given threshold. You could consider hiding any loadings smaller than those reported by Kestilä (2006) to make the output easier to interpret. Click to show code Trust in politics Kestilä extracted three components with VARIMAX rotation. ## Load the psych package: library(psych) ## Run the PCA: pca_trust <- select(ess, trstlgl:trstplt) %>% principal(nfactors = 3, rotate = "varimax") ## Print the results: print(pca_trust, cut = 0.3, digits = 3) ## Principal Components Analysis ## Call: principal(r = ., nfactors = 3, rotate = "varimax") ## Standardized loadings (pattern matrix) based upon correlation matrix ## RC3 RC2 RC1 h2 u2 com ## trstlgl 0.779 0.669 0.331 1.21 ## trstplc 0.761 0.633 0.367 1.18 ## trstun 0.675 0.556 0.444 1.44 ## trstep 0.651 0.332 0.549 0.451 1.57 ## trstprl 0.569 0.489 0.650 0.350 2.49 ## stfhlth 0.745 0.567 0.433 1.04 ## stfedu 0.750 0.603 0.397 1.14 ## stfeco 0.711 0.300 0.616 0.384 1.44 ## stfgov 0.634 0.377 0.587 0.413 1.88 ## stfdem 0.369 0.568 0.325 0.564 0.436 2.38 ## pltinvt 0.817 0.695 0.305 1.08 ## pltcare 0.811 0.695 0.305 1.11 ## trstplt 0.510 0.611 0.716 0.284 2.40 ## ## RC3 RC2 RC1 ## SS loadings 2.942 2.668 2.490 ## Proportion Var 0.226 0.205 0.192 ## Cumulative Var 0.226 0.432 0.623 ## Proportion Explained 0.363 0.329 0.307 ## Cumulative Proportion 0.363 0.693 1.000 ## ## Mean item complexity = 1.6 ## Test of the hypothesis that 3 components are sufficient. ## ## The root mean square of the residuals (RMSR) is 0.07 ## with the empirical chi square 15240.94 with prob < 0 ## ## Fit based upon off diagonal values = 0.967 Attitudes toward immigration Kestilä extracted five components with VARIMAX rotation. pca_att <- select(ess, imsmetn:rfgbfml) %>% principal(nfactors = 5, rotate = "varimax") print(pca_att, cut = 0.3, digits = 3) ## Principal Components Analysis ## Call: principal(r = ., nfactors = 5, rotate = "varimax") ## Standardized loadings (pattern matrix) based upon correlation matrix ## RC2 RC1 RC5 RC3 RC4 h2 u2 com ## imsmetn 0.797 0.725 0.275 1.30 ## imdfetn 0.775 0.794 0.206 1.70 ## eimrcnt 0.827 0.715 0.285 1.09 ## eimpcnt 0.800 0.789 0.211 1.49 ## imrcntr 0.835 0.747 0.253 1.15 ## impcntr 0.777 0.782 0.218 1.63 ## qfimchr 0.813 0.688 0.312 1.08 ## qfimwht 0.752 0.637 0.363 1.26 ## imwgdwn 0.807 0.712 0.288 1.19 ## imhecop 0.747 0.669 0.331 1.42 ## imtcjob 0.569 0.334 0.484 0.516 1.99 ## imbleco 0.703 0.554 0.446 1.25 ## imbgeco 0.698 0.605 0.395 1.52 ## imueclt 0.568 -0.340 0.545 0.455 2.43 ## imwbcnt 0.673 0.633 0.367 1.87 ## imwbcrm 0.655 0.478 0.522 1.23 ## imrsprc 0.614 0.440 0.560 1.34 ## pplstrd 0.324 -0.551 0.468 0.532 2.11 ## vrtrlg -0.345 0.471 0.419 0.581 2.67 ## shrrfg 0.365 -0.352 0.418 0.582 4.16 ## rfgawrk 0.614 0.396 0.604 1.10 ## gvrfgap 0.691 0.559 0.441 1.35 ## rfgfrpc -0.387 0.327 0.673 3.34 ## rfggvfn 0.585 0.417 0.583 1.46 ## rfgbfml 0.596 0.460 0.540 1.61 ## ## RC2 RC1 RC5 RC3 RC4 ## SS loadings 4.374 3.393 2.774 2.199 1.723 ## Proportion Var 0.175 0.136 0.111 0.088 0.069 ## Cumulative Var 0.175 0.311 0.422 0.510 0.579 ## Proportion Explained 0.302 0.235 0.192 0.152 0.119 ## Cumulative Proportion 0.302 0.537 0.729 0.881 1.000 ## ## Mean item complexity = 1.7 ## Test of the hypothesis that 5 components are sufficient. ## ## The root mean square of the residuals (RMSR) is 0.05 ## with the empirical chi square 29496.06 with prob < 0 ## ## Fit based upon off diagonal values = 0.976 Feature engineering (i.e., creating new variables by combining and/or transforming existing variables) is one of the most common applications of PCA. PCA is a dimension reduction technique that distills the most salient information from a set of variables into a (smaller) set of component scores. Hence, PCA can be a good way of creating aggregate items (analogous to weighted scale scores) when the data are not collected with validated scales. Principal component scores are automatically generated when we run the PCA. If we want to use these scores in subsequent analyses (e.g., as predictors in a regression model), we usually add them to our dataset as additional columns. 4.3.11 Add the component scores produced by the analyses you ran above to the ess data frame. Give each component score an informative name, based on your interpretation of the factor loading matrix I.e., What hypothetical construct do you think each component represents given the items that load onto it? Hints: You can use the data.frame() function to join multiple objects into a single data frame. You can use the colnames() function to assign column names to a matrix or data frame. 1. Extract the component scores Click to show code ## Save the component scores in stand-alone matrices: trust_scores <- pca_trust$scores att_scores <- pca_att$scores ## Inspect the result: head(trust_scores) ## RC3 RC2 RC1 ## [1,] NA NA NA ## [2,] 0.09755193 -0.01552183 0.994954 ## [3,] 0.23069626 -1.53162604 -2.022642 ## [4,] NA NA NA ## [5,] -0.21112678 0.84370377 1.200007 ## [6,] 1.86596955 0.31083233 -1.062603 summary(trust_scores) ## RC3 RC2 RC1 ## Min. :-4.035 Min. :-3.706 Min. :-3.139 ## 1st Qu.:-0.527 1st Qu.:-0.652 1st Qu.:-0.649 ## Median : 0.155 Median : 0.094 Median : 0.092 ## Mean : 0.055 Mean : 0.015 Mean : 0.049 ## 3rd Qu.: 0.727 3rd Qu.: 0.742 3rd Qu.: 0.742 ## Max. : 3.302 Max. : 3.452 Max. : 3.539 ## NA's :4912 NA's :4912 NA's :4912 head(att_scores) ## RC2 RC1 RC5 RC3 RC4 ## [1,] 1.9873715 1.3233586 -0.8382499 -0.02172765 -0.0908143 ## [2,] 0.1692841 -1.2178436 -0.5016936 -0.21749066 0.6758844 ## [3,] -0.3630480 0.3260383 -1.5133423 -0.51405480 -2.2071787 ## [4,] NA NA NA NA NA ## [5,] -0.1137484 -0.7891232 -1.4732563 -0.05843873 0.4110692 ## [6,] -0.9195530 2.8231404 -0.3480398 -0.75699796 -1.3230602 summary(att_scores) ## RC2 RC1 RC5 RC3 ## Min. :-3.660 Min. :-3.929 Min. :-3.824 Min. :-2.764 ## 1st Qu.:-0.616 1st Qu.:-0.585 1st Qu.:-0.656 1st Qu.:-0.748 ## Median :-0.085 Median : 0.062 Median :-0.008 Median :-0.121 ## Mean :-0.013 Mean : 0.012 Mean : 0.021 Mean : 0.014 ## 3rd Qu.: 0.680 3rd Qu.: 0.654 3rd Qu.: 0.652 3rd Qu.: 0.698 ## Max. : 3.743 Max. : 4.584 Max. : 4.108 Max. : 4.084 ## NA's :5447 NA's :5447 NA's :5447 NA's :5447 ## RC4 ## Min. :-3.784 ## 1st Qu.:-0.683 ## Median : 0.046 ## Mean : 0.003 ## 3rd Qu.: 0.717 ## Max. : 3.254 ## NA's :5447 Click for explanation The object produced by psych::principal() is simply list, and the component scores are already stored therein. So, to extract the component scores, we simply use the $ operator to extract them. 2. Name the component scores Click to show code ## Check names (note the order): colnames(trust_scores) ## [1] "RC3" "RC2" "RC1" colnames(att_scores) ## [1] "RC2" "RC1" "RC5" "RC3" "RC4" ## Give informative names: colnames(trust_scores) <- c("Trust_Institutions", "Satisfaction", "Trust_Politicians") colnames(att_scores) <- c("Quantity", "Effects", "Refugees", "Diversity", "Economic") 3. Add the component scores to the dataset Click to show code # Add the component scores to the 'ess' data: ess <- data.frame(ess, trust_scores, att_scores) 4.3.12 Were you able to replicate the results of Kestilä (2006)? Click for explanation Yes, more-or-less. Although the exact estimates differ somewhat, the general pattern of factor loadings in Kestilä (2006) matches what we found here. End of At-Home Exercises "],["in-class-exercises-3.html", "4.4 In-Class Exercises", " 4.4 In-Class Exercises In these exercises, we will continue with our re-analysis/replication of the Kestilä (2006) results. Rather than attempting a direct replication, we will now redo the analysis using exploratory factor analysis (EFA). 4.4.1 Load the ess_round1.rds dataset. These are the data that we saved after the data processing in the At-Home Exercises. Click to show code ess <- readRDS("ess_round1.rds") 4.4.2 Kestilä (2006) claimed that running a PCA is a good way to test if the questions in the ESS measure attitudes towards immigration and trust in politics. Based on what you’ve learned from the readings and lectures, do you agree with this position? Click for explanation Hopefully not. PCA is not a method for estimating latent measurement structure; PCA is a dimension reduction technique that tries to summarize a set of data with a smaller set of component scores. If we really want to estimate the factor structure underlying a set of observed variables, we should use EFA. 4.4.3 Suppose you had to construct the trust in politics and attitude towards immigration scales described by Kestilä (2006) based on the theory and background information presented in that article. What type of analysis would you choose? What key factors would influence your decision? Click for explanation We are trying to estimate meaningful latent factors, so EFA would be an appropriate method. The theory presented by Kestilä (2006) did not hypothesize a particular number of factors, so we would need to use appropriate techniques to estimate the best number. In particular, combining information from: Scree plots Parallel analysis Substantive interpretability of the (rotated) factor loadings Since the factors are almost certainly correlated, we should apply an oblique rotation. We will now rerun the two PCAs that you conducted for the At-Home Exercises using EFA. We will estimate the EFA models using the psych::fa() function, but we need to know how many factors to extract. We could simply estimate a range of solutions and compare the results. We can restrict the range of plausible solutions and save some time by first checking/plotting the eigenvalues and running parallel analysis. 4.4.4 Estimate the number of latent factors underlying the Trust items based on the eigenvalues, the scree plot, and parallel analysis. How many factors are suggested by each method? 1. Eigenvalue estimation Click to show code ## Load the psych package: library(psych) ## Run a trivial EFA on the 'trust' items efa_trust0 <- select(ess, trstlgl:trstplt) %>% fa(nfactors = 1, rotate = "none") Click for explanation (EFA) First, we run a trivial EFA using the psych::fa() function to estimate the eigenvalues. We don’t care about the factors yet, so we can extract a single factor. We also don’t care about interpretable solutions, so we don’t need rotation. ## View the estimated eigenvalues: round(efa_trust0$values, digits = 3) ## [1] 4.980 0.716 0.482 0.165 0.069 0.014 -0.066 -0.092 -0.182 -0.207 ## [11] -0.284 -0.296 -0.319 Click for explanation (eigenvalue extraction) We can check the eigenvalues to see what proportion of the observed variance is accounted for by each additional factor we may extract. Since only one eigenvalue is greater than one, the so-called “Kaiser Criterion” would suggest extracting a single factor. The Kaiser Criterion is not a valid way to select the number of factors in EFA. So, we don’t want to rely on this information alone. We can still use the eigenvalues to help us with factor enumeration, though. One way to do so is by plotting the eigenvalues in a scree plot. 2. Scree plot Click to show code Given a vector of estimated eigenvalues, we can create a scree plot using ggplot() and the geom_line() or geom_path() geometry. library(ggplot2) library(magrittr) efa_trust0 %$% data.frame(y = values, x = 1:length(values)) %>% ggplot(aes(x, y)) + geom_line() + xlab("No. of Factors") + ylab("Eigenvalues") We can also use the psych::scree() function to create a scree plot directly from the data. select(ess, trstlgl:trstplt) %>% scree(pc = FALSE) Click for explanation (scree plot) Although the scree plot provides useful information, we need to interpret that information subjectively, and the conclusions are sometimes ambiguous, in this case. In this case, the plot seems to suggest either one or three components, depending on where we consider the “elbow” to lie. As recommended in the lecture, we can also use “parallel analysis” (Horn, 1965) to provide more objective information about the number of factors. We’ll use the psych::fa.parallel() function to implement parallel analysis. Parallel analysis relies on randomly simulated/permuted data, so we should set a seed to make sure our results are reproducible. We can set the fa = \"fa\" option to get only the results for EFA. 3. Parallel Analysis Click to show code ## Set the random number seed: set.seed(235711) ## Run the parallel analysis: pa_trust <- select(ess, trstlgl:trstplt) %>% fa.parallel(fa = "fa") ## Parallel analysis suggests that the number of factors = 6 and the number of components = NA Click for explanation The results of the parallel analysis suggest 6 factors. If you’ve been paying close attention, you may have noticed that we need to compute the eigenvalues from the original data to run parallel analysis. Hence, we don’t actually need to run a separate EFA to estimate the eigenvalues. ## View the eigenvalues estimated during the parallel analysis: pa_trust$fa.values ## [1] 4.97995262 0.71644127 0.48201040 0.16517645 0.06885820 0.01422241 ## [7] -0.06606777 -0.09225113 -0.18231333 -0.20740917 -0.28415857 -0.29573407 ## [13] -0.31877470 ## Compare to the version from the EFA: pa_trust$fa.values - efa_trust0$values ## [1] 0 0 0 0 0 0 0 0 0 0 0 0 0 ## Recreate the scree plot from above: pa_trust %$% data.frame(y = fa.values, x = 1:length(fa.values)) %>% ggplot(aes(x, y)) + geom_line() + xlab("No. of Factors") + ylab("Eigenvalues") Of course, we also see the same scree plot printed as part of the parallel analysis. So, there’s really no reason to create a separate scree plot, at all, if we’re doing parallel analysis. 4. Conclusion Click for explanation The different criteria disagree on how many factors we should extract, but we have narrowed the range. Based on the scree plot and parallel analysis, we should consider solutions for 3 to 6 factors. We need to examine the factor loadings to see which solution makes the most substantive sense. 4.4.5 Do the same analysis for the attitudes toward immigration items. Click to show code This time, we’ll start by running the parallel analysis and get the eigenvalues and scree plot from psych::fa.parallel(). ## Set the seed: set.seed(235711) ## Run parallel analysis on the 'attitudes' items: pa_att <- select(ess, imsmetn:rfgbfml) %>% fa.parallel(fa = "fa") ## Parallel analysis suggests that the number of factors = 7 and the number of components = NA ## Check the eigenvalues: round(pa_att$fa.values, digits = 3) ## [1] 7.895 1.449 0.734 0.533 0.313 0.156 0.121 0.019 -0.001 -0.064 ## [11] -0.083 -0.103 -0.119 -0.131 -0.150 -0.175 -0.185 -0.200 -0.212 -0.233 ## [21] -0.239 -0.247 -0.334 -0.422 -0.427 Click for explanation For the attitudes toward immigration analysis, the results are even more ambiguous than they were for the trust items. The Kaiser Criterion suggests 2 factors. The scree plot is hopelessly ambiguous. At least 3 factors? No more than 9 factors? Parallel analysis suggests 7 factors Based on the scree plot and parallel analysis, it seems reasonable to consider solutions for 3 to 7 factors. Again, we need to check the substantive interpretation to choose the most reasonable solution. To evaluate the substantive interpretability of the different solutions, we need to estimate the full EFA models for each candidate number of factors. We then compare the factor loadings across solutions to see which set of loadings define the most reasonable set of latent variables. 4.4.6 For the trust items, estimate the EFA models for each plausible number of components that you identified above. Use the psych::fa() function to estimate the models. You will need to specify a few key options. The data (including only the variables you want to analyze) The number of factors that you want to extract The rotation method The estimation method The method of estimating factor scores Hint: You can save yourself a lot of typing/copy-pasting (and the attendant chances of errors) by using a for() loop to iterate through numbers of factors. Click to show code ## Define an empty list to hold all of our fitted EFA objects: efa_trust <- list() ## Loop through the interesting numbers of factors and estimate an EFA for each: for(i in 3:6) efa_trust[[as.character(i)]] <- ess %>% select(trstlgl:trstplt) %>% fa(nfactors = i, # Number of factors = Loop index rotate = "promax", # Oblique rotation scores = "Bartlett") # Estimate factor scores with WLS 4.4.7 Repeat the above analysis for the attitudes items. Click to show code efa_att <- list() for(i in 3:7) efa_att[[as.character(i)]] <- ess %>% select(imsmetn:rfgbfml) %>% fa(nfactors = i, rotate = "promax", scores = "Bartlett") 4.4.8 Compare the factor loading matrices from the models estimated from the Trust items, and select the best solution. Hints: The factor loadings are stored in the loadings slot of the object returned by psych::fa(). Looping can also be useful here. Click to show code for(x in efa_trust) print(x$loadings) ## ## Loadings: ## MR3 MR2 MR1 ## trstlgl 0.839 -0.115 ## trstplc 0.763 -0.218 ## trstun 0.579 0.161 ## trstep 0.554 0.198 ## trstprl 0.444 0.342 ## stfhlth 0.656 -0.125 ## stfedu 0.695 -0.157 ## stfeco -0.102 0.704 0.146 ## stfgov 0.593 0.226 ## stfdem 0.183 0.476 0.150 ## pltinvt 0.813 ## pltcare 0.808 ## trstplt 0.330 0.526 ## ## MR3 MR2 MR1 ## SS loadings 2.299 2.016 1.970 ## Proportion Var 0.177 0.155 0.152 ## Cumulative Var 0.177 0.332 0.483 ## ## Loadings: ## MR2 MR1 MR4 MR3 ## trstlgl 0.797 ## trstplc 0.725 ## trstun 0.656 0.113 ## trstep 1.003 -0.175 ## trstprl 0.121 0.455 0.200 0.112 ## stfhlth 0.663 -0.106 ## stfedu 0.704 -0.110 0.100 ## stfeco 0.729 ## stfgov 0.631 0.175 -0.149 ## stfdem 0.501 0.107 0.115 ## pltinvt 0.855 ## pltcare -0.103 0.863 ## trstplt 0.479 0.340 ## ## MR2 MR1 MR4 MR3 ## SS loadings 2.161 1.952 1.722 1.239 ## Proportion Var 0.166 0.150 0.132 0.095 ## Cumulative Var 0.166 0.316 0.449 0.544 ## ## Loadings: ## MR1 MR4 MR5 MR3 MR2 ## trstlgl 0.935 ## trstplc 0.810 ## trstun 0.505 0.168 ## trstep -0.138 1.128 -0.108 -0.154 ## trstprl 0.359 0.250 0.140 0.201 -0.104 ## stfhlth 0.557 ## stfedu 0.752 ## stfeco 0.710 -0.118 0.172 ## stfgov 0.973 -0.132 ## stfdem 0.556 0.153 ## pltinvt 0.882 ## pltcare 0.855 ## trstplt 0.288 0.308 0.313 ## ## MR1 MR4 MR5 MR3 MR2 ## SS loadings 2.019 1.716 1.655 1.674 0.936 ## Proportion Var 0.155 0.132 0.127 0.129 0.072 ## Cumulative Var 0.155 0.287 0.415 0.543 0.615 ## ## Loadings: ## MR5 MR1 MR4 MR3 MR2 MR6 ## trstlgl 0.980 ## trstplc 0.655 ## trstun 0.911 ## trstep -0.116 0.739 0.163 ## trstprl 0.197 0.577 0.138 ## stfhlth 0.614 ## stfedu 0.771 ## stfeco 0.689 -0.123 0.144 ## stfgov 0.891 ## stfdem 0.513 0.144 ## pltinvt 0.816 ## pltcare 0.778 ## trstplt 0.706 0.193 ## ## MR5 MR1 MR4 MR3 MR2 MR6 ## SS loadings 1.606 1.417 1.442 1.327 1.014 0.879 ## Proportion Var 0.124 0.109 0.111 0.102 0.078 0.068 ## Cumulative Var 0.124 0.233 0.343 0.446 0.524 0.591 Click for explanation Note: Any factor loadings with magnitude lower than 0.1 are suppressed in above output. The factor loadings matrix indicates how strongly each latent factor (columns) associates with the observed items (rows). We can interpret these factor loadings in the same way that we would interpret regression coefficients (indeed, a factor analytic model can be viewed as a multivariate regression model wherein the latent factors are the predictors and the observed items are the outcomes). A higher factor loading indicates a stronger association between the item and factor linked by that loading. Items with high factor loadings are “good” indicators of the respective factors. Items with only very low loadings do not provide much information about any factor. You may want to exclude such items from your analysis. Note that the size of the factor loadings depends on the number of factors. So, you should only consider excluding an observed item after you have chosen the number of latent factors. When we print the loading matrix, we see additional information printed below the factor loadings. Proportion Var: What proportion of the items’ variance is explained by each of the factors. Cumulative Var: How much variance the factors explain, in total. If you estimated as many factors as items, then the Cumulative Var for the final factor would be 1.00 (i.e., 100%). 4.4.9 Compare the factor loading matrices from the models estimated from the Attitudes items, and select the best solution. Click to show code for(x in efa_att) print(x$loadings) ## ## Loadings: ## MR1 MR2 MR3 ## imsmetn 0.802 ## imdfetn 0.754 0.106 ## eimrcnt 0.843 ## eimpcnt 0.814 ## imrcntr 0.857 ## impcntr 0.769 ## qfimchr 0.235 0.858 ## qfimwht 0.132 0.719 ## imwgdwn 0.293 -0.181 ## imhecop 0.371 -0.162 ## imtcjob 0.619 ## imbleco 0.702 ## imbgeco 0.687 ## imueclt 0.561 -0.207 ## imwbcnt 0.732 ## imwbcrm 0.637 ## imrsprc -0.494 -0.125 ## pplstrd 0.249 -0.413 ## vrtrlg -0.275 0.240 ## shrrfg 0.514 -0.111 ## rfgawrk -0.386 ## gvrfgap -0.601 -0.148 ## rfgfrpc 0.432 ## rfggvfn -0.489 ## rfgbfml -0.545 ## ## MR1 MR2 MR3 ## SS loadings 4.819 3.950 1.683 ## Proportion Var 0.193 0.158 0.067 ## Cumulative Var 0.193 0.351 0.418 ## ## Loadings: ## MR2 MR4 MR1 MR3 ## imsmetn 0.788 ## imdfetn 0.731 0.153 0.110 ## eimrcnt 0.855 -0.143 ## eimpcnt 0.790 0.165 ## imrcntr 0.860 ## impcntr 0.743 0.182 ## qfimchr -0.122 0.853 ## qfimwht 0.723 ## imwgdwn 0.638 0.264 ## imhecop 0.680 0.217 ## imtcjob 0.633 0.136 ## imbleco 0.563 -0.212 0.153 ## imbgeco 0.604 -0.168 ## imueclt 0.392 -0.236 -0.168 ## imwbcnt 0.526 -0.282 ## imwbcrm 0.397 -0.292 ## imrsprc 0.616 ## pplstrd 0.231 -0.378 ## vrtrlg 0.279 0.264 ## shrrfg 0.299 -0.271 ## rfgawrk 0.452 ## gvrfgap 0.123 0.774 ## rfgfrpc 0.193 -0.281 ## rfggvfn 0.467 ## rfgbfml 0.619 ## ## MR2 MR4 MR1 MR3 ## SS loadings 3.828 2.778 2.570 1.602 ## Proportion Var 0.153 0.111 0.103 0.064 ## Cumulative Var 0.153 0.264 0.367 0.431 ## ## Loadings: ## MR2 MR1 MR5 MR3 MR4 ## imsmetn 0.792 ## imdfetn 0.728 0.169 0.113 ## eimrcnt 0.910 -0.150 -0.237 ## eimpcnt 0.779 0.126 0.213 ## imrcntr 0.910 -0.128 -0.187 ## impcntr 0.731 0.131 0.236 ## qfimchr 0.109 -0.156 0.882 ## qfimwht 0.139 0.736 ## imwgdwn 0.740 ## imhecop 0.700 ## imtcjob 0.543 0.124 0.182 ## imbleco 0.682 0.135 ## imbgeco 0.799 ## imueclt 0.572 -0.202 ## imwbcnt 0.712 ## imwbcrm 0.545 -0.124 ## imrsprc 0.620 ## pplstrd 0.207 -0.396 ## vrtrlg -0.198 0.151 0.285 0.116 ## shrrfg 0.208 -0.263 0.139 ## rfgawrk 0.457 ## gvrfgap 0.783 ## rfgfrpc -0.338 0.156 ## rfggvfn 0.477 ## rfgbfml -0.125 0.538 ## ## MR2 MR1 MR5 MR3 MR4 ## SS loadings 3.970 2.790 2.215 1.693 1.166 ## Proportion Var 0.159 0.112 0.089 0.068 0.047 ## Cumulative Var 0.159 0.270 0.359 0.427 0.473 ## ## Loadings: ## MR2 MR1 MR6 MR3 MR5 MR4 ## imsmetn 0.705 0.166 ## imdfetn 0.833 ## eimrcnt 0.249 0.859 ## eimpcnt 0.946 ## imrcntr 0.456 0.517 ## impcntr 0.951 ## qfimchr 0.134 -0.122 0.875 ## qfimwht 0.151 0.725 ## imwgdwn 0.748 ## imhecop 0.678 ## imtcjob 0.566 0.123 0.175 ## imbleco 0.753 0.144 ## imbgeco 0.822 ## imueclt 0.580 -0.201 ## imwbcnt 0.751 ## imwbcrm 0.597 ## imrsprc 0.146 0.527 ## pplstrd 0.204 -0.392 ## vrtrlg -0.204 0.143 0.281 0.115 ## shrrfg 0.198 -0.275 0.141 ## rfgawrk 0.517 ## gvrfgap 0.784 ## rfgfrpc -0.294 0.144 ## rfggvfn 0.512 ## rfgbfml 0.596 ## ## MR2 MR1 MR6 MR3 MR5 MR4 ## SS loadings 3.304 3.013 1.994 1.649 1.065 1.133 ## Proportion Var 0.132 0.121 0.080 0.066 0.043 0.045 ## Cumulative Var 0.132 0.253 0.332 0.398 0.441 0.486 ## ## Loadings: ## MR2 MR1 MR6 MR3 MR5 MR7 MR4 ## imsmetn 0.700 0.162 ## imdfetn 0.821 ## eimrcnt 0.245 0.879 ## eimpcnt 0.935 ## imrcntr 0.452 0.523 ## impcntr 0.938 ## qfimchr 0.751 ## qfimwht 0.720 ## imwgdwn 0.700 ## imhecop 0.172 0.624 ## imtcjob 0.574 -0.120 0.174 ## imbleco 0.679 0.108 ## imbgeco 0.832 -0.145 ## imueclt 0.531 -0.191 ## imwbcnt 0.649 0.138 ## imwbcrm 0.464 0.131 0.290 ## imrsprc 0.146 0.440 -0.100 ## pplstrd -0.274 0.392 ## vrtrlg -0.121 0.190 -0.297 0.115 ## shrrfg -0.124 0.437 0.131 ## rfgawrk 0.538 ## gvrfgap 0.616 -0.237 ## rfgfrpc -0.131 0.437 0.135 ## rfggvfn 0.504 ## rfgbfml 0.526 ## ## MR2 MR1 MR6 MR3 MR5 MR7 MR4 ## SS loadings 3.224 2.467 1.456 1.305 1.105 0.901 0.984 ## Proportion Var 0.129 0.099 0.058 0.052 0.044 0.036 0.039 ## Cumulative Var 0.129 0.228 0.286 0.338 0.382 0.418 0.458 It is very possible that you selected a different numbers of factors than Kestilä (2006). We need to keep these exercises consistent, though. So, the remaining questions will all assume you have extract three factors from the Trust items and five factors from the Attitudes items, to parallel the Kestilä (2006) results. ## Select the three-factor solution for 'trust': efa_trust <- efa_trust[["3"]] ## Select the five-factor solution for 'attitudes': efa_att <- efa_att[["5"]] 4.4.10 Give the factor scores meaningful names, and add the scores to the ess dataset as new columns. Hint: If you’re not sure of what do to, check 4.3.11. Click to show code ## Rename the factor scores: colnames(efa_trust$scores) <- c("trust_inst", "satisfy", "trust_pol") colnames(efa_att$scores) <- c("effects", "allowance", "refugees", "ethnic", "europe") ## Add factor scores to the dataset as new columns: ess <- data.frame(ess, efa_trust$scores, efa_att$scores) Kestilä (2006) used the component scores to descriptively evaluate country-level differences in Attitudes toward Immigration and Political Trust. So, now it’s time to replicate those analyses. 4.4.11 Repeat the Kestilä (2006) between-country comparison using the factor scores you created in 4.4.10 and an appropriate statistical test. Click to show code Here, we’ll only demonstrate a possible approach to analyzing one of the Trust dimensions. We can use a linear model to test whether the countries differ in average levels of Trust in Institutions (as quantified by the relevant factor score). ## Estimate the model: out <- lm(trust_inst ~ country, data = ess) ## View the regression-style summary: summary(out) ## ## Call: ## lm(formula = trust_inst ~ country, data = ess) ## ## Residuals: ## Min 1Q Median 3Q Max ## -4.2295 -0.6226 0.1171 0.7194 3.3061 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) -0.09028 0.02445 -3.692 0.000224 *** ## countryBelgium -0.28923 0.03642 -7.942 2.12e-15 *** ## countryGermany -0.05966 0.03211 -1.858 0.063205 . ## countryDenmark 0.75509 0.03882 19.452 < 2e-16 *** ## countryFinland 0.59235 0.03439 17.224 < 2e-16 *** ## countryItaly 0.10991 0.04071 2.700 0.006939 ** ## countryNetherlands -0.05357 0.03379 -1.585 0.112893 ## countryNorway 0.36922 0.03493 10.570 < 2e-16 *** ## countrySweden 0.28560 0.03613 7.904 2.89e-15 *** ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 1.029 on 14769 degrees of freedom ## (4912 observations deleted due to missingness) ## Multiple R-squared: 0.082, Adjusted R-squared: 0.0815 ## F-statistic: 164.9 on 8 and 14769 DF, p-value: < 2.2e-16 ## View the results as an ANOVA table: anova(out) ## Post-hoc tests out %>% aov() %>% TukeyHSD() ## Tukey multiple comparisons of means ## 95% family-wise confidence level ## ## Fit: aov(formula = .) ## ## $country ## diff lwr upr p adj ## Belgium-Austria -0.289225482 -0.40219224 -0.17625873 0.0000000 ## Germany-Austria -0.059655996 -0.15926604 0.03995405 0.6429963 ## Denmark-Austria 0.755089552 0.63466911 0.87551000 0.0000000 ## Finland-Austria 0.592348290 0.48565882 0.69903776 0.0000000 ## Italy-Austria 0.109910185 -0.01636587 0.23618624 0.1476635 ## Netherlands-Austria -0.053567808 -0.15838407 0.05124846 0.8131104 ## Norway-Austria 0.369224250 0.26085692 0.47759158 0.0000000 ## Sweden-Austria 0.285601197 0.17350905 0.39769334 0.0000000 ## Germany-Belgium 0.229569486 0.12386351 0.33527546 0.0000000 ## Denmark-Belgium 1.044315033 0.91880537 1.16982470 0.0000000 ## Finland-Belgium 0.881573772 0.76917165 0.99397589 0.0000000 ## Italy-Belgium 0.399135667 0.26799745 0.53027389 0.0000000 ## Netherlands-Belgium 0.235657673 0.12503199 0.34628336 0.0000000 ## Norway-Belgium 0.658449732 0.54445381 0.77244566 0.0000000 ## Sweden-Belgium 0.574826679 0.45728417 0.69236918 0.0000000 ## Denmark-Germany 0.814745547 0.70110863 0.92838247 0.0000000 ## Finland-Germany 0.652004286 0.55303505 0.75097352 0.0000000 ## Italy-Germany 0.169566181 0.04974170 0.28939066 0.0003895 ## Netherlands-Germany 0.006088188 -0.09085878 0.10303516 0.9999999 ## Norway-Germany 0.428880246 0.32810453 0.52965596 0.0000000 ## Sweden-Germany 0.345257193 0.24048642 0.45002796 0.0000000 ## Finland-Denmark -0.162741262 -0.28263218 -0.04285034 0.0008579 ## Italy-Denmark -0.645179366 -0.78279052 -0.50756821 0.0000000 ## Netherlands-Denmark -0.808657360 -0.92688442 -0.69043030 0.0000000 ## Norway-Denmark -0.385865301 -0.50725174 -0.26447886 0.0000000 ## Sweden-Denmark -0.469488354 -0.59421139 -0.34476531 0.0000000 ## Italy-Finland -0.482438105 -0.60820928 -0.35666693 0.0000000 ## Netherlands-Finland -0.645916098 -0.75012357 -0.54170862 0.0000000 ## Norway-Finland -0.223124040 -0.33090264 -0.11534544 0.0000000 ## Sweden-Finland -0.306747093 -0.41827017 -0.19522402 0.0000000 ## Netherlands-Italy -0.163477993 -0.28766412 -0.03929186 0.0014719 ## Norway-Italy 0.259314065 0.13211649 0.38651164 0.0000000 ## Sweden-Italy 0.175691012 0.04530545 0.30607657 0.0009794 ## Norway-Netherlands 0.422792059 0.31686740 0.52871671 0.0000000 ## Sweden-Netherlands 0.339169005 0.22943659 0.44890142 0.0000000 ## Sweden-Norway -0.083623053 -0.19675232 0.02950622 0.3462227 Click for explanation According to the omnibus F-test, average levels of Trust in Institutions significantly differ between countries, but this test cannot tell us between which countries the differences lie. Similarly, the t statistics associated with each dummy code in the regression-style summary only tell us if that country differs significantly from the reference country (i.e., Austria), but we cannot see, for example, if there is a significant difference in average trust levels between Belgium and the Netherlands. One way to test for differences between the individual countries would be a post hoc test of all pairwise comparisons. Since we’ll be doing 45 tests, we need to apply a correction for repeated testing. Above, we use the TukeyHSD() function to conduct all pairwise comparisons while applying Tukey’s HSD correction. The TukeyHSD() function only accepts models estimated with the aov() function, so we first pass our fitted lm object through aov(). The second part of the Kestilä (2006) analysis was to evaluate how socio-demographic characteristics affected attitudes towards immigrants and trust in politics among the Finnish electorate. Before we can replicate this part of the analysis, we need to subset the data to only the Finnish cases. 4.4.12 Create a new data frame that contains only the Finnish cases from ess. Hint: You can use logical indexing based on the country variable. Click to show code ess_finland <- filter(ess, country == "Finland") We still have one more step before we can estimate any models. We must prepare our variables for analysis. Our dependent variables will be the factor scores generated above. So, we do not need to apply any further processing. We have not yet used any of the independent variables, though. So, we should inspect those variables to see if they require any processing. In our processed ess data, the relevant variables have the following names: sex yrbrn eduyrs polintr lrscale 4.4.13 Inspect the independent variables listed above. Click to show code library(tidySEM) select(ess_finland, sex, yrbrn, eduyrs, polintr, lrscale) %>% descriptives() Click for explanation It looks like we still need some recoding. 4.4.14 Apply any necessary recoding/transformations. 1. Age Click to show code ess_finland <- mutate(ess_finland, age = 2002 - yrbrn) Click for explanation The data contain the participants’ years of birth instead of their age, but Kestilä analyzed age. Fortunately, we know that the data were collected in 2002, so we can simply subtract each participant’s value of yrbrn from the 2002 to compute their age. 2. Political Interest Click to show code First, we’ll transform polintr. ## Recode the four factor levels into two factor levels: ess_finland <- mutate(ess_finland, polintr_bin = recode_factor(polintr, "Not at all interested" = "Low Interest", "Hardly interested" = "Low Interest", "Quite interested" = "High Interest", "Very interested" = "High Interest") ) ## Check the conversion: with(ess_finland, table(old = polintr, new = polintr_bin, useNA = "always")) ## new ## old Low Interest High Interest <NA> ## Very interested 0 144 0 ## Quite interested 0 785 0 ## Hardly interested 842 0 0 ## Not at all interested 228 0 0 ## <NA> 0 0 1 Click for explanation Kestilä (2006) dichotomized polintr by combining the lowest two and highest two categories. So, we don’t actually want to convert the polint variable into a numeric, Likert-type variable. We want polint to be a binary factor. The recode_factor() function from dplyr() will automatically convert our result into a factor. As with the ess_round1.rds data, we will be coming back to this Finnish subsample data in future practical exercises. So, we should save our work by writing the processed dataset to disk. 4.4.15 Use the saveRDS() function to save the processed Finnish subsample data. Click to see code ## Save the processed Finnish data: saveRDS(ess_finland, "ess_finland.rds") Now, we’re finally ready to replicate the regression analysis from Kestilä (2006). Creating a single aggregate score by summing the individual component scores is a pretty silly thing to do, though. So, we won’t reproduce that aspect of the analysis. 4.4.16 Run a series of multiple linear regression analyses with the factor scores you created in 4.4.10 as the dependent variables and the same predictors used by Kestilä (2006). Do your results agree with those reported by Kestilä (2006)? Click to show code ## Predicting 'Trust in Institutions': out_trust_inst <- lm(trust_inst ~ sex + age + eduyrs + polintr_bin + lrscale, data = ess_finland) summary(out_trust_inst) ## ## Call: ## lm(formula = trust_inst ~ sex + age + eduyrs + polintr_bin + ## lrscale, data = ess_finland) ## ## Residuals: ## Min 1Q Median 3Q Max ## -3.9499 -0.5102 0.1337 0.6638 2.5919 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) -0.057518 0.124294 -0.463 0.643595 ## sexFemale 0.004091 0.045170 0.091 0.927849 ## age -0.003071 0.001380 -2.225 0.026219 * ## eduyrs 0.023223 0.006388 3.635 0.000286 *** ## polintr_binHigh Interest 0.166860 0.046448 3.592 0.000337 *** ## lrscale 0.058951 0.011232 5.249 1.72e-07 *** ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 0.9321 on 1734 degrees of freedom ## (260 observations deleted due to missingness) ## Multiple R-squared: 0.04155, Adjusted R-squared: 0.03879 ## F-statistic: 15.03 on 5 and 1734 DF, p-value: 1.78e-14 ## Predicting 'Trust in Politicians': out_trust_pol <- lm(trust_pol ~ sex + age + eduyrs + polintr_bin + lrscale, data = ess_finland) summary(out_trust_pol) ## ## Call: ## lm(formula = trust_pol ~ sex + age + eduyrs + polintr_bin + lrscale, ## data = ess_finland) ## ## Residuals: ## Min 1Q Median 3Q Max ## -3.03673 -0.67306 0.05346 0.69666 2.38771 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) -0.165989 0.126840 -1.309 0.19083 ## sexFemale 0.015572 0.046095 0.338 0.73554 ## age -0.009112 0.001409 -6.469 1.28e-10 *** ## eduyrs 0.018476 0.006519 2.834 0.00465 ** ## polintr_binHigh Interest 0.463763 0.047399 9.784 < 2e-16 *** ## lrscale 0.054932 0.011462 4.793 1.79e-06 *** ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 0.9512 on 1734 degrees of freedom ## (260 observations deleted due to missingness) ## Multiple R-squared: 0.09806, Adjusted R-squared: 0.09546 ## F-statistic: 37.71 on 5 and 1734 DF, p-value: < 2.2e-16 ## Predicting 'Attitudes toward Refugees': out_refugees <- lm(refugees ~ sex + age + eduyrs + polintr_bin + lrscale, data = ess_finland) summary(out_refugees) ## ## Call: ## lm(formula = refugees ~ sex + age + eduyrs + polintr_bin + lrscale, ## data = ess_finland) ## ## Residuals: ## Min 1Q Median 3Q Max ## -2.9118 -0.6860 -0.0594 0.6904 4.1044 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) -1.690e-01 1.438e-01 -1.175 0.240080 ## sexFemale -4.828e-01 5.181e-02 -9.318 < 2e-16 *** ## age 2.903e-05 1.604e-03 0.018 0.985561 ## eduyrs -2.537e-02 7.459e-03 -3.401 0.000688 *** ## polintr_binHigh Interest -2.131e-01 5.345e-02 -3.986 6.99e-05 *** ## lrscale 9.359e-02 1.296e-02 7.223 7.65e-13 *** ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 1.06 on 1699 degrees of freedom ## (295 observations deleted due to missingness) ## Multiple R-squared: 0.09535, Adjusted R-squared: 0.09269 ## F-statistic: 35.81 on 5 and 1699 DF, p-value: < 2.2e-16 That does it for our replication of the Kesilä (2006) analyses, but we still have one more topic to consider in this practical. One of the most common applications of EFA is scale development. Given a pool of items without a known factor structure, we try to estimate the underlying latent factors that define the (sub)scales represented by our items. In such applications, we use the factor loading matrix for our optimal solution to make “bright-line” assignments of items to putative factors according to the simple structure represented by the estimated factor loading matrix. In other words, we disregard small factor loadings and assign observed items to only the single latent factor upon which they load most strongly. We then hypothesize that those items are true indicators of that latent factor. We can use confirmatory factor analysis (which you will learn about next week) to test rigorously this hypothesis, but we can already get started by estimating the internal consistency (a type of reliability) of the hypothesized subscales. 4.4.17 Estimate the internal consistency of the three Trust subscales and five Attitudes subscales implied by your EFA solutions from above. Use Cronbach’s Alpha to quantify internal consistency. Use the alpha() function from the psych package to conduct the analysis. Run your analysis on the full ess dataset, not the Finnish subset. Are the subscales implied by your EFA reliable, in the sense of good internal consistency? Note that \\(\\alpha > 0.7\\) is generally considered acceptable, and \\(\\alpha > 0.8\\) is usually considered good. Click to show code ## Run the reliability analysis on the subscale data: ( out <- select(ess, starts_with("stf")) %>% psych::alpha() ) ## ## Reliability analysis ## Call: psych::alpha(x = .) ## ## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r ## 0.79 0.79 0.77 0.44 3.9 0.0023 5.4 1.7 0.41 ## ## 95% confidence boundaries ## lower alpha upper ## Feldt 0.79 0.79 0.8 ## Duhachek 0.79 0.79 0.8 ## ## Reliability if an item is dropped: ## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r ## stfhlth 0.78 0.78 0.73 0.47 3.5 0.0026 0.0064 0.46 ## stfedu 0.76 0.76 0.72 0.45 3.2 0.0028 0.0109 0.44 ## stfeco 0.74 0.74 0.70 0.41 2.8 0.0031 0.0069 0.39 ## stfgov 0.74 0.74 0.69 0.42 2.9 0.0030 0.0035 0.41 ## stfdem 0.75 0.75 0.71 0.43 3.0 0.0029 0.0074 0.40 ## ## Item statistics ## n raw.r std.r r.cor r.drop mean sd ## stfhlth 19481 0.69 0.69 0.56 0.50 5.8 2.3 ## stfedu 18844 0.73 0.73 0.62 0.55 5.9 2.3 ## stfeco 19211 0.78 0.78 0.70 0.63 5.0 2.4 ## stfgov 19106 0.77 0.76 0.69 0.61 4.5 2.3 ## stfdem 19106 0.75 0.75 0.67 0.59 5.7 2.3 Click for explanation Here, we estimate the reliability of the Satisfaction subscale from the Trust analysis. According to our EFA, the Satisfaction subscale should be indicated by the following five variables: stfeco stfgov stfdem stfedu stfhlth We select these variables using the tidy-select function starts_with() to extract all variables beginning with the three characters “stf”. To estimate the internal consistency of this subscale, we simply provide a data frame containing only the subscale data to the alpha() function. The raw_alpha value is the estimate of Cronbach’s Alpha. In this case \\(\\alpha = 0.794\\), so the subscale is pretty reliable. The table labeled “Reliability if an item is dropped” shows what Cronbach’s Alpha would be if each item were excluded from the scale. If this value is notably higher than the raw_alpha value, it could indicate a bad item. Note that reliability is only one aspect of scale quality, though. So, you shouldn’t throw out items just because they perform poorly in reliability analysis. End of In-Class Exercises "],["cfa.html", "5 CFA", " 5 CFA This week, we will introduce confirmatory factor analysis (CFA) and discuss how it differs from EFA. Furthermore, we will revisit the idea of model fit and introduce into the R-package lavaan. Homework before the lecture Watch the Lecture Recording for this week. Complete the Reading for this week, and answer the associated reading questions. Homework before the practical Complete the At-Home Exercises. Practical content During the practical you will work on the In-Class Exercises. "],["lecture-4.html", "5.1 Lecture", " 5.1 Lecture Often, we work with scales that have a validated or hypothesized factor structure. In the former case, the scale structure has been validated through previous psychometric studies. In the latter case, we may have conducted an EFA to estimate the factor structure on prior data, or theory/intuition may suggest a plausible structure. Regardless of how we come to expect a given factor structure, such situations represent confirmatory modeling problems, because we are attempting to empirically confirm an a priori expectation. Hence, exploratory methods like EFA are not appropriate, and we should employ confirmatory modeling techniques. This week we consider one such technique: confirmatory factor analysis (CFA). As the name suggests, CFA is related to the EFA methods we discussed last week in that both methods are flavors of factor analysis. However, the two methods address fundamentally different research questions. Rather than attempting to estimate an unknown factor structure (as in EFA), we now want to compare a hypothesized measurement model (i.e., factor structure) to observed data in order to evaluate the model’s plausibility. 5.1.1 Recording Note: When Caspar discusses the complexity of the second-order CFA model, it’s easy to misunderstand his statements. We need to be careful not to over-generalize. In general, a second-order CFA is not more complex than a first-order CFA. Actually, in most practical applications, the opposite is true. A second-order CFA is more complex than a first-order CFA, when the factors in the first-order CFA are uncorrelated. This is the situation Caspar references in the recording when claiming that the second-order model is more complex. We hardly ever want to fit such first-order CFA, though. The default CFA fully saturates the latent covariance structure. If the factors in the first-order CFA are fully correlated (according to standard practice), and we include a single second-order factor, the following statements hold. If the first-order CFA has more than three factors, the first-order model is more complex than the second-order model. If the first-order model has three or fewer factors, the first- and second-order models are equivalent (due to scaling constraints we need to impose to identify the second-order model). The second-order model cannot be more complex than the first-order model (assuming both models are correctly identified and no extra constraints are imposed). The above statements may not hold in more complex situations (e.g., more than one second-order factor, partially saturated first-order correlation structure, etc.). You can always identify the more complex model by calculating the degrees of freedom for both models. The model with fewer degrees of freedom is more complex. 5.1.2 Slides You can download the lecture slides here "],["reading-4.html", "5.2 Reading", " 5.2 Reading Reference Byrne, B. (2005). Factor analytic models: Viewing the structure of an assessment instrument from three perspectives, Journal of Personality Assessment, 85(1), 17–32. Questions What are the main differences between exploratory factor analysis (EFA) and confirmatory factor analysis (CFA)? In which circumstances should a researcher use EFA, and in which should they use CFA? What are the five main limitations of EFA that CFA overcomes? In which circumstances can a second order CFA model be useful? Consider the following four techniques: PCA, EFA, CFA, second order CFA. For each of the following three research situations, which of the above techniques would you use and why? A researcher has developed a new questionnaire that should measure personality and wants to know how many factors underlie the items in their new measure. A researcher is modeling data collected with a seven-item scale that has been used since the 1960s to measure authoritarianism. A researcher has recorded highest completed level of education, years of education, and highest level of education attempted for all respondents in a survey. The researcher wants to include some operationalization of the concept of ‘education’ in their model but is unsure of which observed variable to use. "],["at-home-exercises-4.html", "5.3 At-Home Exercises", " 5.3 At-Home Exercises This week, we will wrap up our re-analysis of the Kestilä (2006) results. During this practical, you will conduct a CFA of the Trust in Politics items and compare the results to those obtained from your previous EFA- and PCA-based replications of Kestilä (2006). 5.3.1 Load the ESS data. The relevant data are contained in the ess_round1.rds file. This file is in R Data Set (RDS) format. The dataset is already stored as a data frame with the processing and cleaning that you should have done for previous practicals completed. Click to show code ess <- readRDS("ess_round1.rds") Although you may have settled on any number of EFA solutions during the Week 4 In-Class Exercises, we are going to base the following CFA on a three-factor model of Trust in Politics similar to the original PCA results from Kestilä (2006). Note: Unless otherwise specified, all following questions refer to the Trust in Politics items. We will not consider the Attitudes toward Immigration items in these exercises. 5.3.2 Define the lavaan model syntax for the CFA implied by the three-factor EFA solution you found in the Week 4 In-Class Exercises. Covary the three latent factors. Do not specify any mean structure. Save this model syntax as an object in your environment. Click to show code mod_3f <- ' institutions =~ trstlgl + trstplc + trstun + trstep + trstprl satisfaction =~ stfhlth + stfedu + stfeco + stfgov + stfdem politicians =~ pltinvt + pltcare + trstplt ' Click for explanation We don’t have to specify the latent covariances in the model syntax, we can tell lavaan to estimate all latent covariances when we fit the model. 5.3.3 Estimate the CFA model you defined above, and summarize the results. Use the lavaan::cfa() function to estimate the model. Use the default settings for the cfa() function. Request the model fit statistics with the summary by supplying the fit.measures = TRUE argument to summary(). Request the standardized parameter estimates with the summary by supplying the standardized = TRUE argument to summary(). Check the results, and answer the following questions: Does the model fit the data well? How are the latent variances and covariances specified when using the default settings? How is the model identified when using the default settings? Click the code ## Load the lavaan package: library(lavaan) ## Estimate the CFA model: fit_3f <- cfa(mod_3f, data = ess) ## Summarize the fitted model: summary(fit_3f, fit.measures = TRUE, standardized = TRUE) ## lavaan 0.6.16 ended normally after 46 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 29 ## ## Used Total ## Number of observations 14778 19690 ## ## Model Test User Model: ## ## Test statistic 10652.207 ## Degrees of freedom 62 ## P-value (Chi-square) 0.000 ## ## Model Test Baseline Model: ## ## Test statistic 81699.096 ## Degrees of freedom 78 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.870 ## Tucker-Lewis Index (TLI) 0.837 ## ## Loglikelihood and Information Criteria: ## ## Loglikelihood user model (H0) -371404.658 ## Loglikelihood unrestricted model (H1) -366078.555 ## ## Akaike (AIC) 742867.317 ## Bayesian (BIC) 743087.743 ## Sample-size adjusted Bayesian (SABIC) 742995.583 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.108 ## 90 Percent confidence interval - lower 0.106 ## 90 Percent confidence interval - upper 0.109 ## P-value H_0: RMSEA <= 0.050 0.000 ## P-value H_0: RMSEA >= 0.080 1.000 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.059 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## institutions =~ ## trstlgl 1.000 1.613 0.677 ## trstplc 0.770 0.012 61.866 0.000 1.241 0.567 ## trstun 0.929 0.013 69.227 0.000 1.498 0.642 ## trstep 0.908 0.013 70.929 0.000 1.464 0.660 ## trstprl 1.139 0.014 84.084 0.000 1.837 0.809 ## satisfaction =~ ## stfhlth 1.000 1.173 0.521 ## stfedu 1.106 0.022 50.840 0.000 1.297 0.577 ## stfeco 1.415 0.025 57.214 0.000 1.659 0.713 ## stfgov 1.480 0.025 58.764 0.000 1.736 0.756 ## stfdem 1.384 0.024 57.904 0.000 1.623 0.731 ## politicians =~ ## pltinvt 1.000 0.646 0.613 ## pltcare 1.021 0.016 62.862 0.000 0.660 0.628 ## trstplt 3.012 0.039 76.838 0.000 1.946 0.891 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## institutions ~~ ## satisfaction 1.391 0.032 43.206 0.000 0.736 0.736 ## politicians 0.909 0.018 49.934 0.000 0.872 0.872 ## satisfaction ~~ ## politicians 0.539 0.013 41.053 0.000 0.711 0.711 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## .trstlgl 3.068 0.041 75.262 0.000 3.068 0.541 ## .trstplc 3.248 0.041 80.037 0.000 3.248 0.678 ## .trstun 3.197 0.041 77.141 0.000 3.197 0.588 ## .trstep 2.776 0.036 76.243 0.000 2.776 0.564 ## .trstprl 1.776 0.029 61.361 0.000 1.776 0.345 ## .stfhlth 3.695 0.046 79.989 0.000 3.695 0.729 ## .stfedu 3.368 0.043 77.916 0.000 3.368 0.667 ## .stfeco 2.656 0.038 69.070 0.000 2.656 0.491 ## .stfgov 2.264 0.035 64.201 0.000 2.264 0.429 ## .stfdem 2.289 0.034 67.172 0.000 2.289 0.465 ## .pltinvt 0.694 0.009 78.255 0.000 0.694 0.624 ## .pltcare 0.668 0.009 77.562 0.000 0.668 0.605 ## .trstplt 0.978 0.028 34.461 0.000 0.978 0.205 ## institutions 2.601 0.059 44.198 0.000 1.000 1.000 ## satisfaction 1.375 0.044 31.407 0.000 1.000 1.000 ## politicians 0.417 0.011 38.843 0.000 1.000 1.000 Click for explanation No, the model does not seem to fit the data well. The SRMR looks good, but one good looking fit statistic is not enough. The RMSEA, TLI, and CFI are all in the “unacceptable” range. The \\(\\chi^2\\) is highly significant, but we don’t care. The cfa() function is just a wrapper for the lavaan() function with several options set at the defaults you would want for a standard CFA. By default: All latent variances and covariances are freely estimated (due to the argument auto.cov.lv.x = TRUE) The model is identified by fixing the first factor loading of each factor to 1 (due to the argument auto.fix.first = TRUE) To see a full list of the (many) options you can specify to tweak the behavior of lavaan estimation functions run ?lavOptions. Now, we will consider a couple of alternative factor structures for the Trust in Politics CFA. First, we will go extremely simple by estimating a one-factor model wherein all Trust items are explained by a single latent variable. 5.3.4 Define the lavaan model syntax for a one-factor model of the Trust items. Save this syntax as an object in your environment. Click to show code mod_1f <- ' political_trust =~ trstlgl + trstplc + trstun + trstep + trstprl + stfhlth + stfedu + stfeco + stfgov + stfdem + pltinvt + pltcare + trstplt ' 5.3.5 Estimate the one-factor model, and summarize the results. Does this model appear to fit better or worse than the three-factor model? Note: You can use the lavaan::fitMeasures() function to extract only the model fit information from a fitted lavaan object. Click to show code ## Estimate the one factor model: fit_1f <- cfa(mod_1f, data = ess) ## Summarize the results: summary(fit_1f, fit.measures = TRUE) ## lavaan 0.6.16 ended normally after 33 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 26 ## ## Used Total ## Number of observations 14778 19690 ## ## Model Test User Model: ## ## Test statistic 17667.304 ## Degrees of freedom 65 ## P-value (Chi-square) 0.000 ## ## Model Test Baseline Model: ## ## Test statistic 81699.096 ## Degrees of freedom 78 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.784 ## Tucker-Lewis Index (TLI) 0.741 ## ## Loglikelihood and Information Criteria: ## ## Loglikelihood user model (H0) -374912.206 ## Loglikelihood unrestricted model (H1) -366078.555 ## ## Akaike (AIC) 749876.413 ## Bayesian (BIC) 750074.036 ## Sample-size adjusted Bayesian (SABIC) 749991.410 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.135 ## 90 Percent confidence interval - lower 0.134 ## 90 Percent confidence interval - upper 0.137 ## P-value H_0: RMSEA <= 0.050 0.000 ## P-value H_0: RMSEA >= 0.080 1.000 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.080 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) ## political_trust =~ ## trstlgl 1.000 ## trstplc 0.774 0.013 57.949 0.000 ## trstun 0.930 0.014 64.200 0.000 ## trstep 0.909 0.014 65.679 0.000 ## trstprl 1.182 0.015 79.401 0.000 ## stfhlth 0.615 0.013 45.947 0.000 ## stfedu 0.695 0.014 51.424 0.000 ## stfeco 0.895 0.014 62.316 0.000 ## stfgov 0.985 0.014 68.200 0.000 ## stfdem 0.998 0.014 70.899 0.000 ## pltinvt 0.382 0.006 59.215 0.000 ## pltcare 0.396 0.006 61.195 0.000 ## trstplt 1.183 0.014 81.716 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .trstlgl 3.370 0.042 79.787 0.000 ## .trstplc 3.410 0.041 82.311 0.000 ## .trstun 3.451 0.043 80.749 0.000 ## .trstep 3.019 0.038 80.272 0.000 ## .trstprl 1.938 0.027 70.878 0.000 ## .stfhlth 4.201 0.050 84.093 0.000 ## .stfedu 3.941 0.047 83.419 0.000 ## .stfeco 3.565 0.044 81.289 0.000 ## .stfgov 3.044 0.038 79.326 0.000 ## .stfdem 2.631 0.034 78.072 0.000 ## .pltinvt 0.775 0.009 82.043 0.000 ## .pltcare 0.743 0.009 81.579 0.000 ## .trstplt 1.548 0.023 67.052 0.000 ## political_trst 2.299 0.055 41.569 0.000 ## Compare fit statistics: fitMeasures(fit_3f) ## npar fmin chisq ## 29.000 0.360 10652.207 ## df pvalue baseline.chisq ## 62.000 0.000 81699.096 ## baseline.df baseline.pvalue cfi ## 78.000 0.000 0.870 ## tli nnfi rfi ## 0.837 0.837 0.836 ## nfi pnfi ifi ## 0.870 0.691 0.870 ## rni logl unrestricted.logl ## 0.870 -371404.658 -366078.555 ## aic bic ntotal ## 742867.317 743087.743 14778.000 ## bic2 rmsea rmsea.ci.lower ## 742995.583 0.108 0.106 ## rmsea.ci.upper rmsea.ci.level rmsea.pvalue ## 0.109 0.900 0.000 ## rmsea.close.h0 rmsea.notclose.pvalue rmsea.notclose.h0 ## 0.050 1.000 0.080 ## rmr rmr_nomean srmr ## 0.255 0.255 0.059 ## srmr_bentler srmr_bentler_nomean crmr ## 0.059 0.059 0.064 ## crmr_nomean srmr_mplus srmr_mplus_nomean ## 0.064 0.059 0.059 ## cn_05 cn_01 gfi ## 113.901 126.971 0.897 ## agfi pgfi mfi ## 0.849 0.611 0.699 ## ecvi ## 0.725 fitMeasures(fit_1f) ## npar fmin chisq ## 26.000 0.598 17667.304 ## df pvalue baseline.chisq ## 65.000 0.000 81699.096 ## baseline.df baseline.pvalue cfi ## 78.000 0.000 0.784 ## tli nnfi rfi ## 0.741 0.741 0.741 ## nfi pnfi ifi ## 0.784 0.653 0.784 ## rni logl unrestricted.logl ## 0.784 -374912.206 -366078.555 ## aic bic ntotal ## 749876.413 750074.036 14778.000 ## bic2 rmsea rmsea.ci.lower ## 749991.410 0.135 0.134 ## rmsea.ci.upper rmsea.ci.level rmsea.pvalue ## 0.137 0.900 0.000 ## rmsea.close.h0 rmsea.notclose.pvalue rmsea.notclose.h0 ## 0.050 1.000 0.080 ## rmr rmr_nomean srmr ## 0.364 0.364 0.080 ## srmr_bentler srmr_bentler_nomean crmr ## 0.080 0.080 0.087 ## crmr_nomean srmr_mplus srmr_mplus_nomean ## 0.087 0.080 0.080 ## cn_05 cn_01 gfi ## 71.949 79.980 0.825 ## agfi pgfi mfi ## 0.756 0.590 0.551 ## ecvi ## 1.199 Click for explanation The one-factor model definitely seems to fit worse than the three-factor model. A second order CFA model is another way of representing the latent structure underlying a set of items. As you read in Byrne (2005), however, the second order CFA is only appropriate in certain circumstances. 5.3.6 Given the CFA results above, would a second order CFA be appropriate for the Trust data? Why or why not? Click for explanation Yes, a second order CFA model is a theoretically appropriate representation of the Trust items. The first order latent variables in the three-factor model are all significantly correlated. The first order latent variables in the three-factor model seem to tap different aspects of some single underlying construct. 5.3.7 Define the lavaan model syntax for a second-order CFA model of the Trust items. Use the three factors defined in 5.3.2 as the first order factors. Click to show code mod_2nd <- ' institutions =~ trstlgl + trstplc + trstun + trstep + trstprl satisfaction =~ stfhlth + stfedu + stfeco + stfgov + stfdem politicians =~ pltinvt + pltcare + trstplt trust =~ politicians + satisfaction + institutions ' Click for explanation To define the second order factor, we use the same syntactic conventions that we employ to define a first order factor. The only differences is that the “indicators” of the second order factor (i.e., the variables listed on the RHS of the =~ operator) are previously defined first order latent variables. 5.3.8 Estimate the second order CFA model, and summarize the results. Does this model fit better or worse than the three-factor model? Is this model more or less complex than the three-factor model? What information can you use to quantify this difference in complexity? Click to show code fit_2nd <- cfa(mod_2nd, data = ess) summary(fit_2nd, fit.measures = TRUE, standardized = TRUE) ## lavaan 0.6.16 ended normally after 44 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 29 ## ## Used Total ## Number of observations 14778 19690 ## ## Model Test User Model: ## ## Test statistic 10652.207 ## Degrees of freedom 62 ## P-value (Chi-square) 0.000 ## ## Model Test Baseline Model: ## ## Test statistic 81699.096 ## Degrees of freedom 78 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.870 ## Tucker-Lewis Index (TLI) 0.837 ## ## Loglikelihood and Information Criteria: ## ## Loglikelihood user model (H0) -371404.658 ## Loglikelihood unrestricted model (H1) -366078.555 ## ## Akaike (AIC) 742867.317 ## Bayesian (BIC) 743087.743 ## Sample-size adjusted Bayesian (SABIC) 742995.583 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.108 ## 90 Percent confidence interval - lower 0.106 ## 90 Percent confidence interval - upper 0.109 ## P-value H_0: RMSEA <= 0.050 0.000 ## P-value H_0: RMSEA >= 0.080 1.000 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.059 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## institutions =~ ## trstlgl 1.000 1.613 0.677 ## trstplc 0.770 0.012 61.866 0.000 1.241 0.567 ## trstun 0.929 0.013 69.227 0.000 1.498 0.642 ## trstep 0.908 0.013 70.929 0.000 1.464 0.660 ## trstprl 1.139 0.014 84.084 0.000 1.837 0.809 ## satisfaction =~ ## stfhlth 1.000 1.173 0.521 ## stfedu 1.106 0.022 50.840 0.000 1.297 0.577 ## stfeco 1.415 0.025 57.214 0.000 1.659 0.713 ## stfgov 1.480 0.025 58.764 0.000 1.736 0.756 ## stfdem 1.384 0.024 57.904 0.000 1.623 0.731 ## politicians =~ ## pltinvt 1.000 0.646 0.613 ## pltcare 1.021 0.016 62.862 0.000 0.660 0.628 ## trstplt 3.012 0.039 76.838 0.000 1.946 0.891 ## trust =~ ## politicians 1.000 0.918 0.918 ## satisfaction 1.531 0.033 46.494 0.000 0.774 0.774 ## institutions 2.583 0.045 56.796 0.000 0.950 0.950 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## .trstlgl 3.068 0.041 75.262 0.000 3.068 0.541 ## .trstplc 3.248 0.041 80.037 0.000 3.248 0.678 ## .trstun 3.197 0.041 77.141 0.000 3.197 0.588 ## .trstep 2.776 0.036 76.243 0.000 2.776 0.564 ## .trstprl 1.776 0.029 61.361 0.000 1.776 0.345 ## .stfhlth 3.695 0.046 79.989 0.000 3.695 0.729 ## .stfedu 3.368 0.043 77.916 0.000 3.368 0.667 ## .stfeco 2.656 0.038 69.070 0.000 2.656 0.491 ## .stfgov 2.264 0.035 64.201 0.000 2.264 0.429 ## .stfdem 2.289 0.034 67.172 0.000 2.289 0.465 ## .pltinvt 0.694 0.009 78.255 0.000 0.694 0.624 ## .pltcare 0.668 0.009 77.562 0.000 0.668 0.605 ## .trstplt 0.978 0.028 34.461 0.000 0.978 0.205 ## .institutions 0.255 0.022 11.691 0.000 0.098 0.098 ## .satisfaction 0.551 0.020 27.846 0.000 0.400 0.400 ## .politicians 0.065 0.004 17.091 0.000 0.157 0.157 ## trust 0.352 0.010 35.005 0.000 1.000 1.000 ## Compare fit between the first and second order models: fitMeasures(fit_3f) ## npar fmin chisq ## 29.000 0.360 10652.207 ## df pvalue baseline.chisq ## 62.000 0.000 81699.096 ## baseline.df baseline.pvalue cfi ## 78.000 0.000 0.870 ## tli nnfi rfi ## 0.837 0.837 0.836 ## nfi pnfi ifi ## 0.870 0.691 0.870 ## rni logl unrestricted.logl ## 0.870 -371404.658 -366078.555 ## aic bic ntotal ## 742867.317 743087.743 14778.000 ## bic2 rmsea rmsea.ci.lower ## 742995.583 0.108 0.106 ## rmsea.ci.upper rmsea.ci.level rmsea.pvalue ## 0.109 0.900 0.000 ## rmsea.close.h0 rmsea.notclose.pvalue rmsea.notclose.h0 ## 0.050 1.000 0.080 ## rmr rmr_nomean srmr ## 0.255 0.255 0.059 ## srmr_bentler srmr_bentler_nomean crmr ## 0.059 0.059 0.064 ## crmr_nomean srmr_mplus srmr_mplus_nomean ## 0.064 0.059 0.059 ## cn_05 cn_01 gfi ## 113.901 126.971 0.897 ## agfi pgfi mfi ## 0.849 0.611 0.699 ## ecvi ## 0.725 fitMeasures(fit_2nd) ## npar fmin chisq ## 29.000 0.360 10652.207 ## df pvalue baseline.chisq ## 62.000 0.000 81699.096 ## baseline.df baseline.pvalue cfi ## 78.000 0.000 0.870 ## tli nnfi rfi ## 0.837 0.837 0.836 ## nfi pnfi ifi ## 0.870 0.691 0.870 ## rni logl unrestricted.logl ## 0.870 -371404.658 -366078.555 ## aic bic ntotal ## 742867.317 743087.743 14778.000 ## bic2 rmsea rmsea.ci.lower ## 742995.583 0.108 0.106 ## rmsea.ci.upper rmsea.ci.level rmsea.pvalue ## 0.109 0.900 0.000 ## rmsea.close.h0 rmsea.notclose.pvalue rmsea.notclose.h0 ## 0.050 1.000 0.080 ## rmr rmr_nomean srmr ## 0.255 0.255 0.059 ## srmr_bentler srmr_bentler_nomean crmr ## 0.059 0.059 0.064 ## crmr_nomean srmr_mplus srmr_mplus_nomean ## 0.064 0.059 0.059 ## cn_05 cn_01 gfi ## 113.901 126.971 0.897 ## agfi pgfi mfi ## 0.849 0.611 0.699 ## ecvi ## 0.725 Click for explanation We don’t have to do anything special here. We can estimate and summarize the second order CFA exactly as we did the first order CFA. You should quickly notice something strange about the model fit statistics compared above. If you don’t see it, consider the following: fitMeasures(fit_3f) - fitMeasures(fit_2nd) ## npar fmin chisq ## 0 0 0 ## df pvalue baseline.chisq ## 0 0 0 ## baseline.df baseline.pvalue cfi ## 0 0 0 ## tli nnfi rfi ## 0 0 0 ## nfi pnfi ifi ## 0 0 0 ## rni logl unrestricted.logl ## 0 0 0 ## aic bic ntotal ## 0 0 0 ## bic2 rmsea rmsea.ci.lower ## 0 0 0 ## rmsea.ci.upper rmsea.ci.level rmsea.pvalue ## 0 0 0 ## rmsea.close.h0 rmsea.notclose.pvalue rmsea.notclose.h0 ## 0 0 0 ## rmr rmr_nomean srmr ## 0 0 0 ## srmr_bentler srmr_bentler_nomean crmr ## 0 0 0 ## crmr_nomean srmr_mplus srmr_mplus_nomean ## 0 0 0 ## cn_05 cn_01 gfi ## 0 0 0 ## agfi pgfi mfi ## 0 0 0 ## ecvi ## 0 The two models produce identical fit statistics! We also see that the degrees of freedom are identical between the two models. Hence, the two models have equal complexity. This result taps into a critical idea in statistical modeling, namely, model equivalency. It turns out the two models we’re comparing here are equivalent in the sense that they are statistically indistinguishable representations of the data. Since this is a very important idea, I want to spend some time discussing it in person. So, spend some time between now and the Week 6 lecture session thinking about the implications of this model equivalence. Specifically, consider the following questions: What do we mean when we say that these two models are equivalent? How is it possible for these two models to be equivalent when one contains an additional latent variable? Why are the degrees of freedom equal for these two models? Why are the fit statistics equal for these two models? We’ll take some time to discuss these ideas in the Week 6 lecture session. End of At-Home Exercises "],["in-class-exercises-4.html", "5.4 In-Class Exercises", " 5.4 In-Class Exercises This week, we will wrap up our re-analysis of the Kestilä (2006) results. During this practical, you will conduct a CFA of the Attitudes toward Immigration items and compare the results to those obtained from your previous EFA- and PCA-based replications of Kestilä (2006). 5.4.1 Load the ESS data. The relevant data are contained in the ess_round1.rds file. Click to show code ess <- readRDS("ess_round1.rds") We are going to conduct a CFA to evaluate the measurement model implied by the five-factor representation of the Attitudes toward Immigration items that you should have found via the EFA you conducted in the Week 4 In-Class Exercises. Caveat: Technically, the following CFA result have no confirmatory value because we’ll be estimating our CFA models from the same data that we used for our EFA. Practicing the techniques will still be useful, though. 5.4.2 Define the lavaan model syntax for the CFA implied by the five-factor solution from 4.4.9. Enforce a simple structure; do not allow any cross-loadings. Covary the five latent factors. Do not specify any mean structure. Save this model syntax as an object in your environment. Hints: You can algorithmically enforce a simple structure by assigning each item to the factor upon which it loads most strongly. You can download the fitted psych::efa() object for the five-factor solution here. The pattern matrix for the five-factor EFA solution in our Week 4 exercises is equivalent to the solution presented in Table 3 of Kestilä (2006). Click to show code mod_5f <- ' ## Immigration Policy: ip =~ imrcntr + eimrcnt + eimpcnt + imsmetn + impcntr + imdfetn ## Social Threat: st =~ imbgeco + imbleco + imwbcnt + imwbcrm + imtcjob + imueclt ## Refugee Policy: rp =~ gvrfgap + imrsprc + rfgbfml + rfggvfn + rfgawrk + rfgfrpc + shrrfg ## Cultural Threat: ct =~ qfimchr + qfimwht + pplstrd + vrtrlg ## Economic Threat: et =~ imwgdwn + imhecop ' Note: We don’t have to specify the latent covariances in the model syntax, we can tell lavaan to estimate all latent covariances when we fit the model. 5.4.3 Estimate the CFA model you defined above, and summarize the results. Use the lavaan::cfa() function to estimate the model. Use the default settings for the cfa() function. Request the model fit statistics with the summary by supplying the fit.measures = TRUE argument to summary(). Request the standardized parameter estimates with the summary by supplying the standardized = TRUE argument to summary(). Check the results, and answer the following questions: Does the model fit the data well? How are the latent variances and covariances specified when using the default settings? How is the model identified when using the default settings? Click to show code ## Load the lavaan package: library(lavaan) ## Estimate the CFA model: fit_5f <- cfa(mod_5f, data = ess) ## Summarize the fitted model: summary(fit_5f, fit.measures = TRUE, standardized = TRUE) ## lavaan 0.6.16 ended normally after 72 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 60 ## ## Used Total ## Number of observations 14243 19690 ## ## Model Test User Model: ## ## Test statistic 18631.556 ## Degrees of freedom 265 ## P-value (Chi-square) 0.000 ## ## Model Test Baseline Model: ## ## Test statistic 159619.058 ## Degrees of freedom 300 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.885 ## Tucker-Lewis Index (TLI) 0.869 ## ## Loglikelihood and Information Criteria: ## ## Loglikelihood user model (H0) -520035.133 ## Loglikelihood unrestricted model (H1) -510719.354 ## ## Akaike (AIC) 1040190.265 ## Bayesian (BIC) 1040644.106 ## Sample-size adjusted Bayesian (SABIC) 1040453.432 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.070 ## 90 Percent confidence interval - lower 0.069 ## 90 Percent confidence interval - upper 0.071 ## P-value H_0: RMSEA <= 0.050 0.000 ## P-value H_0: RMSEA >= 0.080 0.000 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.048 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## ip =~ ## imrcntr 1.000 0.617 0.748 ## eimrcnt 0.942 0.011 84.943 0.000 0.582 0.696 ## eimpcnt 1.127 0.010 113.413 0.000 0.695 0.898 ## imsmetn 0.982 0.010 98.753 0.000 0.606 0.796 ## impcntr 1.150 0.010 113.623 0.000 0.710 0.900 ## imdfetn 1.132 0.010 111.802 0.000 0.698 0.887 ## st =~ ## imbgeco 1.000 1.608 0.728 ## imbleco 0.826 0.012 69.222 0.000 1.327 0.619 ## imwbcnt 1.046 0.012 88.056 0.000 1.682 0.792 ## imwbcrm 0.713 0.011 63.102 0.000 1.146 0.564 ## imtcjob 0.751 0.011 66.787 0.000 1.207 0.597 ## imueclt 1.008 0.013 78.043 0.000 1.621 0.698 ## rp =~ ## gvrfgap 1.000 0.659 0.610 ## imrsprc 0.855 0.016 51.881 0.000 0.563 0.535 ## rfgbfml 1.047 0.019 56.174 0.000 0.690 0.593 ## rfggvfn 0.849 0.016 51.714 0.000 0.559 0.533 ## rfgawrk 0.653 0.016 41.044 0.000 0.430 0.405 ## rfgfrpc -0.810 0.016 -51.095 0.000 -0.534 -0.525 ## shrrfg -0.999 0.017 -58.381 0.000 -0.658 -0.625 ## ct =~ ## qfimchr 1.000 1.836 0.629 ## qfimwht 0.941 0.017 54.250 0.000 1.728 0.659 ## pplstrd -0.366 0.007 -51.585 0.000 -0.673 -0.600 ## vrtrlg 0.252 0.006 41.294 0.000 0.462 0.443 ## et =~ ## imwgdwn 1.000 0.723 0.667 ## imhecop 1.151 0.023 49.736 0.000 0.832 0.771 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## ip ~~ ## st -0.605 0.012 -48.693 0.000 -0.610 -0.610 ## rp 0.264 0.006 45.566 0.000 0.648 0.648 ## ct 0.634 0.015 41.007 0.000 0.560 0.560 ## et -0.206 0.006 -35.411 0.000 -0.462 -0.462 ## st ~~ ## rp -0.838 0.017 -48.329 0.000 -0.792 -0.792 ## ct -1.622 0.041 -39.091 0.000 -0.550 -0.550 ## et 0.675 0.017 39.083 0.000 0.580 0.580 ## rp ~~ ## ct 0.626 0.018 34.950 0.000 0.518 0.518 ## et -0.233 0.007 -33.007 0.000 -0.490 -0.490 ## ct ~~ ## et -0.592 0.020 -30.127 0.000 -0.446 -0.446 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## .imrcntr 0.299 0.004 77.941 0.000 0.299 0.440 ## .eimrcnt 0.359 0.005 79.638 0.000 0.359 0.515 ## .eimpcnt 0.116 0.002 62.821 0.000 0.116 0.193 ## .imsmetn 0.212 0.003 75.580 0.000 0.212 0.366 ## .impcntr 0.119 0.002 62.454 0.000 0.119 0.191 ## .imdfetn 0.132 0.002 65.344 0.000 0.132 0.213 ## .imbgeco 2.288 0.033 70.261 0.000 2.288 0.470 ## .imbleco 2.837 0.037 76.688 0.000 2.837 0.617 ## .imwbcnt 1.677 0.027 63.198 0.000 1.677 0.372 ## .imwbcrm 2.810 0.036 78.612 0.000 2.810 0.682 ## .imtcjob 2.630 0.034 77.524 0.000 2.630 0.643 ## .imueclt 2.761 0.038 72.515 0.000 2.761 0.512 ## .gvrfgap 0.733 0.010 73.584 0.000 0.733 0.628 ## .imrsprc 0.791 0.010 77.119 0.000 0.791 0.714 ## .rfgbfml 0.877 0.012 74.508 0.000 0.877 0.648 ## .rfggvfn 0.788 0.010 77.203 0.000 0.788 0.716 ## .rfgawrk 0.945 0.012 80.870 0.000 0.945 0.836 ## .rfgfrpc 0.749 0.010 77.501 0.000 0.749 0.724 ## .shrrfg 0.676 0.009 72.682 0.000 0.676 0.609 ## .qfimchr 5.142 0.080 64.113 0.000 5.142 0.604 ## .qfimwht 3.891 0.064 60.623 0.000 3.891 0.566 ## .pplstrd 0.804 0.012 67.054 0.000 0.804 0.640 ## .vrtrlg 0.872 0.011 76.990 0.000 0.872 0.804 ## .imwgdwn 0.652 0.012 53.300 0.000 0.652 0.555 ## .imhecop 0.472 0.014 34.353 0.000 0.472 0.405 ## ip 0.381 0.007 51.578 0.000 1.000 1.000 ## st 2.584 0.054 47.795 0.000 1.000 1.000 ## rp 0.434 0.012 36.748 0.000 1.000 1.000 ## ct 3.371 0.096 35.174 0.000 1.000 1.000 ## et 0.523 0.015 34.944 0.000 1.000 1.000 Click for explanation No, the model does not seem to fit the data well. The SRMR looks good, but one good looking fit statistic is not enough. The TLI and CFI are in the “unacceptable” range. RMSEA is in the “questionable” range. The \\(\\chi^2\\) is highly significant, but we don’t care. The cfa() function is just a wrapper for the lavaan() function with several options set at the defaults you would want for a standard CFA. By default: All latent variances and covariances are freely estimated (due to the argument auto.cov.lv.x = TRUE) The model is identified by fixing the first factor loading of each factor to 1 (due to the argument auto.fix.first = TRUE) To see a full list of the (many) options you can specify to tweak the behavior of lavaan estimation functions run ?lavOptions. Now, we will consider a couple of alternative factor structures for the Attitudes toward Immigration CFA. First, we will go extremely simple by estimating a one-factor model wherein all Attitude items are explained by a single latent variable. 5.4.4 Define the lavaan model syntax for a one-factor model of the Immigration items. Save this syntax as an object in your environment. Click to show code mod_1f <- ' ati =~ imrcntr + eimrcnt + eimpcnt + imsmetn + impcntr + imdfetn + imbgeco + imbleco + imwbcnt + imwbcrm + imtcjob + imueclt + gvrfgap + imrsprc + rfgbfml + rfggvfn + rfgawrk + rfgfrpc + shrrfg + qfimchr + qfimwht + pplstrd + vrtrlg + imwgdwn + imhecop ' 5.4.5 Estimate the one-factor model, and summarize the results. Compare the fit measures for the one-factor and five-factor models Which model better fits the data? Note: Remember, you can use the lavaan::fitMeasures() function to extract only the model fit information from a fitted lavaan object. Click to show code ## Estimate the one factor model: fit_1f <- cfa(mod_1f, data = ess) ## Summarize the results: summary(fit_1f) ## lavaan 0.6.16 ended normally after 47 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 50 ## ## Used Total ## Number of observations 14243 19690 ## ## Model Test User Model: ## ## Test statistic 49510.917 ## Degrees of freedom 275 ## P-value (Chi-square) 0.000 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) ## ati =~ ## imrcntr 1.000 ## eimrcnt 0.937 0.012 78.324 0.000 ## eimpcnt 1.114 0.011 101.263 0.000 ## imsmetn 0.987 0.011 90.990 0.000 ## impcntr 1.147 0.011 102.371 0.000 ## imdfetn 1.153 0.011 103.148 0.000 ## imbgeco -2.055 0.032 -64.749 0.000 ## imbleco -1.625 0.031 -52.533 0.000 ## imwbcnt -2.173 0.030 -71.324 0.000 ## imwbcrm -1.432 0.029 -48.849 0.000 ## imtcjob -1.532 0.029 -52.519 0.000 ## imueclt -2.198 0.033 -65.876 0.000 ## gvrfgap 0.807 0.016 51.746 0.000 ## imrsprc 0.757 0.015 49.790 0.000 ## rfgbfml 0.861 0.017 51.272 0.000 ## rfggvfn 0.722 0.015 47.671 0.000 ## rfgawrk 0.530 0.015 34.448 0.000 ## rfgfrpc -0.755 0.015 -51.462 0.000 ## shrrfg -0.931 0.015 -61.438 0.000 ## qfimchr 1.597 0.042 37.835 0.000 ## qfimwht 1.769 0.038 46.697 0.000 ## pplstrd -0.873 0.016 -53.994 0.000 ## vrtrlg 0.602 0.015 39.940 0.000 ## imwgdwn -0.682 0.016 -43.576 0.000 ## imhecop -0.773 0.016 -49.611 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .imrcntr 0.327 0.004 79.021 0.000 ## .eimrcnt 0.388 0.005 80.422 0.000 ## .eimpcnt 0.161 0.002 70.832 0.000 ## .imsmetn 0.235 0.003 77.101 0.000 ## .impcntr 0.158 0.002 69.688 0.000 ## .imdfetn 0.150 0.002 68.791 0.000 ## .imbgeco 3.381 0.041 82.203 0.000 ## .imbleco 3.666 0.044 83.130 0.000 ## .imwbcnt 2.839 0.035 81.477 0.000 ## .imwbcrm 3.399 0.041 83.334 0.000 ## .imtcjob 3.260 0.039 83.130 0.000 ## .imueclt 3.683 0.045 82.092 0.000 ## .gvrfgap 0.938 0.011 83.176 0.000 ## .imrsprc 0.906 0.011 83.285 0.000 ## .rfgbfml 1.092 0.013 83.203 0.000 ## .rfggvfn 0.917 0.011 83.394 0.000 ## .rfgawrk 1.031 0.012 83.913 0.000 ## .rfgfrpc 0.832 0.010 83.192 0.000 ## .shrrfg 0.803 0.010 82.499 0.000 ## .qfimchr 7.613 0.091 83.803 0.000 ## .qfimwht 5.772 0.069 83.442 0.000 ## .pplstrd 0.988 0.012 83.040 0.000 ## .vrtrlg 0.958 0.011 83.728 0.000 ## .imwgdwn 1.010 0.012 83.583 0.000 ## .imhecop 0.954 0.011 83.294 0.000 ## ati 0.353 0.007 48.941 0.000 ## Compare fit statistics: fitMeasures(fit_5f, fit.measures = c("npar", # Estimated parameters "chisq", "df", "pvalue", # Model fit vs. saturated "cfi", "tli", # Model fit vs. baseline "rmsea", "srmr"), # Model fit vs. saturated output = "text") ## ## Model Test User Model: ## ## Test statistic 18631.556 ## Degrees of freedom 265 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.885 ## Tucker-Lewis Index (TLI) 0.869 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.070 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.048 fitMeasures(fit_1f, fit.measures = c("npar", # Estimated parameters "chisq", "df", "pvalue", # Model fit vs. saturated "cfi", "tli", # Model fit vs. baseline "rmsea", "srmr"), # Model fit vs. saturated output = "text") ## ## Model Test User Model: ## ## Test statistic 49510.917 ## Degrees of freedom 275 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.691 ## Tucker-Lewis Index (TLI) 0.663 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.112 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.087 Click for explanation The one-factor model definitely seems to fit worse than the five-factor model. 5.4.6 Given the CFA results from the five factor model, would a second-order CFA be appropriate for the Attitudes towards Immigration data? Why or why not? Click for explanation Yes, a second-order CFA model is a theoretically appropriate representation of the Attitudes towards Immigration items. The first order latent variables in the five-factor model are all significantly correlated. The first order latent variables in the five-factor model seem to tap different aspects of some single underlying construct. 5.4.7 Define the lavaan model syntax for a second-order CFA model of the Attitudes towards Immigration items, estimate it, and inspect the results. Use the five factors defined in 5.4.2 as the first order factors. Click to show code mod_2o <- paste(mod_5f, 'ati =~ ip + rp + st + ct + et', sep = '\\n') fit_2o <- cfa(mod_2o, data = ess) summary(fit_2o, fit.measures = TRUE) ## lavaan 0.6.16 ended normally after 94 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 55 ## ## Used Total ## Number of observations 14243 19690 ## ## Model Test User Model: ## ## Test statistic 19121.111 ## Degrees of freedom 270 ## P-value (Chi-square) 0.000 ## ## Model Test Baseline Model: ## ## Test statistic 159619.058 ## Degrees of freedom 300 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.882 ## Tucker-Lewis Index (TLI) 0.869 ## ## Loglikelihood and Information Criteria: ## ## Loglikelihood user model (H0) -520279.910 ## Loglikelihood unrestricted model (H1) -510719.354 ## ## Akaike (AIC) 1040669.820 ## Bayesian (BIC) 1041085.841 ## Sample-size adjusted Bayesian (SABIC) 1040911.056 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.070 ## 90 Percent confidence interval - lower 0.069 ## 90 Percent confidence interval - upper 0.071 ## P-value H_0: RMSEA <= 0.050 0.000 ## P-value H_0: RMSEA >= 0.080 0.000 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.048 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) ## ip =~ ## imrcntr 1.000 ## eimrcnt 0.943 0.011 85.095 0.000 ## eimpcnt 1.126 0.010 113.523 0.000 ## imsmetn 0.982 0.010 98.910 0.000 ## impcntr 1.149 0.010 113.651 0.000 ## imdfetn 1.130 0.010 111.789 0.000 ## st =~ ## imbgeco 1.000 ## imbleco 0.822 0.012 68.916 0.000 ## imwbcnt 1.047 0.012 88.172 0.000 ## imwbcrm 0.709 0.011 62.846 0.000 ## imtcjob 0.747 0.011 66.424 0.000 ## imueclt 1.013 0.013 78.434 0.000 ## rp =~ ## gvrfgap 1.000 ## imrsprc 0.854 0.017 51.127 0.000 ## rfgbfml 1.048 0.019 55.377 0.000 ## rfggvfn 0.853 0.017 51.170 0.000 ## rfgawrk 0.657 0.016 40.785 0.000 ## rfgfrpc -0.828 0.016 -51.249 0.000 ## shrrfg -1.020 0.017 -58.369 0.000 ## ct =~ ## qfimchr 1.000 ## qfimwht 0.939 0.018 51.902 0.000 ## pplstrd -0.389 0.008 -51.072 0.000 ## vrtrlg 0.271 0.006 41.908 0.000 ## et =~ ## imwgdwn 1.000 ## imhecop 1.158 0.024 48.877 0.000 ## ati =~ ## ip 1.000 ## rp 1.264 0.024 53.732 0.000 ## st -3.123 0.051 -61.058 0.000 ## ct 2.638 0.058 45.467 0.000 ## et -1.000 0.024 -42.490 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .imrcntr 0.299 0.004 77.900 0.000 ## .eimrcnt 0.359 0.005 79.597 0.000 ## .eimpcnt 0.116 0.002 62.698 0.000 ## .imsmetn 0.211 0.003 75.502 0.000 ## .impcntr 0.119 0.002 62.476 0.000 ## .imdfetn 0.133 0.002 65.406 0.000 ## .imbgeco 2.285 0.033 70.158 0.000 ## .imbleco 2.852 0.037 76.762 0.000 ## .imwbcnt 1.668 0.027 62.920 0.000 ## .imwbcrm 2.821 0.036 78.653 0.000 ## .imtcjob 2.646 0.034 77.607 0.000 ## .imueclt 2.734 0.038 72.213 0.000 ## .gvrfgap 0.740 0.010 73.738 0.000 ## .imrsprc 0.797 0.010 77.211 0.000 ## .rfgbfml 0.885 0.012 74.621 0.000 ## .rfggvfn 0.791 0.010 77.189 0.000 ## .rfgawrk 0.946 0.012 80.833 0.000 ## .rfgfrpc 0.741 0.010 77.149 0.000 ## .shrrfg 0.665 0.009 72.020 0.000 ## .qfimchr 5.347 0.081 65.623 0.000 ## .qfimwht 4.084 0.065 62.673 0.000 ## .pplstrd 0.778 0.012 64.838 0.000 ## .vrtrlg 0.854 0.011 75.931 0.000 ## .imwgdwn 0.655 0.012 52.977 0.000 ## .imhecop 0.468 0.014 33.353 0.000 ## .ip 0.177 0.004 44.418 0.000 ## .st 0.596 0.023 26.030 0.000 ## .rp 0.101 0.005 21.784 0.000 ## .ct 1.745 0.060 29.185 0.000 ## .et 0.316 0.010 31.813 0.000 ## ati 0.204 0.005 37.371 0.000 5.4.8 Compare the model fit of the first- and second-order five-factor models using the fitMeasures() function. Which model offers the better fit? Which model is more complex? Click to show code fitMeasures(fit_5f, fit.measures = c("npar", # Estimated parameters "chisq", "df", "pvalue", # Model fit vs. saturated "cfi", "tli", # Model fit vs. baseline "rmsea", "srmr"), # Model fit vs. saturated output = "text") ## ## Model Test User Model: ## ## Test statistic 18631.556 ## Degrees of freedom 265 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.885 ## Tucker-Lewis Index (TLI) 0.869 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.070 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.048 fitMeasures(fit_2o, fit.measures = c("npar", # Estimated parameters "chisq", "df", "pvalue", # Model fit vs. saturated "cfi", "tli", # Model fit vs. baseline "rmsea", "srmr"), # Model fit vs. saturated output = "text") ## ## Model Test User Model: ## ## Test statistic 19121.111 ## Degrees of freedom 270 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.882 ## Tucker-Lewis Index (TLI) 0.869 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.070 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.048 Click for explanation The CFI and TLI are both slightly better in the original five factor model, but the RMSEA and SRMR of both models don’t differ out to three decimal places. As usual, both models have a significant \\(\\chi^2\\), but that doesn’t tell us much. Qualitative comparisons of model fit are fine, but we’d like to have an actual statistical test for these fit differences. As it happens, we have just such a test: a nested model \\(\\Delta \\chi^2\\) test (AKA, chi-squared difference test, change in chi-squared test, likelihood ratio test). In the Week 7 lecture, we’ll cover nested models and tests thereof, but it will be useful to start thinking about these concepts now. Two models are said to be nested if you can define one model by placing constraints on the other model. By way of example, consider the following two CFA models. The second model is nested within the first model, because we can define the second model by fixing the latent covariance to zero in the first model. Notice that the data contain \\(6(6 + 1) / 2 = 21\\) unique pieces of information. The first model estimates 13 parameters, and the second model estimates 12 parameters. Hence the first model has 8 degrees of freedom, and the second model has 9 degrees of freedom. In general, the following must hold whenever Model B is nested within Model A. Model B will have fewer estimated parameters than Model A. Model B will have more degrees of freedom than Model A. Model A will be more complex than model B. Model A will fit the data better than model B. Saturated Model All models are nested within the saturated model, because the saturated model estimates all possible relations among the variables. Regardless of what model we may be considering, we can always convert that model to a saturated model by estimating all possible associations. Hence, all models are nested within the saturated model. Baseline Model Similarly, the baseline model (AKA, independence model) is nested within all other models. In the baseline model, we only estimate the variances of the observed items; all associations are constrained to zero. We can always convert our model to the baseline model by fixing all associations to zero. Hence, the baseline model is nested within all other models. When two models are nested, we can use a \\(\\Delta \\chi^2\\) test to check if the nested model fits significantly worse than its parent model. Whenever we place constraints on the model, the fit will deteriorate, but we want to know if the constraints we imposed to define the nested model have produced too much loss of fit. We can use the anova() function to easily conduct \\(\\Delta \\chi^2\\) tests comparing models that we’ve estimated with cfa() or sem(). 5.4.9 Use the anova() function to compare the five-factor model from 5.4.2 and one-factor model from 5.4.4. Explain what Df, Chisq, Chisq diff, Df diff, and Pr(>Chisq) mean. Which model is more complex? Which model fits better? What is the conclusion of the test? Click to show code anova(fit_1f, fit_5f) Click for explanation The Df column contains the degrees of freedom of each model. Higher df \\(\\Rightarrow\\) Less complex model The Chisq column shows the \\(\\chi^2\\) statistics (AKA, likelihood ratio statistics) for each model. \\(\\chi^2\\) = The ratio of the likelihoods for the estimated model and the saturated model). Larger \\(\\chi^2\\) \\(\\Rightarrow\\) Worse fit Chisq diff is the difference between the two \\(\\chi^2\\) values (i.e., \\(\\Delta \\chi^2\\)). How much better the more complex model fits the data Larger \\(\\Delta \\chi^2\\) values indicate greater losses of fit induced by the constraints needed to define the nested model. Df diff is the difference in the degrees of freedom between the models. Since both models must be estimated from the same pool of variables, this difference also represents the number of parameters that were constrained to define the nested model. Pr(>Chisq) is a p-value for the \\(\\Delta \\chi^2\\) test. \\(H_0: \\Delta \\chi^2 = 0\\) \\(H_1: \\Delta \\chi^2 > 0\\) The five-factor model is more complex than the one-factor model, but the extra complexity is justified The five-factor model fits significantly better than the one-factor model. 5.4.10 Use the anova() function to compare the first- and second-order five-factor models from 5.4.2 and 5.4.7. Which model is more complex? What is the conclusion of the test? Click to show code anova(fit_5f, fit_2o) Click for explanation The first-order model is more complex than the second-order model (df = 265 vs. df = 270), and the extra complexity is necessary. The first-order model fits significantly better than the second-order model. 5.4.11 Based on the results above, would you say that you have successfully confirmed the five-factor structure implied by the EFA? Click for explanation Nope, not so much. The first-order five-factor model may fit the data best out of the three models considered here, but it still fits terribly. None of these models is an adequate representation of the Attitudes toward Immigration items. This result is particularly embarrassing when you consider that we’ve stacked the deck in our favor by using the same data to conduct the EFA and the CFA. When we fail to support the hypothesized measurement model, the confirmatory phase of our analysis is over. At this point, we’ve essentially rejected our hypothesized measurement structure, and that’s the conclusion of our analysis. We don’t have to throw up our hands in despair, however. We can still contribute something useful by modifying the theoretical measurement model through an exploratory, data-driven, post-hoc analysis. We’ll give that a shot below. 5.4.12 Modify the five-factor CFA from 5.4.2 by freeing the following parameters. The residual covariance between imrcntr and eimrcnt These questions both ask about allowing immigration from wealthy countries. It makes sense that answers on these two items share some additional, unique variance above-and-beyond what they contribute to the common factors. The residual covariance between qfimchr and qfimwht These questions are both about imposing qualifications on immigration (specifically Christian religion and “white” race). Click to show code fit_5f_cov <- paste(mod_5f, 'imrcntr ~~ eimrcnt', 'qfimchr ~~ qfimwht', sep = '\\n') %>% cfa(data = ess) summary(fit_5f_cov, fit.measures = TRUE) ## lavaan 0.6.16 ended normally after 77 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 62 ## ## Used Total ## Number of observations 14243 19690 ## ## Model Test User Model: ## ## Test statistic 9740.512 ## Degrees of freedom 263 ## P-value (Chi-square) 0.000 ## ## Model Test Baseline Model: ## ## Test statistic 159619.058 ## Degrees of freedom 300 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.941 ## Tucker-Lewis Index (TLI) 0.932 ## ## Loglikelihood and Information Criteria: ## ## Loglikelihood user model (H0) -515589.611 ## Loglikelihood unrestricted model (H1) -510719.354 ## ## Akaike (AIC) 1031303.221 ## Bayesian (BIC) 1031772.190 ## Sample-size adjusted Bayesian (SABIC) 1031575.160 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.050 ## 90 Percent confidence interval - lower 0.049 ## 90 Percent confidence interval - upper 0.051 ## P-value H_0: RMSEA <= 0.050 0.280 ## P-value H_0: RMSEA >= 0.080 0.000 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.036 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) ## ip =~ ## imrcntr 1.000 ## eimrcnt 0.928 0.007 126.255 0.000 ## eimpcnt 1.184 0.011 106.508 0.000 ## imsmetn 1.012 0.011 92.436 0.000 ## impcntr 1.213 0.011 107.078 0.000 ## imdfetn 1.181 0.011 104.566 0.000 ## st =~ ## imbgeco 1.000 ## imbleco 0.826 0.012 69.006 0.000 ## imwbcnt 1.050 0.012 88.051 0.000 ## imwbcrm 0.715 0.011 63.128 0.000 ## imtcjob 0.751 0.011 66.542 0.000 ## imueclt 1.015 0.013 78.256 0.000 ## rp =~ ## gvrfgap 1.000 ## imrsprc 0.858 0.017 51.965 0.000 ## rfgbfml 1.046 0.019 56.104 0.000 ## rfggvfn 0.848 0.016 51.644 0.000 ## rfgawrk 0.652 0.016 40.998 0.000 ## rfgfrpc -0.813 0.016 -51.233 0.000 ## shrrfg -1.002 0.017 -58.499 0.000 ## ct =~ ## qfimchr 1.000 ## qfimwht 0.979 0.020 48.332 0.000 ## pplstrd -0.586 0.014 -40.685 0.000 ## vrtrlg 0.397 0.011 36.273 0.000 ## et =~ ## imwgdwn 1.000 ## imhecop 1.157 0.023 49.549 0.000 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ## .imrcntr ~~ ## .eimrcnt 0.230 0.004 59.907 0.000 ## .qfimchr ~~ ## .qfimwht 2.558 0.064 40.233 0.000 ## ip ~~ ## st -0.580 0.012 -48.041 0.000 ## rp 0.255 0.006 45.185 0.000 ## ct 0.467 0.014 34.425 0.000 ## et -0.197 0.006 -35.077 0.000 ## st ~~ ## rp -0.835 0.017 -48.285 0.000 ## ct -1.394 0.040 -35.128 0.000 ## et 0.670 0.017 38.935 0.000 ## rp ~~ ## ct 0.538 0.017 32.407 0.000 ## et -0.232 0.007 -32.949 0.000 ## ct ~~ ## et -0.469 0.017 -27.959 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .imrcntr 0.330 0.004 78.903 0.000 ## .eimrcnt 0.396 0.005 80.392 0.000 ## .eimpcnt 0.109 0.002 60.401 0.000 ## .imsmetn 0.220 0.003 75.979 0.000 ## .impcntr 0.107 0.002 58.874 0.000 ## .imdfetn 0.131 0.002 64.630 0.000 ## .imbgeco 2.301 0.033 70.568 0.000 ## .imbleco 2.845 0.037 76.832 0.000 ## .imwbcnt 1.669 0.026 63.272 0.000 ## .imwbcrm 2.808 0.036 78.659 0.000 ## .imtcjob 2.639 0.034 77.663 0.000 ## .imueclt 2.741 0.038 72.463 0.000 ## .gvrfgap 0.734 0.010 73.743 0.000 ## .imrsprc 0.790 0.010 77.164 0.000 ## .rfgbfml 0.880 0.012 74.676 0.000 ## .rfggvfn 0.790 0.010 77.322 0.000 ## .rfgawrk 0.946 0.012 80.924 0.000 ## .rfgfrpc 0.747 0.010 77.519 0.000 ## .shrrfg 0.674 0.009 72.713 0.000 ## .qfimchr 6.815 0.090 75.362 0.000 ## .qfimwht 5.250 0.072 73.378 0.000 ## .pplstrd 0.674 0.013 52.766 0.000 ## .vrtrlg 0.818 0.011 73.191 0.000 ## .imwgdwn 0.655 0.012 53.496 0.000 ## .imhecop 0.468 0.014 33.845 0.000 ## ip 0.350 0.007 48.646 0.000 ## st 2.571 0.054 47.662 0.000 ## rp 0.433 0.012 36.718 0.000 ## ct 1.698 0.073 23.296 0.000 ## et 0.520 0.015 34.814 0.000 5.4.13 Evaluate the model modifications. Did the model fit significantly improve? Is the fit of the modified model acceptable? Click to show code anova(fit_5f_cov, fit_5f) fitMeasures(fit_5f_cov) ## npar fmin chisq ## 62.000 0.342 9740.512 ## df pvalue baseline.chisq ## 263.000 0.000 159619.058 ## baseline.df baseline.pvalue cfi ## 300.000 0.000 0.941 ## tli nnfi rfi ## 0.932 0.932 0.930 ## nfi pnfi ifi ## 0.939 0.823 0.941 ## rni logl unrestricted.logl ## 0.941 -515589.611 -510719.354 ## aic bic ntotal ## 1031303.221 1031772.190 14243.000 ## bic2 rmsea rmsea.ci.lower ## 1031575.160 0.050 0.049 ## rmsea.ci.upper rmsea.ci.level rmsea.pvalue ## 0.051 0.900 0.280 ## rmsea.close.h0 rmsea.notclose.pvalue rmsea.notclose.h0 ## 0.050 0.000 0.080 ## rmr rmr_nomean srmr ## 0.103 0.103 0.036 ## srmr_bentler srmr_bentler_nomean crmr ## 0.036 0.036 0.037 ## crmr_nomean srmr_mplus srmr_mplus_nomean ## 0.037 0.036 0.036 ## cn_05 cn_01 gfi ## 442.344 467.858 0.944 ## agfi pgfi mfi ## 0.931 0.764 0.717 ## ecvi ## 0.693 Click for explanation Yes, the model fit improved significantly. In this case, the original five-factor model is nested within the modified model. So, our \\(\\Delta \\chi^2\\) test is evaluating the improvement in fit contributed by freeing the two residual covariances. The \\(\\Delta \\chi^2\\) test is significant, so we can conclude that including the two new parameter estimates has significantly improved the model fit. I.e., Estimating these two residual covariances is “worth it” in the sense of balancing model fit and model complexity. Also, the fit of the modified model is now acceptable. Caveat If we had found this result when testing our original model, we would be well-situated to proceed with our analysis. In this case, however, we are no longer justified in generalizing these estimates to the population. We only arrived at this well-fitting model by modifying our original theoretical model to better fit the data using estimates derived from those same data to guide our model modifications. We’ve conducted this post-hoc analysis to help inform future research, and this result is useful as a starting point for future studies. Now, anyone analyzing these scales in the future could incorporate these residual covariances into their initial theoretical model. Basically, we conduct these types of post-hoc analyses to help future researchers learn from our mistakes. End of In-Class Exercises "],["full-sem.html", "6 Full SEM", " 6 Full SEM This week, we will focus on integrating all of the disparate methods we’ve covered so far into full-fledged structural equation models. Homework before the lecture Watch the Lecture Recording for this week. Complete the Reading for this week, and answer the associated reading questions. Homework before the practical Complete the At-Home Exercises. Practical content During the practical you will work on the In-Class Exercises. "],["lecture-5.html", "6.1 Lecture", " 6.1 Lecture This week, we will begin with our final theme and discuss structural equation modeling (SEM). This powerful technique joins the strengths of CFA and path analysis to produce a highly flexible and theoretically appealing modeling tool. Essentially, SEM allows us to build structural path models using the latent variables defined by a CFA. 6.1.1 Recording 6.1.2 Slides You can download the lectures slides here "],["reading-5.html", "6.2 Reading", " 6.2 Reading Reference Weston, R. & Gore, P. A. (2006). A brief guide to structural equation modeling. The Counseling Psychologist 34, 719–752. Notes: This article is quite general and provides an overview of things we have discussed so far in this course. This article also also adds an important new idea: combining factor analysis with path modeling to produce a full Structural Equation Model (SEM). Skip the part on GFI (p. 741). The GFI has been shown to be too dependent on sample size and is not recommended any longer. Skip the part on missing data. There is nothing wrong with this section, but missing data analysis is a broad and difficult topic that we cannot adequately cover in this course. If you would like to learn more about missing data and how to treat them, you can take two courses offered by our department: Conducting a Survey Missing Data Theory and Causal Effects Questions The authors state three similarities and two big differences between SEM and other multivariate statistical techniques (e.g., ANCOVA, regression). What are these similarities and differences? Do you agree with the relative strengths and weaknesses of SEM vs. other methods that the authors present? The authors miss at least one additional advantage of SEM over other multivariate methods. What is this missing advantage? Explain what the terms “measurement model” and “structural model” mean in the SEM context. What are the 6 steps of doing an SEM-based analysis given by the authors? The authors claim that testing an SEM using cross-validation is a good idea. When is cross-validation helpful in SEM? Hint: You may have to do some independent (internet, literature) research to learn how cross-validation can be implemented in SEM. "],["at-home-exercises-5.html", "6.3 At-Home Exercises", " 6.3 At-Home Exercises This week, we’ll take another look at the Kestilä (2006) results. During this practical, you will conduct an SEM to replicate the regression analysis of the Finnish data that you conducted in the Week 4 In-Class Exercises. 6.3.1 Load the Finnish subsample of ESS data. The relevant data are contained in the ess_finland.rds file. These are the processed Finnish subsample data from the Week 4 exercises. Note: Unless otherwise noted, all the following analyses use these data. Click to show code ess_fin <- readRDS("ess_finland.rds") We need to do a little data processing before we can fit the regression model. At the moment, lavaan will not automatically convert a factor variable into dummy codes. So, we need to create explicit dummy codes for the two factors we’ll use as predictors in our regression analysis: sex and political orientation. 6.3.2 Convert the sex and political interest factors into dummy codes. Click to show code library(dplyr) ## Create a dummy codes by broadcasting a logical test on the factor levels: ess_fin <- mutate(ess_fin, female = ifelse(sex == "Female", 1, 0), hi_pol_interest = ifelse(polintr_bin == "High Interest", 1, 0) ) ## Check the results: with(ess_fin, table(dummy = female, factor = sex)) ## factor ## dummy Male Female ## 0 960 0 ## 1 0 1040 with(ess_fin, table(dummy = hi_pol_interest, factor = polintr_bin)) ## factor ## dummy Low Interest High Interest ## 0 1070 0 ## 1 0 929 Click for explanation In R, we have several ways of converting a factor into an appropriate set of dummy codes. We could use the dplyr::recode() function as we did last week. We can use the model.matrix() function to define a design matrix based on the inherent contrast attribute of the factor. Missing data will cause problems here. We can us as.numeric() to revert the factor to its underlying numeric representation {Male = 1, Female = 2} and use arithmetic to convert {1, 2} \\(\\rightarrow\\) {0, 1}. When our factor only has two levels, though, the ifelse() function is the simplest way. We are now ready to estimate our latent regression model. Specifically, we want to combine the three OLS regression models that you ran in 4.4.16 into a single SEM that we will estimate in lavaan. The following path diagram shows the intended theoretical model. Although the variances are not included in this path diagram, all variables in the model (including the observed predictor variables) are random. 6.3.3 Define the lavaan model syntax for the SEM shown above. Use the definition of the institutions, satsifaction, and politicians factors from 5.3.2 to define the DVs. Covary the three latent factors. Covary the five predictors. Click to show code mod_sem <- ' ## Define the latent DVs: institutions =~ trstlgl + trstplc + trstun + trstep + trstprl satisfaction =~ stfhlth + stfedu + stfeco + stfgov + stfdem politicians =~ pltinvt + pltcare + trstplt ## Specify the structural relations: institutions + satisfaction + politicians ~ female + age + eduyrs + hi_pol_interest + lrscale ' Click for explanation We simply need to add a line defining the latent regression paths to our old CFA syntax. We don’t need to specify the covariances in the syntax. We can use options in the sem() function to request those estimates. 6.3.4 Estimate the SEM, and summarize the results. Fit the model to the processed Finnish subsample from above. Estimate the model using lavaan::sem(). Request the standardized parameter estimates with the summary. Request the \\(R^2\\) estimates with the summary. Click to show code library(lavaan) ## Fit the SEM: fit_sem <- sem(mod_sem, data = ess_fin, fixed.x = FALSE) ## Summarize the results: summary(fit_sem, fit.measures = TRUE, standardized = TRUE, rsquare = TRUE) ## lavaan 0.6.16 ended normally after 82 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 59 ## ## Used Total ## Number of observations 1740 2000 ## ## Model Test User Model: ## ## Test statistic 1287.421 ## Degrees of freedom 112 ## P-value (Chi-square) 0.000 ## ## Model Test Baseline Model: ## ## Test statistic 10534.649 ## Degrees of freedom 143 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.887 ## Tucker-Lewis Index (TLI) 0.856 ## ## Loglikelihood and Information Criteria: ## ## Loglikelihood user model (H0) -57914.779 ## Loglikelihood unrestricted model (H1) -57271.068 ## ## Akaike (AIC) 115947.557 ## Bayesian (BIC) 116269.794 ## Sample-size adjusted Bayesian (SABIC) 116082.357 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.078 ## 90 Percent confidence interval - lower 0.074 ## 90 Percent confidence interval - upper 0.082 ## P-value H_0: RMSEA <= 0.050 0.000 ## P-value H_0: RMSEA >= 0.080 0.160 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.045 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## institutions =~ ## trstlgl 1.000 1.418 0.669 ## trstplc 0.609 0.031 19.403 0.000 0.863 0.508 ## trstun 0.887 0.038 23.484 0.000 1.257 0.626 ## trstep 1.134 0.041 27.652 0.000 1.607 0.755 ## trstprl 1.192 0.040 29.444 0.000 1.689 0.815 ## satisfaction =~ ## stfhlth 1.000 0.979 0.497 ## stfedu 0.602 0.043 13.872 0.000 0.589 0.416 ## stfeco 1.266 0.067 18.848 0.000 1.240 0.681 ## stfgov 1.639 0.079 20.638 0.000 1.605 0.846 ## stfdem 1.521 0.075 20.180 0.000 1.489 0.793 ## politicians =~ ## pltinvt 1.000 0.567 0.566 ## pltcare 0.953 0.048 19.653 0.000 0.540 0.590 ## trstplt 3.281 0.133 24.675 0.000 1.860 0.915 ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## institutions ~ ## female 0.019 0.073 0.259 0.796 0.013 0.007 ## age -0.008 0.002 -3.740 0.000 -0.006 -0.105 ## eduyrs 0.034 0.010 3.233 0.001 0.024 0.091 ## hi_pol_interst 0.358 0.076 4.730 0.000 0.253 0.126 ## lrscale 0.104 0.018 5.634 0.000 0.073 0.147 ## satisfaction ~ ## female -0.147 0.050 -2.910 0.004 -0.150 -0.075 ## age -0.007 0.002 -4.598 0.000 -0.007 -0.129 ## eduyrs 0.005 0.007 0.775 0.439 0.006 0.022 ## hi_pol_interst 0.164 0.052 3.162 0.002 0.167 0.084 ## lrscale 0.099 0.013 7.501 0.000 0.101 0.202 ## politicians ~ ## female 0.010 0.029 0.349 0.727 0.018 0.009 ## age -0.004 0.001 -4.490 0.000 -0.007 -0.124 ## eduyrs 0.007 0.004 1.697 0.090 0.012 0.047 ## hi_pol_interst 0.258 0.031 8.364 0.000 0.455 0.227 ## lrscale 0.039 0.007 5.370 0.000 0.068 0.138 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## .institutions ~~ ## .satisfaction 1.030 0.069 14.933 0.000 0.796 0.796 ## .politicians 0.675 0.041 16.628 0.000 0.908 0.908 ## .satisfaction ~~ ## .politicians 0.365 0.027 13.544 0.000 0.713 0.713 ## female ~~ ## age 0.071 0.212 0.335 0.738 0.071 0.008 ## eduyrs 0.179 0.046 3.869 0.000 0.179 0.093 ## hi_pol_interst -0.017 0.006 -2.767 0.006 -0.017 -0.066 ## lrscale -0.032 0.024 -1.316 0.188 -0.032 -0.032 ## age ~~ ## eduyrs -22.750 1.722 -13.212 0.000 -22.750 -0.334 ## hi_pol_interst 1.377 0.215 6.413 0.000 1.377 0.156 ## lrscale 1.774 0.853 2.079 0.038 1.774 0.050 ## eduyrs ~~ ## hi_pol_interst 0.270 0.047 5.787 0.000 0.270 0.140 ## lrscale 0.735 0.186 3.946 0.000 0.735 0.095 ## hi_pol_interest ~~ ## lrscale 0.016 0.024 0.672 0.501 0.016 0.016 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## .trstlgl 2.477 0.093 26.743 0.000 2.477 0.552 ## .trstplc 2.140 0.076 28.334 0.000 2.140 0.742 ## .trstun 2.453 0.090 27.322 0.000 2.453 0.608 ## .trstep 1.950 0.078 24.906 0.000 1.950 0.430 ## .trstprl 1.443 0.064 22.437 0.000 1.443 0.336 ## .stfhlth 2.922 0.104 28.103 0.000 2.922 0.753 ## .stfedu 1.663 0.058 28.613 0.000 1.663 0.827 ## .stfeco 1.775 0.069 25.755 0.000 1.775 0.536 ## .stfgov 1.020 0.056 18.371 0.000 1.020 0.284 ## .stfdem 1.307 0.060 21.953 0.000 1.307 0.371 ## .pltinvt 0.682 0.024 27.818 0.000 0.682 0.680 ## .pltcare 0.547 0.020 27.582 0.000 0.547 0.652 ## .trstplt 0.672 0.069 9.676 0.000 0.672 0.163 ## .institutions 1.881 0.125 15.077 0.000 0.936 0.936 ## .satisfaction 0.892 0.086 10.386 0.000 0.930 0.930 ## .politicians 0.294 0.024 12.224 0.000 0.914 0.914 ## female 0.250 0.008 29.496 0.000 0.250 1.000 ## age 313.238 10.620 29.496 0.000 313.238 1.000 ## eduyrs 14.818 0.502 29.496 0.000 14.818 1.000 ## hi_pol_interst 0.250 0.008 29.496 0.000 0.250 1.000 ## lrscale 4.034 0.137 29.496 0.000 4.034 1.000 ## ## R-Square: ## Estimate ## trstlgl 0.448 ## trstplc 0.258 ## trstun 0.392 ## trstep 0.570 ## trstprl 0.664 ## stfhlth 0.247 ## stfedu 0.173 ## stfeco 0.464 ## stfgov 0.716 ## stfdem 0.629 ## pltinvt 0.320 ## pltcare 0.348 ## trstplt 0.837 ## institutions 0.064 ## satisfaction 0.070 ## politicians 0.086 Click for explanation The fixed.x = FALSE argument tells lavaan to model the predictors as random variables. By default, lavaan will covary any random predictor variables. So, we don’t need to make any other changes to the usual procedure. 6.3.5 Finally, we will rerun the latent regression model from above as a path model with the factor scores from 4.4.10 acting as the DVs. Rerun the above SEM as a path model wherein the EFA-derived Trust in Institutions, Satisfaction with Political Systems, and Trust in Politicians factor scores act as the DVs. Request the standardized parameter estimates with the summary. Request the \\(R^2\\) estimates with the summary. Click to show code ## Define the model syntax for the path analysis: mod_pa <- ' trust_inst + satisfy + trust_pol ~ female + age + eduyrs + hi_pol_interest + lrscale' ## Estimate the path model: fit_pa <- sem(mod_pa, data = ess_fin, fixed.x = FALSE) ## Summarize the results: summary(fit_pa, standardized = TRUE, rsquare = TRUE) ## lavaan 0.6.16 ended normally after 44 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 36 ## ## Used Total ## Number of observations 1740 2000 ## ## Model Test User Model: ## ## Test statistic 0.000 ## Degrees of freedom 0 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## trust_inst ~ ## female 0.004 0.045 0.091 0.928 0.004 0.002 ## age -0.003 0.001 -2.229 0.026 -0.003 -0.057 ## eduyrs 0.023 0.006 3.642 0.000 0.023 0.094 ## hi_pol_interst 0.167 0.046 3.599 0.000 0.167 0.088 ## lrscale 0.059 0.011 5.258 0.000 0.059 0.125 ## satisfy ~ ## female -0.125 0.040 -3.115 0.002 -0.125 -0.073 ## age -0.005 0.001 -4.102 0.000 -0.005 -0.105 ## eduyrs -0.003 0.006 -0.534 0.594 -0.003 -0.014 ## hi_pol_interst 0.073 0.041 1.782 0.075 0.073 0.043 ## lrscale 0.085 0.010 8.510 0.000 0.085 0.200 ## trust_pol ~ ## female 0.016 0.046 0.338 0.735 0.016 0.008 ## age -0.009 0.001 -6.480 0.000 -0.009 -0.161 ## eduyrs 0.018 0.007 2.839 0.005 0.018 0.071 ## hi_pol_interst 0.464 0.047 9.801 0.000 0.464 0.232 ## lrscale 0.055 0.011 4.801 0.000 0.055 0.110 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## .trust_inst ~~ ## .satisfy 0.437 0.021 20.609 0.000 0.437 0.568 ## .trust_pol 0.498 0.024 20.480 0.000 0.498 0.564 ## .satisfy ~~ ## .trust_pol 0.367 0.021 17.664 0.000 0.367 0.467 ## female ~~ ## age 0.071 0.212 0.335 0.738 0.071 0.008 ## eduyrs 0.179 0.046 3.869 0.000 0.179 0.093 ## hi_pol_interst -0.017 0.006 -2.767 0.006 -0.017 -0.066 ## lrscale -0.032 0.024 -1.316 0.188 -0.032 -0.032 ## age ~~ ## eduyrs -22.750 1.722 -13.212 0.000 -22.750 -0.334 ## hi_pol_interst 1.377 0.215 6.413 0.000 1.377 0.156 ## lrscale 1.774 0.853 2.079 0.038 1.774 0.050 ## eduyrs ~~ ## hi_pol_interst 0.270 0.047 5.787 0.000 0.270 0.140 ## lrscale 0.735 0.186 3.946 0.000 0.735 0.095 ## hi_pol_interest ~~ ## lrscale 0.016 0.024 0.672 0.501 0.016 0.016 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## .trust_inst 0.866 0.029 29.496 0.000 0.866 0.958 ## .satisfy 0.684 0.023 29.496 0.000 0.684 0.945 ## .trust_pol 0.902 0.031 29.496 0.000 0.902 0.902 ## female 0.250 0.008 29.496 0.000 0.250 1.000 ## age 313.238 10.620 29.496 0.000 313.238 1.000 ## eduyrs 14.818 0.502 29.496 0.000 14.818 1.000 ## hi_pol_interst 0.250 0.008 29.496 0.000 0.250 1.000 ## lrscale 4.034 0.137 29.496 0.000 4.034 1.000 ## ## R-Square: ## Estimate ## trust_inst 0.042 ## satisfy 0.055 ## trust_pol 0.098 Click to show explanation We don’t so anything particularly special here. We simply rerun our latent regression as a path analysis with the EFA-derived factor scores as the DVs. 6.3.6 Compare the results from the path analysis to the SEM-based results. Does it matter whether we use a latent variable or a factor score to define the DV? Hint: When comparing parameter estimates, use the fully standardized estimates (i.e., the values in the column labeled Std.all). Click to show code Note: The “supportFunction.R” script that we source below isn’t a necessary part of the solution. This script defines a bunch of convenience functions. One of these functions, partSummary(), allows us to print selected pieces of the model summary. ## Source a script of convenience function definitions: source("supportFunctions.R") ## View the regression estimates from the SEM: partSummary(fit_sem, 8, standardized = TRUE) ## Regressions: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## institutions ~ ## female 0.019 0.073 0.259 0.796 0.013 0.007 ## age -0.008 0.002 -3.740 0.000 -0.006 -0.105 ## eduyrs 0.034 0.010 3.233 0.001 0.024 0.091 ## hi_pol_interst 0.358 0.076 4.730 0.000 0.253 0.126 ## lrscale 0.104 0.018 5.634 0.000 0.073 0.147 ## satisfaction ~ ## female -0.147 0.050 -2.910 0.004 -0.150 -0.075 ## age -0.007 0.002 -4.598 0.000 -0.007 -0.129 ## eduyrs 0.005 0.007 0.775 0.439 0.006 0.022 ## hi_pol_interst 0.164 0.052 3.162 0.002 0.167 0.084 ## lrscale 0.099 0.013 7.501 0.000 0.101 0.202 ## politicians ~ ## female 0.010 0.029 0.349 0.727 0.018 0.009 ## age -0.004 0.001 -4.490 0.000 -0.007 -0.124 ## eduyrs 0.007 0.004 1.697 0.090 0.012 0.047 ## hi_pol_interst 0.258 0.031 8.364 0.000 0.455 0.227 ## lrscale 0.039 0.007 5.370 0.000 0.068 0.138 ## View the regression estimates from the path analysis: partSummary(fit_pa, 7, standardized = TRUE) ## Regressions: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## trust_inst ~ ## female 0.004 0.045 0.091 0.928 0.004 0.002 ## age -0.003 0.001 -2.229 0.026 -0.003 -0.057 ## eduyrs 0.023 0.006 3.642 0.000 0.023 0.094 ## hi_pol_interst 0.167 0.046 3.599 0.000 0.167 0.088 ## lrscale 0.059 0.011 5.258 0.000 0.059 0.125 ## satisfy ~ ## female -0.125 0.040 -3.115 0.002 -0.125 -0.073 ## age -0.005 0.001 -4.102 0.000 -0.005 -0.105 ## eduyrs -0.003 0.006 -0.534 0.594 -0.003 -0.014 ## hi_pol_interst 0.073 0.041 1.782 0.075 0.073 0.043 ## lrscale 0.085 0.010 8.510 0.000 0.085 0.200 ## trust_pol ~ ## female 0.016 0.046 0.338 0.735 0.016 0.008 ## age -0.009 0.001 -6.480 0.000 -0.009 -0.161 ## eduyrs 0.018 0.007 2.839 0.005 0.018 0.071 ## hi_pol_interst 0.464 0.047 9.801 0.000 0.464 0.232 ## lrscale 0.055 0.011 4.801 0.000 0.055 0.110 ## View the R-squared estimates from the SEM: partSummary(fit_sem, 11, rsquare = TRUE) ## R-Square: ## Estimate ## trstlgl 0.448 ## trstplc 0.258 ## trstun 0.392 ## trstep 0.570 ## trstprl 0.664 ## stfhlth 0.247 ## stfedu 0.173 ## stfeco 0.464 ## stfgov 0.716 ## stfdem 0.629 ## pltinvt 0.320 ## pltcare 0.348 ## trstplt 0.837 ## institutions 0.064 ## satisfaction 0.070 ## politicians 0.086 ## View the R-squared estimates from the SEM: partSummary(fit_pa, 10, rsquare = TRUE) ## R-Square: ## Estimate ## trust_inst 0.042 ## satisfy 0.055 ## trust_pol 0.098 Click for explanation It certainly looks like the way we define the DV has a meaningful impact. The patterns of significance differ between the two sets of regression slopes, and the \\(R^2\\) values are larger for the Institutions and Satisfaction factors in the SEM, and the \\(R^2\\) for the Politicians factor is higher in the path analysis. End of At-Home Exercises "],["in-class-exercises-5.html", "6.4 In-Class Exercises", " 6.4 In-Class Exercises In these exercises, you will use full structural equation modeling (SEM) to evaluate the Theory of Reasoned Action (TORA), which is a popular psychological theory of social behavior developed by Ajzen and Fishbein. The theory states that actual behavior is predicted by behavioral intention, which is in turn predicted by the attitude toward the behavior and subjective norms about the behavior. Later, a third determinant was added, perceived behavioral control. The extent to which people feel that they have control over their behavior also influences their behavior. The data we will use for this practical are available in the toradata.csv file. These data were synthesized according to the results of Reinecke (1998)’s investigation of condom use by young people between 16 and 24 years old. The data contain the following variables: respnr: Numeric participant ID behavior: The dependent variable condom use Measured on a 5-point frequency scale (How often do you…) intent: A single item assessing behavioral intention Measured on a similar 5-point scale (In general, do you intend to…). attit_1:attit_3: Three indicators of attitudes about condom use Measured on a 5-point rating scale (e.g., using a condom is awkward) norm_1:norm_3: Three indicators of social norms about condom use Measured on a 5-point rating scale (e.g., I think most of my friends would use…) control_1:control_3: Three indicators of perceived behavioral control Measured on a 5-point rating scale (e.g., I know well how to use a condom) sex: Binary factor indicating biological sex 6.4.1 Load the data contained in the toradata.csv file. Click for explanation condom <- read.csv("toradata.csv", stringsAsFactors = TRUE) 6.4.2 The data contain multiple indicators of attitudes, norms, and control. Run a CFA for these three latent variables. Correlate the latent factors. Do the data support the measurement model for these latent factors? Are the three latent factors significantly correlated? Is it reasonable to proceed with our evaluation of the TORA theory? Click for explanation library(lavaan) mod_cfa <- ' attitudes =~ attit_1 + attit_2 + attit_3 norms =~ norm_1 + norm_2 + norm_3 control =~ control_1 + control_2 + control_3 ' fit <- cfa(mod_cfa, data = condom) summary(fit, fit.measures = TRUE) ## lavaan 0.6.16 ended normally after 29 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 21 ## ## Number of observations 250 ## ## Model Test User Model: ## ## Test statistic 35.611 ## Degrees of freedom 24 ## P-value (Chi-square) 0.060 ## ## Model Test Baseline Model: ## ## Test statistic 910.621 ## Degrees of freedom 36 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.987 ## Tucker-Lewis Index (TLI) 0.980 ## ## Loglikelihood and Information Criteria: ## ## Loglikelihood user model (H0) -2998.290 ## Loglikelihood unrestricted model (H1) -2980.484 ## ## Akaike (AIC) 6038.580 ## Bayesian (BIC) 6112.530 ## Sample-size adjusted Bayesian (SABIC) 6045.959 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.044 ## 90 Percent confidence interval - lower 0.000 ## 90 Percent confidence interval - upper 0.073 ## P-value H_0: RMSEA <= 0.050 0.599 ## P-value H_0: RMSEA >= 0.080 0.017 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.037 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) ## attitudes =~ ## attit_1 1.000 ## attit_2 1.036 0.068 15.308 0.000 ## attit_3 -1.002 0.067 -14.856 0.000 ## norms =~ ## norm_1 1.000 ## norm_2 1.031 0.098 10.574 0.000 ## norm_3 0.932 0.093 10.013 0.000 ## control =~ ## control_1 1.000 ## control_2 0.862 0.129 6.699 0.000 ## control_3 0.968 0.133 7.290 0.000 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ## attitudes ~~ ## norms 0.340 0.069 4.957 0.000 ## control 0.475 0.073 6.468 0.000 ## norms ~~ ## control 0.338 0.064 5.254 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .attit_1 0.418 0.052 8.047 0.000 ## .attit_2 0.310 0.047 6.633 0.000 ## .attit_3 0.369 0.049 7.577 0.000 ## .norm_1 0.504 0.071 7.130 0.000 ## .norm_2 0.469 0.071 6.591 0.000 ## .norm_3 0.635 0.075 8.465 0.000 ## .control_1 0.614 0.078 7.905 0.000 ## .control_2 0.865 0.091 9.520 0.000 ## .control_3 0.762 0.087 8.758 0.000 ## attitudes 0.885 0.116 7.620 0.000 ## norms 0.743 0.116 6.423 0.000 ## control 0.497 0.099 5.002 0.000 Yes, the model fits the data well, and the measurement parameters (e.g., factor loadings, residual variances) look reasonable. So, the data seem to support this measurement structure. Yes, all three latent variables are significantly, positively correlated. Yes. The measurement structure is supported, so we can use the latent variables to represent the respective constructs in our subsequent SEM. The TORA doesn’t actually say anything about the associations between these three factors, but it makes sense that they would be positively associated. So, we should find this result comforting. 6.4.3 Estimate the basic TORA model as an SEM. Predict intention from attitudes and norms. Predict condom use from intention. Use the latent versions of attitudes and norms. Covary the attitudes and norms factors. Does the model fit well? Do the estimates align with the TORA? How much variance in intention and condom use are explained by the model? Click for explanation mod <- ' ## Define the latent variables: attitudes =~ attit_1 + attit_2 + attit_3 norms =~ norm_1 + norm_2 + norm_3 ## Define the structural model: intent ~ attitudes + norms behavior ~ intent ' fit <- sem(mod, data = condom) summary(fit, fit.measures = TRUE, rsquare = TRUE) ## lavaan 0.6.16 ended normally after 24 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 18 ## ## Number of observations 250 ## ## Model Test User Model: ## ## Test statistic 27.890 ## Degrees of freedom 18 ## P-value (Chi-square) 0.064 ## ## Model Test Baseline Model: ## ## Test statistic 1089.407 ## Degrees of freedom 28 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.991 ## Tucker-Lewis Index (TLI) 0.986 ## ## Loglikelihood and Information Criteria: ## ## Loglikelihood user model (H0) -2533.616 ## Loglikelihood unrestricted model (H1) -2519.671 ## ## Akaike (AIC) 5103.232 ## Bayesian (BIC) 5166.618 ## Sample-size adjusted Bayesian (SABIC) 5109.557 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.047 ## 90 Percent confidence interval - lower 0.000 ## 90 Percent confidence interval - upper 0.079 ## P-value H_0: RMSEA <= 0.050 0.523 ## P-value H_0: RMSEA >= 0.080 0.046 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.036 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) ## attitudes =~ ## attit_1 1.000 ## attit_2 1.039 0.068 15.365 0.000 ## attit_3 -1.002 0.067 -14.850 0.000 ## norms =~ ## norm_1 1.000 ## norm_2 0.983 0.087 11.333 0.000 ## norm_3 0.935 0.087 10.778 0.000 ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## intent ~ ## attitudes 0.439 0.063 6.990 0.000 ## norms 0.693 0.077 8.977 0.000 ## behavior ~ ## intent 0.746 0.045 16.443 0.000 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ## attitudes ~~ ## norms 0.347 0.069 5.027 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .attit_1 0.420 0.052 8.103 0.000 ## .attit_2 0.306 0.046 6.604 0.000 ## .attit_3 0.372 0.049 7.651 0.000 ## .norm_1 0.483 0.064 7.581 0.000 ## .norm_2 0.521 0.065 7.954 0.000 ## .norm_3 0.610 0.070 8.713 0.000 ## .intent 0.423 0.048 8.769 0.000 ## .behavior 0.603 0.054 11.180 0.000 ## attitudes 0.884 0.116 7.614 0.000 ## norms 0.765 0.113 6.767 0.000 ## ## R-Square: ## Estimate ## attit_1 0.678 ## attit_2 0.757 ## attit_3 0.705 ## norm_1 0.613 ## norm_2 0.587 ## norm_3 0.523 ## intent 0.639 ## behavior 0.520 Yes, the model still fits the data very well. Yes, the estimates all align with the TORA. Specifically, attitudes and norms both significantly predict intention, and intention significantly predicts condom use. The model explains 63.93% of the variance in intention and 51.96% of the variance in condom use. 6.4.4 Update your model to represent the extended TORA model that includes perceived behavioral control. Regress condom use onto perceived behavioral control. Use the latent variable representation of control. Covary all three exogenous latent factors. Does the model fit well? Do the estimates align with the updated TORA? How much variance in intention and condom use are explained by the model? Click for explanation mod_tora <- ' attitudes =~ attit_1 + attit_2 + attit_3 norms =~ norm_1 + norm_2 + norm_3 control =~ control_1 + control_2 + control_3 intent ~ attitudes + norms behavior ~ intent + control ' fit_tora <- sem(mod_tora, data = condom) summary(fit_tora, fit.measures = TRUE, rsquare = TRUE) ## lavaan 0.6.16 ended normally after 31 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 27 ## ## Number of observations 250 ## ## Model Test User Model: ## ## Test statistic 48.757 ## Degrees of freedom 39 ## P-value (Chi-square) 0.136 ## ## Model Test Baseline Model: ## ## Test statistic 1333.695 ## Degrees of freedom 55 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.992 ## Tucker-Lewis Index (TLI) 0.989 ## ## Loglikelihood and Information Criteria: ## ## Loglikelihood user model (H0) -3551.160 ## Loglikelihood unrestricted model (H1) -3526.782 ## ## Akaike (AIC) 7156.320 ## Bayesian (BIC) 7251.400 ## Sample-size adjusted Bayesian (SABIC) 7165.807 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.032 ## 90 Percent confidence interval - lower 0.000 ## 90 Percent confidence interval - upper 0.057 ## P-value H_0: RMSEA <= 0.050 0.870 ## P-value H_0: RMSEA >= 0.080 0.000 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.033 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) ## attitudes =~ ## attit_1 1.000 ## attit_2 1.033 0.068 15.221 0.000 ## attit_3 -1.025 0.068 -15.097 0.000 ## norms =~ ## norm_1 1.000 ## norm_2 0.984 0.087 11.256 0.000 ## norm_3 0.955 0.088 10.881 0.000 ## control =~ ## control_1 1.000 ## control_2 0.859 0.127 6.789 0.000 ## control_3 0.997 0.131 7.609 0.000 ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## intent ~ ## attitudes 0.447 0.063 7.100 0.000 ## norms 0.706 0.078 9.078 0.000 ## behavior ~ ## intent 0.563 0.063 8.923 0.000 ## control 0.454 0.119 3.805 0.000 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ## attitudes ~~ ## norms 0.342 0.068 5.011 0.000 ## control 0.474 0.072 6.548 0.000 ## norms ~~ ## control 0.352 0.064 5.521 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .attit_1 0.432 0.052 8.381 0.000 ## .attit_2 0.330 0.046 7.220 0.000 ## .attit_3 0.344 0.046 7.439 0.000 ## .norm_1 0.496 0.063 7.820 0.000 ## .norm_2 0.533 0.065 8.152 0.000 ## .norm_3 0.595 0.069 8.643 0.000 ## .control_1 0.625 0.075 8.372 0.000 ## .control_2 0.876 0.090 9.757 0.000 ## .control_3 0.746 0.084 8.874 0.000 ## .intent 0.409 0.047 8.769 0.000 ## .behavior 0.542 0.052 10.423 0.000 ## attitudes 0.872 0.115 7.566 0.000 ## norms 0.751 0.112 6.709 0.000 ## control 0.485 0.096 5.059 0.000 ## ## R-Square: ## Estimate ## attit_1 0.668 ## attit_2 0.738 ## attit_3 0.727 ## norm_1 0.602 ## norm_2 0.577 ## norm_3 0.535 ## control_1 0.437 ## control_2 0.290 ## control_3 0.392 ## intent 0.651 ## behavior 0.566 Yes, the model still fits the data very well. Yes, the estimates all align with the updated TORA. Specifically, attitudes and norms both significantly predict intention, while intention and control both significantly predict condom use. The model explains 65.11% of the variance in intention and 56.62% of the variance in condom use. The TORA model explicitly forbids direct paths from attitudes and norms to behaviors; these effects should be fully mediated by the behavioral intention. The theory does not specify how perceived behavioral control should affect behaviors. There may be a direct effect of control on behavior, or the effect may be (partially) mediated by intention. 6.4.5 Evaluate the hypothesized indirect effects of attitudes and norms. Include attitudes, norms, and control in your model as in 6.4.4. Does intention significantly mediate the effects of attitudes and norms on behavior? Don’t forget to follow all the steps we covered for testing mediation. Are both of the above effects completely mediated? Do these results comport with the TORA? Why or why not? Click for explanation mod <- ' attitudes =~ attit_1 + attit_2 + attit_3 norms =~ norm_1 + norm_2 + norm_3 control =~ control_1 + control_2 + control_3 intent ~ a1 * attitudes + a2 * norms behavior ~ b * intent + control + attitudes + norms ie_att := a1 * b ie_norm := a2 * b ' set.seed(235711) fit <- sem(mod, data = condom, se = "bootstrap", bootstrap = 1000) summary(fit, ci = TRUE) ## lavaan 0.6.16 ended normally after 36 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 29 ## ## Number of observations 250 ## ## Model Test User Model: ## ## Test statistic 48.629 ## Degrees of freedom 37 ## P-value (Chi-square) 0.096 ## ## Parameter Estimates: ## ## Standard errors Bootstrap ## Number of requested bootstrap draws 1000 ## Number of successful bootstrap draws 1000 ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) ci.lower ci.upper ## attitudes =~ ## attit_1 1.000 1.000 1.000 ## attit_2 1.033 0.060 17.261 0.000 0.925 1.165 ## attit_3 -1.025 0.064 -15.894 0.000 -1.163 -0.902 ## norms =~ ## norm_1 1.000 1.000 1.000 ## norm_2 0.984 0.071 13.794 0.000 0.843 1.127 ## norm_3 0.955 0.093 10.324 0.000 0.792 1.157 ## control =~ ## control_1 1.000 1.000 1.000 ## control_2 0.860 0.113 7.624 0.000 0.653 1.098 ## control_3 0.996 0.147 6.790 0.000 0.748 1.320 ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ci.lower ci.upper ## intent ~ ## attitudes (a1) 0.447 0.067 6.674 0.000 0.324 0.585 ## norms (a2) 0.706 0.078 9.094 0.000 0.569 0.878 ## behavior ~ ## intent (b) 0.545 0.075 7.282 0.000 0.389 0.686 ## control 0.428 0.232 1.847 0.065 0.046 0.934 ## attitudes 0.010 0.122 0.084 0.933 -0.249 0.226 ## norms 0.041 0.118 0.345 0.730 -0.194 0.266 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ci.lower ci.upper ## attitudes ~~ ## norms 0.342 0.070 4.883 0.000 0.208 0.480 ## control 0.475 0.069 6.850 0.000 0.344 0.612 ## norms ~~ ## control 0.350 0.067 5.218 0.000 0.221 0.484 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ci.lower ci.upper ## .attit_1 0.432 0.050 8.720 0.000 0.331 0.526 ## .attit_2 0.330 0.045 7.382 0.000 0.238 0.415 ## .attit_3 0.343 0.049 6.992 0.000 0.244 0.444 ## .norm_1 0.496 0.060 8.305 0.000 0.376 0.614 ## .norm_2 0.533 0.077 6.951 0.000 0.390 0.687 ## .norm_3 0.594 0.069 8.597 0.000 0.443 0.719 ## .control_1 0.624 0.076 8.216 0.000 0.477 0.763 ## .control_2 0.875 0.092 9.495 0.000 0.686 1.052 ## .control_3 0.745 0.079 9.398 0.000 0.574 0.889 ## .intent 0.409 0.050 8.169 0.000 0.309 0.507 ## .behavior 0.544 0.058 9.379 0.000 0.415 0.639 ## attitudes 0.872 0.104 8.387 0.000 0.675 1.077 ## norms 0.751 0.099 7.557 0.000 0.556 0.941 ## control 0.486 0.096 5.042 0.000 0.303 0.684 ## ## Defined Parameters: ## Estimate Std.Err z-value P(>|z|) ci.lower ci.upper ## ie_att 0.244 0.050 4.860 0.000 0.150 0.352 ## ie_norm 0.385 0.066 5.835 0.000 0.268 0.527 Yes, both indirect effects are significant according to the 95% bootstrapped CIs. Yes, both effects are completely moderated by behavioral intention. We can infer as much because the direct effects of attitudes and norms on condom use are both nonsignificant. Yes, these results comport with the TORA. Both effects are fully mediated, as the theory stipulates. In addition to evaluating the significance of the indirect and direct effects, we can also take a model-comparison perspective. We can use model comparisons to test if removing the direct effects of attitudes and norms on condom use significantly decreases model fit. In other words, are those paths needed to accurately represent the data, or are they “dead weight”. 6.4.6 Use a \\(\\Delta \\chi^2\\) test to evaluate the necessity of including the direct effects of attitudes and norms on condom use in the model. What is your conclusion? Click for explanation We only need to compare the fit of the model with the direct effects included to the fit of the model without the direct effects. We’ve already estimated both models, so we can simply submit the fitted lavaan objects to the anova() function. anova(fit, fit_tora) The \\(\\Delta \\chi^2\\) test is not significant. So, we have not lost a significant amount of fit by fixing the direct effects to zero. In other words, the complete mediation model explains the data just as well as the partial mediation model. So, we should probably prefer the more parsimonious model. 6.4.7 Use some statistical means of evaluating the most plausible way to include perceived behavioral control into the model. Choose between the following three options: control predicts behavior via a direct, un-mediated effect. control predicts behavior via an indirect effect that is completely mediated by intention. control predicts behavior via both an indirect effect through intention and a residual direct effect. Hint: There is more than one way to approach this problem. Click for explanation We can tackle this problem in a few different ways. One possibility entails testing the indirect, direct, and total effects. ## Allow for partial mediation: mod1 <- ' attitudes =~ attit_1 + attit_2 + attit_3 norms =~ norm_1 + norm_2 + norm_3 control =~ control_1 + control_2 + control_3 intent ~ attitudes + norms + a * control behavior ~ b * intent + c * control ie := a * b total := ie + c ' set.seed(235711) fit1 <- sem(mod1, data = condom, se = "bootstrap", bootstrap = 1000) summary(fit1, ci = TRUE) ## lavaan 0.6.16 ended normally after 33 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 28 ## ## Number of observations 250 ## ## Model Test User Model: ## ## Test statistic 47.389 ## Degrees of freedom 38 ## P-value (Chi-square) 0.141 ## ## Parameter Estimates: ## ## Standard errors Bootstrap ## Number of requested bootstrap draws 1000 ## Number of successful bootstrap draws 1000 ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) ci.lower ci.upper ## attitudes =~ ## attit_1 1.000 1.000 1.000 ## attit_2 1.034 0.060 17.222 0.000 0.925 1.167 ## attit_3 -1.021 0.064 -15.877 0.000 -1.158 -0.898 ## norms =~ ## norm_1 1.000 1.000 1.000 ## norm_2 0.985 0.071 13.803 0.000 0.848 1.133 ## norm_3 0.948 0.093 10.204 0.000 0.786 1.155 ## control =~ ## control_1 1.000 1.000 1.000 ## control_2 0.861 0.113 7.635 0.000 0.653 1.100 ## control_3 0.996 0.142 7.020 0.000 0.760 1.318 ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ci.lower ci.upper ## intent ~ ## attitudes 0.357 0.115 3.113 0.002 0.146 0.603 ## norms 0.646 0.095 6.794 0.000 0.473 0.859 ## control (a) 0.199 0.199 1.002 0.317 -0.188 0.633 ## behavior ~ ## intent (b) 0.551 0.074 7.487 0.000 0.391 0.683 ## control (c) 0.469 0.142 3.298 0.001 0.231 0.791 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ci.lower ci.upper ## attitudes ~~ ## norms 0.344 0.070 4.905 0.000 0.210 0.481 ## control 0.471 0.069 6.838 0.000 0.342 0.608 ## norms ~~ ## control 0.345 0.066 5.240 0.000 0.215 0.481 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ci.lower ci.upper ## .attit_1 0.429 0.050 8.628 0.000 0.329 0.524 ## .attit_2 0.325 0.045 7.230 0.000 0.233 0.408 ## .attit_3 0.347 0.049 7.011 0.000 0.248 0.455 ## .norm_1 0.490 0.060 8.172 0.000 0.373 0.612 ## .norm_2 0.525 0.076 6.869 0.000 0.385 0.684 ## .norm_3 0.599 0.070 8.529 0.000 0.447 0.729 ## .control_1 0.626 0.074 8.429 0.000 0.479 0.761 ## .control_2 0.875 0.092 9.522 0.000 0.689 1.049 ## .control_3 0.748 0.078 9.532 0.000 0.579 0.893 ## .intent 0.412 0.050 8.283 0.000 0.307 0.504 ## .behavior 0.541 0.055 9.873 0.000 0.423 0.639 ## attitudes 0.875 0.104 8.385 0.000 0.676 1.081 ## norms 0.757 0.099 7.616 0.000 0.560 0.949 ## control 0.484 0.095 5.092 0.000 0.306 0.683 ## ## Defined Parameters: ## Estimate Std.Err z-value P(>|z|) ci.lower ci.upper ## ie 0.110 0.105 1.048 0.295 -0.105 0.309 ## total 0.578 0.186 3.108 0.002 0.235 0.971 From the above results, we can see that the direct and total effects are both significant, but the indirect effect is not. Hence, it probably makes the most sense to include control via a direct (non-mediated) effect on behavior. We can also approach this problem from a model-comparison perspective. We can fit models that encode each pattern of constraints and check which one best represents the data. ## Force complete mediation: mod2 <- ' attitudes =~ attit_1 + attit_2 + attit_3 norms =~ norm_1 + norm_2 + norm_3 control =~ control_1 + control_2 + control_3 intent ~ attitudes + norms + control behavior ~ intent ' ## Force no mediation: mod3 <- ' attitudes =~ attit_1 + attit_2 + attit_3 norms =~ norm_1 + norm_2 + norm_3 control =~ control_1 + control_2 + control_3 intent ~ attitudes + norms behavior ~ intent + control ' ## Estimate the two restricted models: fit2 <- sem(mod2, data = condom) fit3 <- sem(mod3, data = condom) ## Check the results: summary(fit2) ## lavaan 0.6.16 ended normally after 33 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 27 ## ## Number of observations 250 ## ## Model Test User Model: ## ## Test statistic 62.797 ## Degrees of freedom 39 ## P-value (Chi-square) 0.009 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) ## attitudes =~ ## attit_1 1.000 ## attit_2 1.033 0.068 15.295 0.000 ## attit_3 -1.018 0.068 -15.087 0.000 ## norms =~ ## norm_1 1.000 ## norm_2 0.985 0.087 11.305 0.000 ## norm_3 0.947 0.087 10.845 0.000 ## control =~ ## control_1 1.000 ## control_2 0.864 0.126 6.855 0.000 ## control_3 0.958 0.129 7.417 0.000 ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## intent ~ ## attitudes 0.352 0.096 3.669 0.000 ## norms 0.644 0.088 7.347 0.000 ## control 0.207 0.163 1.268 0.205 ## behavior ~ ## intent 0.746 0.045 16.443 0.000 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ## attitudes ~~ ## norms 0.345 0.069 5.023 0.000 ## control 0.476 0.073 6.513 0.000 ## norms ~~ ## control 0.346 0.065 5.361 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .attit_1 0.427 0.051 8.295 0.000 ## .attit_2 0.325 0.046 7.101 0.000 ## .attit_3 0.349 0.047 7.477 0.000 ## .norm_1 0.490 0.064 7.702 0.000 ## .norm_2 0.524 0.065 8.025 0.000 ## .norm_3 0.600 0.069 8.652 0.000 ## .control_1 0.610 0.076 8.015 0.000 ## .control_2 0.861 0.090 9.580 0.000 ## .control_3 0.769 0.086 8.938 0.000 ## .intent 0.412 0.046 8.890 0.000 ## .behavior 0.603 0.054 11.180 0.000 ## attitudes 0.877 0.115 7.596 0.000 ## norms 0.757 0.112 6.733 0.000 ## control 0.500 0.098 5.076 0.000 summary(fit3) ## lavaan 0.6.16 ended normally after 31 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 27 ## ## Number of observations 250 ## ## Model Test User Model: ## ## Test statistic 48.757 ## Degrees of freedom 39 ## P-value (Chi-square) 0.136 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) ## attitudes =~ ## attit_1 1.000 ## attit_2 1.033 0.068 15.221 0.000 ## attit_3 -1.025 0.068 -15.097 0.000 ## norms =~ ## norm_1 1.000 ## norm_2 0.984 0.087 11.256 0.000 ## norm_3 0.955 0.088 10.881 0.000 ## control =~ ## control_1 1.000 ## control_2 0.859 0.127 6.789 0.000 ## control_3 0.997 0.131 7.609 0.000 ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## intent ~ ## attitudes 0.447 0.063 7.100 0.000 ## norms 0.706 0.078 9.078 0.000 ## behavior ~ ## intent 0.563 0.063 8.923 0.000 ## control 0.454 0.119 3.805 0.000 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ## attitudes ~~ ## norms 0.342 0.068 5.011 0.000 ## control 0.474 0.072 6.548 0.000 ## norms ~~ ## control 0.352 0.064 5.521 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .attit_1 0.432 0.052 8.381 0.000 ## .attit_2 0.330 0.046 7.220 0.000 ## .attit_3 0.344 0.046 7.439 0.000 ## .norm_1 0.496 0.063 7.820 0.000 ## .norm_2 0.533 0.065 8.152 0.000 ## .norm_3 0.595 0.069 8.643 0.000 ## .control_1 0.625 0.075 8.372 0.000 ## .control_2 0.876 0.090 9.757 0.000 ## .control_3 0.746 0.084 8.874 0.000 ## .intent 0.409 0.047 8.769 0.000 ## .behavior 0.542 0.052 10.423 0.000 ## attitudes 0.872 0.115 7.566 0.000 ## norms 0.751 0.112 6.709 0.000 ## control 0.485 0.096 5.059 0.000 ## Do either of the restricted models fit worse than the partial mediation model? anova(fit1, fit2) anova(fit1, fit3) The above \\(\\Delta \\chi^2\\) tests tell us that the full mediation model fits significantly worse than the partial mediation model. Hence, forcing full mediation by fixing the direct effect to zero is an unreasonable restraint. The total effect model, on the other hand, does not fit significantly worse than the partial mediation model. So, we can conclude that removing the indirect effect and modeling the influence of control on behavior as an un-mediated direct association represents the data just as well as a model that allows for both indirect and direct effects. Hence, we should prefer the more parsimonious total effects model. While the two tests above lead us to prefer the non-mediated model, we cannot directly say that the the complete mediation model fits significantly worse than the non-mediated model. We have not directly compared those two models, and we cannot do so with the \\(\\Delta \\chi^2\\). We cannot do such a test because these two models are not nested: we must both add and remove a path to get from one model specification to the other. Also, both models have the same degrees of freedom, so we cannot define a sampling distribution against which we would compare the \\(\\Delta \\chi^2\\), anyway. We are not completely without options, though. We can use information criteria to compare non-nested models. The two most popular information criteria are the Akaike’s Information Criterion (AIC) and the Bayesian Information Criterion (BIC). ## Which model is the most parsimonious representation of the data? AIC(fit1, fit2, fit3) BIC(fit1, fit2, fit3) When comparing models based on information criteria, a lower value indicates a better model in the sense of a better balance of fit and parsimony. The above results show that both the AIC and the BIC agree that the no-mediation model is the best. So, in the end, regardless of how we approach the question, all of our results suggest modeling perceived behavioral control as a direct, non-mediated predictor of condom use. End of In-Class Exercises "],["multiple-group-models.html", "7 Multiple Group Models", " 7 Multiple Group Models This week, you will cover multiple group modeling and measurement invariance testing in the SEM/CFA context. Homework before the lecture Watch the Lecture Recording for this week. Complete the Reading for this week, and answer the associated reading questions. Homework before the practical Complete the At-Home Exercises. Practical content During the practical you will work on the In-Class Exercises. "],["lecture-6.html", "7.1 Lecture", " 7.1 Lecture In this lecture, we will explore how you can incorporate grouping factors into your CFA and SEM analyses. We’ll cover three general topics: The multiple group modeling framework Measurement invariance testing Using multiple group models to test for moderation 7.1.1 Recordings Once it’s ready, the lecture recording will be embedded below. 7.1.2 Slides You can download the lecture slides here "],["reading-6.html", "7.2 Reading", " 7.2 Reading Coming soon to a GitBook near you! "],["at-home-exercises-6.html", "7.3 At-Home Exercises", " 7.3 At-Home Exercises Coming soon to a GitBook near you! "],["in-class-exercises-6.html", "7.4 In-Class Exercises", " 7.4 In-Class Exercises Coming soon to a GitBook near you! "],["wrap-up.html", "8 Wrap-Up", " 8 Wrap-Up Information This is an open week that we’ll use to tie up any loose ends and wrap up the course content. "],["404.html", "Page not found", " Page not found The page you requested cannot be found (perhaps it was moved or renamed). You may want to try searching to find the page's new location, or use the table of contents to find the page you are looking for. "]] +[["index.html", "Theory Construction and Statistical Modeling Course Information", " Theory Construction and Statistical Modeling Kyle M. Lang Last updated: 2023-10-15 Course Information In order to test a theory, we must express the theory as a statistical model and then test this model on quantitative (numeric) data. In this course we will use datasets from different disciplines within the social sciences (educational sciences, psychology, and sociology) to explain and illustrate theories and practices that are used in all social science disciplines to statistically model social science theories. This course uses existing tutorial datasets to practice the process of translating verbal theories into testable statistical models. If you are interested in the methods of acquiring high quality data to test your own theory, we recommend following the course Conducting a Survey which is taught from November to January. Most information about the course is available in this GitBook. Course-related communication will be through https://uu.blackboard.com (Log in with your student ID and password). "],["acknowledgement.html", "Acknowledgement", " Acknowledgement This course was originally developed by dr. Caspar van Lissa. I (dr. Kyle M. Lang) have modified Caspar’s original materials and take full responsibility for any errors or inaccuracies introduced through these modifications. Credit for any particularly effective piece of pedagogy should probably go to Caspar. You can view the original version of this course here on Caspar’s GitHub page. "],["instructors.html", "Instructors", " Instructors Coordinator: dr. Kyle M. Lang Lectures: dr. Kyle M. Lang Practicals: Rianne Kraakman Daniëlle Remmerswaal Danielle McCool "],["course-overview.html", "Course overview", " Course overview This course comprises three parts: Path analysis: You will learn how to estimate complex path models of observed variables (e.g., linked linear regressions) as structural equation models. Factor analysis: You will learn different ways of defining and estimating latent (unobserved) constructs. Full structural equation modeling: You will combine the first two topics to estimate path models describing the associations among latent constructs. Each of these three themes will be evaluated with a separate assignment. The first two assignments will be graded on a pass/fail basis. Your course grade will be based on your third assignment grade. "],["schedule.html", "Schedule", " Schedule Course Week Calendar Week Lecture/Practical Topic Workgroup Activity Assignment Deadline 0 36 Pre-course preparation 1 37 Introduction to R 2 38 Statistical modeling, Path analysis 3 39 Mediation, Moderation 4 40 Exploratory factor analysis (EFA) A1 Peer-Review A1: 2023-10-04 @ 23:59 5 41 Confirmatory factor analysis (CFA) 6 42 Structural equation modeling (SEM) A2 Peer-Review A2: 2023-10-18 @ 23:59 7 43 Multiple group models 8 44 Wrap-up A3 Peer-Review 9 45 Exam week: No class meetings A3: 2023-11-10 @ 23:59 NOTE: The schedule (including topics covered and assignment deadlines) is subject to change at the instructors’ discretion. "],["learning-goals.html", "Learning goals", " Learning goals In this course you will learn how to translate a social scientific theory into a statistical model, how to analyze your data with these models, and how to interpret and report your results following APA standards. After completing the course, you will be able to: Translate a verbal theory into a conceptual model, and translate a conceptual model into a statistical model. Independently analyze data using the free, open-source statistical software R. Apply a latent variable model to a real-life problem wherein the observed variables are only indirect indicators of an unobserved construct. Use a path model to represent the hypothesized causal relations among several variables, including relationships such as mediation and moderation. Explain to a fellow student how structural equation modeling combines latent variable models with path models and the benefits of doing so. Reflect critically on the decisions involved in defining and estimating structural equation models. "],["resources.html", "Resources", " Resources Literature You do not need a separate book for this course! Most of the information is contained within this GitBook and the course readings (which you will be able to access via links in this GitBook). All literature is freely available online, as long as you are logging in from within the UU-domain (i.e., from the UU campus or through an appropriate VPN). All readings are linked in this GitBook via either direct download links or DOIs. If you run into any trouble accessing a given article, searching for the title using Google Scholar or the University Library will probably due the trick. Software You will do all of your statistical analyses with the statistical programming language/environment R and the add-on package lavaan. If you want to expand your learning, you can follow this optional lavaan tutorial. "],["reading-questions.html", "Reading questions", " Reading questions Along with every article, we will provide reading questions. You will not be graded on the reading questions, but it is important to prepare the reading questions before every lecture. The reading questions serve several important purposes: Provide relevant background knowledge for the lecture Help you recognize and understand the key terms and concepts Make you aware of important publications that shaped the field Help you extract the relevant insights from the literature "],["weekly-preparation.html", "Weekly preparation", " Weekly preparation Before every class meeting (both lectures and practicals) you need to do the assigned homework (delineated in the GitBook chapter for that week). This course follows a flipped classroom procedure, so you must complete the weekly homework to meaningfully participate in, and benefit from, the class meetings. Background knowledge We assume you have basic knowledge about multivariate statistics before entering this course. You do not need any prior experience working with R. If you wish to refresh your knowledge, we recommend the chapters on ANOVA, multiple regression, and exploratory factor analysis from Field’s Discovering Statistics using R. If you cannot access the Field book, many other introductory statistics textbooks cover these topics equally well. So, use whatever you have lying around from past statistics courses. You could also try one of the following open-access options: Applied Statistics with R Introduction to Modern Statistics Introduction to Statistical Learning "],["grading.html", "Grading", " Grading Your grade for the course is based on a “portfolio” composed of the three take-home assignments: Path modeling Deadline: Wednesday 2023-10-04 at 23:59 Group assignment Pass/Fail Confirmatory factor analysis Deadline: Wednesday 2023-10-18 at 23:59 Group assignment Pass/Fail Full structural equation modeling Deadline: Friday 2023-11-10 at 23:59 Individual assignment Comprises your entire numeric course grade The specifics of the assignments will be explicated in the Assignments chapter of this GitBook "],["attendance.html", "Attendance", " Attendance Attendance is not mandatory, but we strongly encourage you to attend all lectures and practicals. In our experience, students who actively participate tend to pass the course, whereas those who do not participate tend to drop out or fail. The lectures and practicals build on each other, so, in the unfortunate event that you have to miss a class meeting, please make sure you have caught up with the material before the next session. "],["assignments.html", "Assignments", " Assignments This chapter contains the details and binding information about the three assignments that comprise the portfolio upon which your course grade is based. Below, you can find a brief idea of what each assignment will cover. For each assignment, you will use R to analyze some real-world data, and you will write up your results in a concise report (not a full research paper). Guidelines for these analyses/reports are delineated in the following three sections. You will submit your reports via Blackboard. You will complete the first two assignments in your Assignment Groups. You will complete the third assignment individually. The first two assignments are graded as pass/fail. You must pass both of these assignments to pass the course. The third assignment constitutes your course grade. "],["assignment-1-path-analysis.html", "Assignment 1: Path Analysis", " Assignment 1: Path Analysis For the first assignment, you will work in groups to apply a path model that describes how several variables could be causally related. The components of the first assignment are described below. Choose a suitable dataset, and describe the data. You can use any of the 8 datasets linked below. State the research question; define and explicate the theoretical path model. This model must include, at least, three variables. Use a path diagram to show your theoretical model. Translate your theoretical path model into lavaan syntax, and estimate the model. Include the code used to define and estimate your model as an appendix. Explain your rationale for important modeling decisions. Discuss the conceptual fit between your theory and your model. Evaluate the model assumptions. Discuss other important decisions that could have influence your results. Report the results in APA style. Provide relevant output in a suitable format. Include measures of explained variance for the dependent variables. Discuss the results. Use your results to answer the research question. Consider the strengths and limitations of your analysis. Evaluation See the Grading section below for more information on how Assignment 1 will be evaluated. You can access an evaluation matrix for Assignment 1 here. This matrix gives an indication of what level of work constitutes insufficient, sufficient, and excellent responses to the six components described above. Submission Assignment 1 is due at 23:59 on Wednesday 4 October 2023. Submit your report via the Assignment 1 portal on Blackboard. "],["assignment-2-confirmatory-factor-analysis.html", "Assignment 2: Confirmatory Factor Analysis", " Assignment 2: Confirmatory Factor Analysis In the second assignment, you will work in groups to run a CFA wherein the observed variables are indirect indicators of the unobserved constructs you want to analyze. The components of the second assignment are described below. Choose a suitable dataset, and describe the data. Ideally, you will work with the same data that you analyzed in Assignment 1. If you want to switch, you can use any of the 8 datasets linked below. State the research question; define and explicate the theoretical CFA model. This model must include, at least, two latent constructs. Use a path diagram to represent your model. Translate your theoretical model into lavaan syntax, and estimate the model. Include the code used to define and estimate your model as an appendix. Explain your rationale for important modeling decisions. Discuss the conceptual fit between your theory and your model. Evaluate the model assumptions. Discuss other important decisions that could have influence your results. Report the results in APA style. Provide relevant output in a suitable format. Include measures of model fit. Discuss the results. Use your results to answer the research question. Consider the strengths and limitations of your analysis. Evaluation See the Grading section below for more information on how Assignment 2 will be evaluated. You can access an evaluation matrix for Assignment 2 here. This matrix gives an indication of what level of work constitutes insufficient, sufficient, and excellent responses to the six components described above. Submission Assignment 2 is due at 23:59 on Wednesday 18 October 2023. Submit your report via the Assignment 2 portal on Blackboard. "],["a3_components.html", "Assignment 3: Full Structural Equation Model", " Assignment 3: Full Structural Equation Model In the third assignment, you will work individually to apply a full SEM that describes how several (latent) variables could be causally related. The components of the third assignment are described below. Choose a suitable dataset, and describe the data. Ideally, you will work with the same data that you analyzed in Assignments 1 & 2. If you want to switch, you can use any of the 8 datasets linked below. State the research question; define and explicate the theoretical SEM. The structural component of this model must include, at least, three variables. The model must include, at least, two latent variables. Use a path diagram to represent your model. Translate your theoretical SEM into lavaan syntax, and estimate the model. Include the code used to define and estimate your model as an appendix. Explain your rationale for important modeling decisions. Discuss the conceptual fit between your theory and your model. Evaluate the model assumptions. Discuss other important decisions that could have influence your results. Report the results in APA style. Provide relevant output in a suitable format. Include measures of model fit. Include measures of explained variance for the dependent variables. Discuss the results. Use your results to answer the research question. Consider the strengths and limitations of your analysis. Evaluation See the Grading section below for more information on how the component scores represented in the rubric are combined into an overall assignment grade. You can access an evaluation matrix for Assignment 3 here. This matrix gives an indication of what level of work constitutes insufficient, sufficient, and excellent responses to the six components described above. Submission Assignment 3 is due at 23:59 on Friday 10 November 2023. Submit your report via the Assignment 3 portal on Blackboard. "],["elaboration-tips.html", "Elaboration & Tips", " Elaboration & Tips Theoretical Model & Research Question You need to provide some justification for your model and research question, but only enough to demonstrate that you’ve actually conceptualized and estimated a theoretically plausible statistical model (as opposed to randomly combining variables until lavaan returns a pretty picture). You have several ways to show that your model is plausible. Use common-sense arguments. Reference (a small number of) published papers. Replicate an existing model/research question. Don’t provide a rigorous literature-supported theoretical motivation. You don’t have the time to conduct a thorough literature review, and we don’t have the time to read such reviews when grading. Literature review is not one of the learning goals for this course, so you cannot get “bonus points” for an extensive literature review. You are free to test any plausible model that meets the size requirements. You can derive your own model/research question or you can replicate a published analysis. Model Specifications We will not cover methods for modeling categorical outcome variables. So, use only continuous variables as outcomes. DVs in path models and the structural parts of SEMs Observed indicators of latent factors in CFA/SEM NOTE: You may treat ordinal items as continuous, for the purposes of these assignments. We will not cover methods for latent variable interactions. Don’t specify a theoretical model that requires an interaction involving a latent construct. There is one exception to the above prohibition. If the moderator is an observed grouping variable, you can estimate the model as a multiple-group model. We’ll cover these methods in Week 7. Assumptions You need to show that you’re thinking about the assumptions and their impact on your results, but you don’t need to run thorough model diagnostics. Indeed, the task of checking assumptions isn’t nearly as straight forward in path analysis, CFA, and SEM as it is in linear regression modeling. You won’t be able to directly apply the methods you have learned for regression diagnostics, for example. Since all of our models are estimated with normal-theory maximum likelihood, the fundamental assumption of all the models we’ll consider in this course boils down to the following. All random variables in my model are i.i.d. multivariate normally distributed. So, you can get by with basic data screening and checking the observed random variables in your model (i.e., all variables other than fixed predictors) for normality. Since checking for multivariate normality is a bit tricky, we’ll only ask you to evaluate univariate normality. You should do these evaluations via graphical means. To summarize, we’re looking for the following. Data Consider whether the measurement level of your data matches the assumptions of your model. Check your variables for univariate outliers. If you find any outliers, either treat them in some way or explain why you are retaining them for the analysis. Check for missing data. For the purposes of the assignment, you can use complete case analysis to work around the missing data. If you’re up for more of a challenge, feel free to try multiple imputation or full information maximum likelihood. Model Evaluate the univariate normality of any random, observed variables in your model. E.g., DVs in path models, observed IVs modeled as random variables, indicators of latent factors If you fit a multiple-group model for Assignment 3, do this evaluation within groups. Use graphical tools to evaluate the normality assumption. Normal QQ-Plots Histograms Results What do we mean by reporting your results “in a suitable format”? Basically, put some effort into making your results readable, and don’t include a bunch of superfluous information. Part of demonstrating that you understand the analysis is showing that you know which pieces of output convey the important information. Tabulate your results; don’t directly copy the R output. Don’t include everything lavaan gives you. Include only the output needed to understand your results and support your conclusions. "],["data_options.html", "Data", " Data Below, you can find links to a few suitable datasets that you can use for the assignments. You must use one of the following datasets. You may not choose your own data from the wild. Coping with Covid Dataset Codebook Pre-Registration Feminist Perspectives Scale Dataset Article Hypersensitive Narcissism Scale & Dirty Dozen Dataset HSNS Article DD Article Kentucky Inventory of Mindfulness Skills Dataset Article Depression Anxiety Stress Scale Dataset DASS Information Nomophobia Dataset Recylced Water Acceptance Dataset Article "],["procedures.html", "Procedures", " Procedures Formatting You must submit your assignment reports in PDF format. Each report should include a title page. The title page should include the following information: The name of the assignment. The names of all assignment authors (i.e., all group members for Assignments 1 & 2, your name for Assignment 3). The Assignment Group number (only for Assignments 1 & 2). You must include the code used to define and run your model(s) as an appendix. Try to format the text in this appendix clearly. Use a monospace font. Length You may use as many words as necessary to adequately explain yourself; though, concision and parsimony are encouraged. Note that the assignments are not intended to be full-blown papers! The focus should be on the definition of your model, how this model relates to theory (introduction), and what you have learned from your estimated model (discussion). For each of the assignments, you should be able to get the job done in fewer than 10 pages of text (excluding title page, figures, appendices, and references). Submission You will submit your reports through Blackboard. Each assignment has a corresponding item in the “Assignments” section of the BB page through which you will submit your reports. For Assignments 1 & 2, you may only submit one report per group. Designate one group member to submit the report. The grade for this submission will apply to all group members. If something goes wrong with the submission, or you notice a mistake (before the deadline) that you want to correct, you may upload a new version of your report. We will grade the final submitted version. The submissions will be screened with Ouriginal. "],["grading-1.html", "Grading", " Grading Group Assignments Assignments 1 & 2 are simply graded as pass/fail. To pass, your submission must: Do a reasonable job of addressing the relevant components listed above Be submitted before the deadline Otherwise, you will fail the assignment. Individual Assignment Assignment 3 will be fully graded on the usual 10-point scale. Points will be allocated according to the extent to which your submission addresses the six components listed above. The evaluation matrix gives an indication of how these points will be apportioned. Further details over the grading procedures for Assignment 3 (e.g., exactly how your 10-point grade will be defined) will be provided at a later date. Assuming your group passes the first two assignments, your final course grade will simply be your Assignment 3 grade. Resits You must get a “pass” for Assignments 1 & 2 and score at least 5.5 on Assignment 3 to pass the course. If you fail any of the assignments, you will have the opportunity to resit the failed assignment(s). If you resit Assignment 3, your revised graded cannot be higher than 6. Further details on the resit procedure will be provided at a later date. Example Assignment You can find an example of a good submission (for an older version of Assignment 2) here. This example is not perfect (no paper ever is), and several points could be improved. That being said, this submission exemplifies what we’re looking for in your project reports. So, following the spirit of this example would earn you a high grade. "],["rules.html", "Rules", " Rules Resources For all three assignments, you may use any reference materials you like, including: All course materials The course GitBook Additional books and papers The internet Collaboration You will complete the first two assignments in groups. Although you will work in groups, your group may not work together with other groups. You will complete the final assignment individually. For this assignment, you may not work with anyone else. For all three assignments, you are obligated to submit original work (i.e., work conducted for this course by you or your group). Submitting an assignment that violates this condition constitutes fraud. Such cases of fraud will be addressed according to the University’s standard policy. Academic integrity Hopefully, you also feel a moral obligation to obey the rules. For this course, we have implemented an examination that allows you to showcase what you have learned in a more realistic way than a written exam would allow. This assessment format spares you the stress of long exams (the two exams for this course used to be 4 hours each) and the attendant studying/cramming. The assignments will also help you assess your ability to independently analyse data, which is important to know for your future courses and/or career. However, this format also assumes that you complete the assignments in good faith. So, I simply ask that you hold up your end of the bargain, and submit your original work to show us what you’ve learned. Strict stuff By submitting your assignments (both group and individual), you confirm the following: You have completed the assignment yourself (or with your group) You are submitting work that you have written yourself (or with your group) You are using your own UU credentials to submit the assignment You have not had outside help that violates the conditions delineated above while completing the assignment All assignments will be submitted via Ouriginal in Blackboard and, thereby, checked for plagiarism. If fraud or plagiarism is detected or suspected, we will inform the Board of Examiners in the usual manner. In the event of demonstrable fraud, the sanctions delineated in Article 5.15 of the Education and Examination Regulations (EER) will apply. "],["software-setup.html", "Software Setup", " Software Setup This chapter will help you prepare for the course by showing how to install R and RStudio on your computer. If you’re already using R, there may be nothing new for you here. That being said, you should look over this chapter to ensure that your current setup will be compatible with the course requirements. If you have never used R before, this chapter is essential! The information is this chapter will be crucial for getting your computer ready for the course. "],["typographic-conventions.html", "Typographic Conventions", " Typographic Conventions Throughout this GitBook, we (try to) use a consistent set of typographic conventions: Functions are typeset in a code font, and the name of the function is always followed by parentheses E.g., sum(), mean() Other R objects (e.g., data objects, function arguments) are in also typeset in a code font but without parentheses E.g., seTE, method.tau Sometimes, we’ll use the package name followed by two colons (::, the so-called *scope-resolution operator), like lavaan::sem(). This command is valid R code and will run if you copy it into your R console. The lavaan:: part of the command tells R that we want to use the sem() from the lavaan package. "],["installing-software.html", "Installing software", " Installing software Before we start the course, we have to install three things: R: A free program for statistical programming RStudio: An integrated development environment (IDE) which makes it easier to work with R. Several packages: Separate pieces of ‘add-on’ software for R with functions to do specific analyses. Packages also include documentation describing how to use their functions and sample data. Installing R The latest version of R is available here. Click the appropriate link for your operating system and follow the instructions for installing the latest stable release. Depending on which OS you select, you may be given an option to install different components (e.g., base, contrib, Rtools). For this course, you will only need the base package. Installing RStudio Download the Free Desktop version of RStudio from the download page of the RStudio website. Installing packages To participate in this course, you will need a few essential R packages. Here’s an overview of the packages and why we need them: Package Description lavaan A sophisticated and user-friendly package for structural equation modeling dplyr A powerful suite of data-processing tools ggplot2 A flexible and user-friendly package for making graphs tidySEM Plotting and tabulating the output of SEM-models semTools Comparing models, establishing measurement invariance across groups psych Descriptive statistics and EFA rockchalk Probing interactions foreign Loading data from SPSS ‘.sav’ files readxl Loading data from Excel ‘.xslx’ files To install these packages, we use the install.packages() function in R. Open RStudio Inside RStudio, find the window named Console on left side of the screen. Copy the following code into the console and hit Enter/Return to run the command. install.packages(c("lavaan", "dplyr", "ggplot2", "tidySEM", "semTools", "psych", "rockchalk", "foreign", "readxl"), dependencies = TRUE) "],["course-data.html", "Course Data", " Course Data All of the data files you will need for the course are available in this SurfDrive directory. Follow the link to download a ZIP archive containing the data you will need to complete the practical exercises and assignments. Extract these data files to a convenient location on your computer. "],["note-on-data-updates.html", "Note on Data Updates", " Note on Data Updates During the course, we may need to update some of these datasets and/or add some new datasets to the SurfDrive directory. If so, you will need to download the updated data. We will let you know if and when any datasets are modified. In such situations, you are responsible for updating your data. Working with outdated data will probably produce incorrect results. Your answer won’t match the solutions we expect. Your answer will be marked as incorrect, even if the code used to produce the answer is correct. Points lost on an assignment due to using outdated datasets will not be returned. "],["introduction-to-r.html", "1 Introduction to R", " 1 Introduction to R This week is all about getting up-and-running with R and RStudio. Homework before the lecture Complete the preparatory material: Read over the Course Information chapter Work through the Software Setup chapter Watch the Lecture Recording for this week. Homework before the practical Complete the At-Home Exercises. Practical content During the practical you will work on the In-Class Exercises. "],["lecture.html", "1.1 Lecture", " 1.1 Lecture This week, you will learn the basics of R and RStudio. Rather than re-inventing the proverbial wheel, we’re linked to existing resources developed by R-Ladies Sydney. 1.1.1 Recordings Tour of RStudio \\[\\\\[6pt]\\] R Packages \\[\\\\[6pt]\\] Data I/0 1.1.2 Slides You can access the accompanying resources on the R-Ladies Sydney website here. "],["reading.html", "1.2 Reading", " 1.2 Reading There is no official reading this week. If you’d like to deepen your dive into R, feel free to check out Hadley Wickham’s excellent book R for Data Science. Otherwise, you may want to get a jump-start on the At-Home Exercises for this week. \\[\\\\[12pt]\\] "],["at-home-exercises.html", "1.3 At-Home Exercises", " 1.3 At-Home Exercises This week is all about gaining familiarity with R and RStudio. We’ll be using the primers available on Posit Cloud to work through some basic elements of data visualization and statistical programming in R. Although you should already have R working, this week’s at-home and in-class exercises don’t require that you have R installed on your system. If following along within this GitBook doesn’t work for you, you can also find the tutorials online on the Posit Primers page. 1.3.1 Visualizations with R 1.3.2 Programming with R End of At-Home Exercises "],["in-class-exercises.html", "1.4 In-Class Exercises", " 1.4 In-Class Exercises In the practical this week, we’ll go a little further into what it’s possible with R. Don’t worry if you cannot remember everything in these primers—they’re only meant to familiarize you with what is possible and to get you some experience interacting with R and RStudio. The following primers come from Posit Cloud and were created with the learnr package. 1.4.1 Viewing Data This first primer introduces a special data format called a tibble, as well as some functions for viewing your data. 1.4.2 Dissecting Data In the next primer, we’ll explore tools to subset and rearrange you data: select(), filter(), and arrange(). 1.4.3 Grouping and Manipulating Data Advanced If you made it through the previous two sections with ease and want to challenge yourself, go ahead with this next section. If you’re running short on time, you can skip ahead to Exploratory Data Analysis. \\[\\\\[3pt]\\] 1.4.4 Exploratory Data Analysis 1.4.5 Visualizing Data Visualizing data is a great way to start understanding a data set. In this section, we’ll highlight a few examples of how you can use the ggplot2 libarary to visualize your data. Primers on many other visualizations are available on Posit Cloud. Bar Charts for Categorical Variables Scatterplots for Continuous Variables 1.4.6 Tidying Data This primer will provide an overview of what’s meant by “tidy data”. You only need to complete the Tidy Data section—the sections on Gathering and Spreading columns are useful, but we won’t ask you to apply those techniques in this course. Recap Hopefully, you now feel more comfortable using some of R’s basic functionality and packages to work with data. Here’s a brief description of the functions covered above: install.packages() for installing packages Remember to put the package names in quotes library() for loading packages View() for viewing your dataset select() for picking only certain columns filter() for picking only certain rows arrange() for changing the rows order %>% aka “the pipe” for chaining commands together In RStudio, you can hit ctrl+shift+m as a handy key combination ? for help files Logical tests and Boolean operators == equal to != not equal to < less than <= less than or equal to > greater than >= greater than or equal to is.na() is the value NA (not available) !is.na is the value not NA & and (true only if the left and right are both true) | or (true if either the left or right are true) ! not (invert true/false) %in% in (is left in the larger set of right values) any() any (true if any in the set are true) all() all (true if all in the set are true) xor() xor (true if one and only one of the set are true) ggplot2 ggplot() create the basic object from which to building a plot aes() contains the aesthetic mappings (like x and y) geom_bar() bar plots for distributions of categorical variables geom_point() scatterplots for plotting two continuous variables geom_label_repel() for plotting text facet_wrap() for creating sets of conditional plots End of In-Class Exercises "],["statistical-modeling-path-analysis.html", "2 Statistical Modeling & Path Analysis", " 2 Statistical Modeling & Path Analysis This week, we will cover statistical modeling and path analysis. Homework before the lecture Watch the Lecture Recording for this week. Complete the Reading for this week, and answer the associated reading questions. Homework before the practical Complete the At-Home Exercises. Practical content During the practical you will work on the In-Class Exercises. "],["lecture-1.html", "2.1 Lecture", " 2.1 Lecture In this lecture, we will begin by discussing the paradigm and contextualizing statistical modeling relative to other ways that we can conduct statistical analyses. We will conclude with an introduction to . 2.1.1 Recordings Statistical Reasoning Statistical Modeling Path Analysis 2.1.2 Slides You can download the lectures slides here "],["reading-1.html", "2.2 Reading", " 2.2 Reading Reference Smaldino, P. E. (2017). Models are stupid, and we need more of them. In R.R. Vallacher, S.J. Read, & A. Nowakt (Eds.), Computational Social Psychology (pp. 311–331). New York: Routledge. SKIP PAGES 322 - 327 Questions What are the differences between a “verbal model” and a “formal model”? As explained in the paragraph “A Brief Note on Statistical Models”, formal models are not the same as statistical models. Still, we can learn a lot from Smaldino’s approach. Write down three insights from this paper that you would like to apply to your statistical modeling during this course. Suggested Reading (Optional) The following paper is not required, but it’s definitely worth a read. Breiman provides a very interesting perspective on different ways to approach a modeling-based analysis. Breiman, L. (2001). Statistical Modeling: The Two Cultures (with comments and a rejoinder by the author). Statistical Science, 16(3) 199–231. https://doi.org/10.1214/ss/1009213726 "],["at-home-exercises-1.html", "2.3 At-Home Exercises", " 2.3 At-Home Exercises Load the LifeSat.sav data. library(dplyr) library(haven) LifeSat <- read_spss("LifeSat.sav") 2.3.1 Make a table of descriptive statistics for the variables: LifSat, educ, ChildSup, SpouSup, and age. What is the average age in the sample? What is the range (youngest and oldest child)? Hint: Use the tidySEM::descriptives() function.` Click for explanation The package tidySEM contains the descriptives() function for computing descriptive statistics. The describe() function in the psych package is a good alternative. library(tidySEM) descriptives(LifeSat[ , c("LifSat", "educ", "ChildSup", "SpouSup", "age")]) 2.3.2 Run a simple linear regression with LifSat as the dependent variable and educ as the independent variable. Hints: The lm() function (short for linear model) does linear regression. The summary() function provides relevant summary statistics for the model. It can be helpful to store the results of your analysis in an object. Click for explanation results <- lm(LifSat ~ educ, data = LifeSat) summary(results) ## ## Call: ## lm(formula = LifSat ~ educ, data = LifeSat) ## ## Residuals: ## Min 1Q Median 3Q Max ## -43.781 -11.866 2.018 12.418 43.018 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) 35.184 7.874 4.469 2.15e-05 *** ## educ 3.466 1.173 2.956 0.00392 ** ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 17.64 on 96 degrees of freedom ## Multiple R-squared: 0.08344, Adjusted R-squared: 0.0739 ## F-statistic: 8.74 on 1 and 96 DF, p-value: 0.003918 2.3.3 Repeat the analysis from 2.3.2 with age as the independent variable. Click for explanation results <- lm(LifSat ~ age, data = LifeSat) summary(results) ## ## Call: ## lm(formula = LifSat ~ age, data = LifeSat) ## ## Residuals: ## Min 1Q Median 3Q Max ## -35.321 -14.184 3.192 13.593 40.626 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) 200.2302 52.1385 3.840 0.00022 *** ## age -2.0265 0.7417 -2.732 0.00749 ** ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 17.75 on 96 degrees of freedom ## Multiple R-squared: 0.07215, Adjusted R-squared: 0.06249 ## F-statistic: 7.465 on 1 and 96 DF, p-value: 0.007487 2.3.4 Repeat the analysis from 2.3.2 and 2.3.3 with ChildSup as the independent variable. Click for explanation results <- lm(LifSat ~ ChildSup, data = LifeSat) summary(results) ## ## Call: ## lm(formula = LifSat ~ ChildSup, data = LifeSat) ## ## Residuals: ## Min 1Q Median 3Q Max ## -37.32 -12.14 0.66 12.41 44.68 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) 37.559 8.342 4.502 1.89e-05 *** ## ChildSup 2.960 1.188 2.492 0.0144 * ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 17.86 on 96 degrees of freedom ## Multiple R-squared: 0.06076, Adjusted R-squared: 0.05098 ## F-statistic: 6.211 on 1 and 96 DF, p-value: 0.01441 2.3.5 Run a multiple linear regression with LifSat as the dependent variable and educ, age, and ChildSup as the independent variables. Hint: You can use the + sign to add multiple variables to the RHS of your model formula. Click for explanation results <- lm(LifSat ~ educ + age + ChildSup, data = LifeSat) summary(results) ## ## Call: ## lm(formula = LifSat ~ educ + age + ChildSup, data = LifeSat) ## ## Residuals: ## Min 1Q Median 3Q Max ## -32.98 -12.56 2.68 11.03 41.91 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) 134.9801 53.2798 2.533 0.0130 * ## educ 2.8171 1.1436 2.463 0.0156 * ## age -1.5952 0.7188 -2.219 0.0289 * ## ChildSup 2.4092 1.1361 2.121 0.0366 * ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 16.92 on 94 degrees of freedom ## Multiple R-squared: 0.1741, Adjusted R-squared: 0.1477 ## F-statistic: 6.603 on 3 and 94 DF, p-value: 0.0004254 2.3.6 Compare the results from 2.3.5 with those from 2.3.2, 2.3.3, and 2.3.4. What do you notice when you compare the estimated slopes for each of the three predictors in the multiple regression model with the corresponding estimates from the simple regression models? "],["in-class-exercises-1.html", "2.4 In-Class Exercises", " 2.4 In-Class Exercises During this practical, you will work through some exercises meant to expand your statistical reasoning skills and improve your understanding of linear models. For this exercise, having some familiarity with regression will be helpful. If you feel like you need to refresh your knowledge in this area, consider the resources listed in the Background knowledge section. Data: You will use the following dataset for these exercises. Sesam.sav 2.4.1 Data Exploration Open the file “Sesam.sav” # Load `dplyr` for data processing: library(dplyr) # Load the `haven` library for reading in SPSS files: library(haven) ## Load the 'Sesam.sav' data ## Use haven::zap_formats() to remove SPSS attributes sesam <- read_sav(file = "Sesam.sav") %>% zap_formats() This file is part of a larger dataset that evaluates the impact of the first year of the Sesame Street television series. Sesame Street is mainly concerned with teaching preschool related skills to children in the 3–5 year age range. The following variables will be used in this exercise: age: measured in months prelet: knowledge of letters before watching Sesame Street (range 0–58) prenumb: knowledge of numbers before watching Sesame Street (range 0–54) prerelat: knowledge of size/amount/position relationships before watching Sesame Street (range 0–17) peabody: vocabulary maturity before watching Sesame Street (range 20–120) postnumb: knowledge of numbers after a year of Sesame Street (range 0–54) Note: Unless stated otherwise, the following questions refer to the sesam data and the above variables. 2.4.1.1 What is the type of each variable? Hint: The output of the str() function should be helpful here. Click to show code ## Examine the data structure: str(sesam) ## tibble [240 × 8] (S3: tbl_df/tbl/data.frame) ## $ id : num [1:240] 1 2 3 4 5 6 7 8 9 10 ... ## $ age : num [1:240] 66 67 56 49 69 54 47 51 69 53 ... ## $ prelet : num [1:240] 23 26 14 11 47 26 12 48 44 38 ... ## $ prenumb : num [1:240] 40 39 9 14 51 33 13 52 42 31 ... ## $ prerelat: num [1:240] 14 16 9 9 17 14 11 15 15 10 ... ## $ peabody : num [1:240] 62 80 32 27 71 32 28 38 49 32 ... ## $ postnumb: num [1:240] 44 39 40 19 54 39 44 51 48 52 ... ## $ gain : num [1:240] 4 0 31 5 3 6 31 -1 6 21 ... ## ..- attr(*, "display_width")= int 10 Click for explanation All variables are numeric. str() uses the abbreviation “num” to indicate a numeric vector. 2.4.1.2 What is the average age in the sample? What is the age range (youngest and oldest child)? Hint: Use tidySEM::descriptives() Click to show code As in the take home exercises, you can use the descriptives() function from the tidySEM package to describe the data: library(tidySEM) descriptives(sesam) Click for explanation We can get the average age from the “mean” column in the table ( 51.5), and the age range from the columns “min” and “max”, (34 and 69 respectively.) 2.4.1.3 What is the average gain in knowledge of numbers? What is the standard deviation of this gain? Hints: You will need to compute the gain and save the change score as a new object. You can then use the base-R functions mean() and sd() to do the calculations. Click to show code Create a new variable that represents the difference between pre- and post-test scores on knowledge of numbers: sesam <- mutate(sesam, ndif = postnumb - prenumb) Compute the mean and SD of the change score: sesam %>% summarise(mean(ndif), sd(ndif)) 2.4.1.4 Create an appropriate visualization of the gain scores you computed in 2.4.1.3. Justify your choice of visualization. Hint: Some applicable visualizations are explained in the Visualizations with R section. Click to show code library(ggplot2) ## Create an empty baseline plot object: p <- ggplot(sesam, aes(x = ndif)) ## Add some appropriate geoms: p + geom_histogram() p + geom_density() p + geom_boxplot() Click for explanation Because the gain score is numeric, we should use something appropriate for showing the distribution of a continuous variable. In this case, we can use either a density plot, or a histogram (remember from the lecture, this is like a density plot, but binned). We can also use a box plot, which can be a concise way to display a lot of information about a variable in a little less space. 2.4.1.5 Create a visualization that provides information about the bivariate relationship between the pre- and post-test number knowledge. Justify your choice of visualization. Describe the relationship based on what you see in your visualization. Hint: Again, the Visualizations with R section may provide some useful insights. Click to show code ## Create a scatterplot of the pre- and post-test number knowledge ggplot(sesam, aes(x = prenumb, y = postnumb)) + geom_point() Click for explanation A scatterplot is a good tool for showing patterns in the way that two continuous variables relate to each other. From it, we can quickly gather information about whether a relationship exists, its direction, its strength, how much variation there is, and whether or not a relationship might be non-linear. Based on this scatterplot, we see a positive relationship between the prior knowledge of numbers and the knowledge of numbers at the end of the study. Children who started with a higher level of numeracy also ended with a higher level of numeracy. There is a considerable amount of variance in the relationship. Not every child increases their numeracy between pre-test and post-test. Children show differing amounts of increase. 2.4.2 Linear Modeling 2.4.2.1 Are there significant, bivariate associations between postnumb and the following variables? age prelet prenumb prerelat peabody Use Pearson correlations to answer this question. You do not need to check the assumptions here (though you would in real life). Hint: The base-R cor.test() function and the corr.test() function from the psych package will both conduct hypothesis tests for a correlation coefficients (the base-R cor() function only computes the coefficients). Click to show code library(psych) ## Test the correlations using psych::corr.test(): sesam %>% select(postnumb, age, prelet, prenumb, prerelat, peabody) %>% corr.test() ## Call:corr.test(x = .) ## Correlation matrix ## postnumb age prelet prenumb prerelat peabody ## postnumb 1.00 0.34 0.50 0.68 0.54 0.52 ## age 0.34 1.00 0.33 0.43 0.44 0.29 ## prelet 0.50 0.33 1.00 0.72 0.47 0.40 ## prenumb 0.68 0.43 0.72 1.00 0.72 0.61 ## prerelat 0.54 0.44 0.47 0.72 1.00 0.56 ## peabody 0.52 0.29 0.40 0.61 0.56 1.00 ## Sample Size ## [1] 240 ## Probability values (Entries above the diagonal are adjusted for multiple tests.) ## postnumb age prelet prenumb prerelat peabody ## postnumb 0 0 0 0 0 0 ## age 0 0 0 0 0 0 ## prelet 0 0 0 0 0 0 ## prenumb 0 0 0 0 0 0 ## prerelat 0 0 0 0 0 0 ## peabody 0 0 0 0 0 0 ## ## To see confidence intervals of the correlations, print with the short=FALSE option ## OR ## library(magrittr) ## Test the correlations using multiple cor.test() calls: sesam %$% cor.test(postnumb, age) ## ## Pearson's product-moment correlation ## ## data: postnumb and age ## t = 5.5972, df = 238, p-value = 5.979e-08 ## alternative hypothesis: true correlation is not equal to 0 ## 95 percent confidence interval: ## 0.2241066 0.4483253 ## sample estimates: ## cor ## 0.3410578 sesam %$% cor.test(postnumb, prelet) ## ## Pearson's product-moment correlation ## ## data: postnumb and prelet ## t = 8.9986, df = 238, p-value < 2.2e-16 ## alternative hypothesis: true correlation is not equal to 0 ## 95 percent confidence interval: ## 0.4029239 0.5926632 ## sample estimates: ## cor ## 0.5038464 sesam %$% cor.test(postnumb, prenumb) ## ## Pearson's product-moment correlation ## ## data: postnumb and prenumb ## t = 14.133, df = 238, p-value < 2.2e-16 ## alternative hypothesis: true correlation is not equal to 0 ## 95 percent confidence interval: ## 0.6002172 0.7389277 ## sample estimates: ## cor ## 0.6755051 sesam %$% cor.test(postnumb, prerelat) ## ## Pearson's product-moment correlation ## ## data: postnumb and prerelat ## t = 9.9857, df = 238, p-value < 2.2e-16 ## alternative hypothesis: true correlation is not equal to 0 ## 95 percent confidence interval: ## 0.4475469 0.6268773 ## sample estimates: ## cor ## 0.5433818 sesam %$% cor.test(postnumb, peabody) ## ## Pearson's product-moment correlation ## ## data: postnumb and peabody ## t = 9.395, df = 238, p-value < 2.2e-16 ## alternative hypothesis: true correlation is not equal to 0 ## 95 percent confidence interval: ## 0.4212427 0.6067923 ## sample estimates: ## cor ## 0.520128 Click for explanation Yes, based on the p-values (remember that 0 here really means very small, making it less than .05), we would say that there are significant correlations between postnumb and all other variables in the data. (In fact, all variables in the data are significantly correlated with one another.) 2.4.2.2 Do age and prenumb explain a significant proportion of the variance in postnumb? What statistic did you use to justify your conclusion? Interpret the model fit. Use the lm() function to fit your model. Click to show code lmOut <- lm(postnumb ~ age + prenumb, data = sesam) summary(lmOut) ## ## Call: ## lm(formula = postnumb ~ age + prenumb, data = sesam) ## ## Residuals: ## Min 1Q Median 3Q Max ## -38.130 -6.456 -0.456 5.435 22.568 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) 7.4242 5.1854 1.432 0.154 ## age 0.1225 0.1084 1.131 0.259 ## prenumb 0.7809 0.0637 12.259 <2e-16 *** ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 9.486 on 237 degrees of freedom ## Multiple R-squared: 0.4592, Adjusted R-squared: 0.4547 ## F-statistic: 100.6 on 2 and 237 DF, p-value: < 2.2e-16 Click for explanation Yes, age and prenumb explain a significant amount of variability in postnumb (\\(R^2 = 0.459\\), \\(F[2, 237] = 100.629\\), \\(p < 0.001\\)). We use the F statistic for the overall test of model fit to support this conclusion. The variables age and prenumb together explain 45.9% of the variability in postnumb. 2.4.2.3 Write the null and alternative hypotheses tested for in 2.4.2.2. Click for explanation Since we are testing for explained variance, our hypotheses concern the \\(R^2\\). \\[ \\begin{align*} H_0: R^2 = 0\\\\ H_1: R^2 > 0 \\end{align*} \\] Note that this is a directional hypotheses because the \\(R^2\\) cannot be negative. 2.4.2.4 Define the model syntax to estimate the model from 2.4.2.2 as a path analysis using lavaan. Click to show code mod <- 'postnumb ~ 1 + age + prenumb' 2.4.2.5 Estimate the path analytic model you defined above. Use the lavaan::sem() function to estimate the model. Click to show code library(lavaan) lavOut1 <- sem(mod, data = sesam) 2.4.2.6 Summarize the fitted model you estimated above. Use the summary() function to summarize the model. Click to show code summary(lavOut1) ## lavaan 0.6.16 ended normally after 1 iteration ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 4 ## ## Number of observations 240 ## ## Model Test User Model: ## ## Test statistic 0.000 ## Degrees of freedom 0 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## postnumb ~ ## age 0.123 0.108 1.138 0.255 ## prenumb 0.781 0.063 12.336 0.000 ## ## Intercepts: ## Estimate Std.Err z-value P(>|z|) ## .postnumb 7.424 5.153 1.441 0.150 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .postnumb 88.864 8.112 10.954 0.000 In OLS regression, the predictor variables are usually treated as fixed and do not covary. We can easily relax this assumption in path analysis. 2.4.2.7 Re-estimate the path analytic model you defined in 2.4.2.4. Specify the predictors as random, correlated variables. Hint: You can make the predictors random in, at least, two ways: Modify the model syntax to specify the correlation between age and prenumb. Add fixed.x = FALSE to your sem() call. Click to show code lavOut2 <- sem(mod, data = sesam, fixed.x = FALSE) ## OR ## mod <- ' postnumb ~ 1 + age + prenumb age ~~ prenumb ' lavOut2 <- sem(mod, data = sesam) 2.4.2.8 Summarize the fitted model you estimated above. Compare the results to those from the OLS regression in 2.4.2.2 and the path model in 2.4.2.5. Click to show code summary(lavOut2) ## lavaan 0.6.16 ended normally after 26 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 9 ## ## Number of observations 240 ## ## Model Test User Model: ## ## Test statistic 0.000 ## Degrees of freedom 0 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## postnumb ~ ## age 0.123 0.108 1.138 0.255 ## prenumb 0.781 0.063 12.336 0.000 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ## age ~~ ## prenumb 28.930 4.701 6.154 0.000 ## ## Intercepts: ## Estimate Std.Err z-value P(>|z|) ## .postnumb 7.424 5.153 1.441 0.150 ## age 51.525 0.405 127.344 0.000 ## prenumb 20.896 0.688 30.359 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .postnumb 88.864 8.112 10.954 0.000 ## age 39.291 3.587 10.954 0.000 ## prenumb 113.702 10.379 10.954 0.000 summary(lavOut1) ## lavaan 0.6.16 ended normally after 1 iteration ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 4 ## ## Number of observations 240 ## ## Model Test User Model: ## ## Test statistic 0.000 ## Degrees of freedom 0 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## postnumb ~ ## age 0.123 0.108 1.138 0.255 ## prenumb 0.781 0.063 12.336 0.000 ## ## Intercepts: ## Estimate Std.Err z-value P(>|z|) ## .postnumb 7.424 5.153 1.441 0.150 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .postnumb 88.864 8.112 10.954 0.000 summary(lmOut) ## ## Call: ## lm(formula = postnumb ~ age + prenumb, data = sesam) ## ## Residuals: ## Min 1Q Median 3Q Max ## -38.130 -6.456 -0.456 5.435 22.568 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) 7.4242 5.1854 1.432 0.154 ## age 0.1225 0.1084 1.131 0.259 ## prenumb 0.7809 0.0637 12.259 <2e-16 *** ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 9.486 on 237 degrees of freedom ## Multiple R-squared: 0.4592, Adjusted R-squared: 0.4547 ## F-statistic: 100.6 on 2 and 237 DF, p-value: < 2.2e-16 2.4.2.9 Consider the path model below. How many regression coefficients are estimated in this model? How many variances are estimated? How many covariances are estimated? Click for explanation Six regression coefficients (red) Four (residual) variances (blue) No covariances 2.4.2.10 Consider a multiple regression analysis with three continuous independent variables: scores on tests of language, history, and logic, and one continuous dependent variable: score on a math test. We want to know if scores on the language, history, and logic tests can predict the math test score. Sketch a path model that you could use to answer this question How many regression parameters are there? How many variances could you estimate? How many covariances could you estimate? 2.4.3 Categorical IVs Load the Drivers.sav data. # Read the data into a data frame named 'drivers': drivers <- read_sav("Drivers.sav") %>% as_factor() # This preserves the SPSS labels for nominal variables In this section, we will evaluate the following research question: Does talking on the phone interfere with people's driving skills? These data come from an experiment. The condition variable represents the three experimental conditions: Hand-held phone Hands-free phone Control (no phone) We will use condition as the IV in our models. The DV, RT, represents the participant’s reaction time (in milliseconds) during a driving simulation. 2.4.3.1 Use the package ggplot2 to create a density plot for the variable RT. What concept are we representing with this plot? Hint: Consider the lap times example from the statistical modeling section of Lecture 2. Click to show code ggplot(drivers, aes(x = RT)) + geom_density() Click for explanation This shows the distribution of all the combined reaction times from drivers in all three categories. 2.4.3.2 Modify this density plot by mapping the variable condition from your data to the fill aesthetic in ggplot. What is the difference between this plot and the previous plot? Do you think there is evidence for differences between the groups? How might we test this by fitting a model to our sample? Click to show code Hint: To modify the transparency of the densities, use the aesthetic alpha. ggplot(drivers, aes(x = RT, fill = condition)) + geom_density(alpha = .5) Click for explanation This figure models the conditional distribution of reaction time, where the type of cell phone usage is the grouping factor. Things you can look at to visually assess whether the three groups differ are the amount of overlap of the distributions, how much distance there is between the individual means, and whether the combined distribution is much different than the conditional distributions. If we are willing to assume that these conditional distributions are normally distributed and have equivalent variances, we could use a linear model with dummy-coded predictors. Aside: ANOVA vs. Linear Regression As you may know, the mathematical model underlying ANOVA is just a linear regression model with nominal IVs. So, in terms of the underlying statistical models, there is no difference between ANOVA and regression; the differences lie in the focus of the analysis. ANOVA is really a type of statistical test wherein we are testing hypotheses about the effects of some set of nominal grouping factors on some continuous outcome. When doing an ANOVA, we usually don’t interact directly with the parameter estimates from the underlying model. Regression is a type of statistical model (i.e., a way to represent a univariate distribution with a conditional mean and fixed variance). When we do a regression analysis, we primarily focus on the estimated parameters of the underling linear model. When doing ANOVA in R, we estimate the model exactly as we would for linear regression; we simply summarize the results differently. If you want to summarize your model in terms of the sums of squares table you usually see when running an ANOVA, you can supply your fitted lm object to the anova() function. This is a statistical modeling course, not a statistical testing course, so we will not consider ANOVA any further. 2.4.3.3 Estimate a linear model that will answer the research question stated in the beginning of this section. Use lm() to estimate the model. Summarize the fitted model and use the results to answer the research question. Click to show code library(magrittr) lmOut <- drivers %>% mutate(condition = relevel(condition, ref = "control")) %$% lm(RT ~ condition) summary(lmOut) ## ## Call: ## lm(formula = RT ~ condition) ## ## Residuals: ## Min 1Q Median 3Q Max ## -317.50 -71.25 2.98 89.55 243.45 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) 553.75 29.08 19.042 <2e-16 *** ## conditionhand-held 100.75 41.13 2.450 0.0174 * ## conditionhands-free 63.80 41.13 1.551 0.1264 ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 130.1 on 57 degrees of freedom ## Multiple R-squared: 0.09729, Adjusted R-squared: 0.06562 ## F-statistic: 3.072 on 2 and 57 DF, p-value: 0.05408 anova(lmOut) Click for explanation The effect of condition on RT is nonsignificant (\\(F[2, 57] = 3.07\\), \\(p = 0.054\\)). Therefore, based on these results, we do not have evidence for an effect of mobile phone usage on driving performance. 2.4.3.4 Use lavaan to estimate the model from 2.4.3.3 as a path model. Hint: lavaan won’t let us use factors for our categorical predictors. So, you will need to create your own dummy codes. Click to show code mod <- 'RT ~ 1 + HH + HF' lavOut <- drivers %>% mutate(HH = ifelse(condition == "hand-held", 1, 0), # Create dummy code for "hand-held" condition HF = ifelse(condition == "hands-free", 1, 0) # Create dummy code for "hands-free" condition ) %>% sem(mod, data = .) # Estimate the model summary(lavOut) ## lavaan 0.6.16 ended normally after 1 iteration ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 4 ## ## Number of observations 60 ## ## Model Test User Model: ## ## Test statistic 0.000 ## Degrees of freedom 0 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## RT ~ ## HH 100.750 40.085 2.513 0.012 ## HF 63.800 40.085 1.592 0.111 ## ## Intercepts: ## Estimate Std.Err z-value P(>|z|) ## .RT 553.750 28.344 19.537 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .RT 16068.028 2933.607 5.477 0.000 At this point, we haven’t covered the tools you need to conduct the ANOVA-style tests with path models. So, you can’t yet answer the research question with the above model. When we discuss model comparisons, you’ll get the missing tools. End of In-Class Exercises 2 "],["mediation-moderation.html", "3 Mediation & Moderation", " 3 Mediation & Moderation In this lecture, we will discuss two particular types of processes that we can model using path analysis: mediation and moderation. Homework before the lecture Watch the Lecture Recording for this week. Complete the Reading for this week, and answer the associated reading questions. Homework before the practical Complete the At-Home Exercises. Practical content During the practical you will work on the In-Class Exercises. "],["lecture-2.html", "3.1 Lecture", " 3.1 Lecture Researchers often have theories about possible causal processes linking multiple variables. Mediation is a particularly important example of such a process in which in an input variable, X, influences the outcome, Y, through an intermediary variable, M (the mediator). For instance, psychotherapy (X), may affect thoughts (M), which in turn affects mood (Y). We can investigate mediation via a specific sequence of linear regression equations, but path modeling will make our lives much easier. We can use path models to simultaneously estimate multiple related regression equations. So, mediation analysis is an ideal application of path modeling. In this lecture, we consider both approaches and discuss their relative strengths and weaknesses. As with mediation, researchers often posit theories involving moderation. Moderation implies that the effect of X on Y depends on another variable, Z. For instance, the effect of feedback (X) on performance (Y) may depend on age (Z). Older children might process feedback more effectively than younger children. Hence, the feedback is more effective for older children than for younger children, and the effect of feedback on performance is stronger for older children than for younger children. In such a case, we would say that age moderates the effect of feedback on performance. 3.1.1 Recordings Note: In the following recordings, the slide numbers are a bit of a mess, because I made these videos by cutting together recordings that used different slide decks. My apologies to those who are particularly distracted by continuity errors. Mediation Basics Mediation Testing Bootstrapping Moderation Basics Moderation Probing 3.1.2 Slides You can download the lecture slides here "],["reading-2.html", "3.2 Reading", " 3.2 Reading Reference Baron, R. M. & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical Considerations. Journal of Personality and Individual Differences, 51(6), 1173–1182 Questions What is mediation? Give an example of mediation. According to the authors, we must satisfy four criteria to infer mediation. What are these criteria? What is “moderation”, and how is it different from “mediation”? Give an example of moderation. What are the four methods given by Baron and Kenny as suitable ways to to study interaction effects? The authors suggest that one of the most common ways to address unreliability is to use multiple indicators. Thinking back to what you’ve learned about factor analysis, briefly explain why multiple indicators can improve reliability. How can you determine whether a variable is a mediator or moderator? Reference Hayes, A. F. (2009). Beyond Baron and Kenny: Statistical mediation analysis in the new millennium. Communication Monographs, 76(4), 408–420. Questions What is an indirect or mediated effect? What is the difference between the total and direct effect? What is the main problem with the Barron & Kenny “Causal Steps Approach”? What is bootstrapping, and why is it a better way to test mediation than Sobel’s test? Explain how it is possible that “effects that don’t exist can be mediated”. "],["at-home-exercises-2.html", "3.3 At-Home Exercises", " 3.3 At-Home Exercises 3.3.1 Mediation In the first part of this practical, we will analyze the data contained in SelfEsteem.sav. These data comprise 143 observations of the following variables.1 case: Participant ID number ParAtt: Parental Attachment PeerAtt: Peer Attachment Emp: Empathy ProSoc: Prosocial behavior Aggr: Aggression SelfEst: Self-esteem 3.3.1.1 Load the SelfEsteem.sav data. Note: Unless otherwise specified, all analyses in Section 3.3.1 apply to these data. Click to show code library(haven) seData <- read_sav("SelfEsteem.sav") Suppose we are interested in the (indirect) effect of peer attachment on self-esteem, and whether empathy has a mediating effect on this relationship. We might generate the following hypotheses: Better peer relationships promote higher self esteem This effect is mediated by a student’s empathy levels, where better peer relationships increase empathy, and higher levels of empathy lead to higher self-esteem. To evaluate these hypotheses, we will use lavaan to estimate a path model. 3.3.1.2 Draw a path model (on paper) that can be used to test the above hypotheses. Label the input (X), outcome (Y), and mediator/intermediary (M). Label the paths a, b, and c’. Hint: Refer back to the Mediation Basics lecture if you need help here. Click for explanation 3.3.1.3 Specify the lavaan model syntax implied by the path diagram shown above. Save the resulting character string as an object in your environment. Hint: Refer back to the example in which opinions of systematic racism mediate the relationship between political affiliation and support for affirmative action policies from the Mediation Testing lecture this week. Click to show code mod <- ' ## Equation for outcome: SelfEst ~ Emp + PeerAtt ## Equation for the mediator: Emp ~ PeerAtt ' 3.3.1.4 Use the lavaan::sem() function to estimate the model defined in 3.3.1.3. Use the default settings in sem(). Click to show code library(lavaan) out <- sem(mod, data = seData) 3.3.1.5 Explore the summary of the fitted model. Which numbers correspond to the a, b, and c’ paths? Interpret these paths. Do the direction of the effects seem to align with our hypothesis? Click to show code summary(out) ## lavaan 0.6.16 ended normally after 1 iteration ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 5 ## ## Number of observations 143 ## ## Model Test User Model: ## ## Test statistic 0.000 ## Degrees of freedom 0 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## SelfEst ~ ## Emp 0.234 0.091 2.568 0.010 ## PeerAtt 0.174 0.088 1.968 0.049 ## Emp ~ ## PeerAtt 0.349 0.076 4.628 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .SelfEst 0.934 0.110 8.456 0.000 ## .Emp 0.785 0.093 8.456 0.000 Click for explanation The results show estimates of the a path (Emp ~ PeerAtt), the b path (SelfEst ~ Emp), and the c’ path (SelfEst ~ PeerAtt). All three of these effects are positive and significant, including the direct effect of PeerAtt on SelfEst (\\(\\beta = 0.174\\), \\(Z = 1.97\\), \\(p = 0.025\\)), and the parts of the indirect effect made up by the effect of PeerAtt on Emp (\\(\\beta = 0.349\\), \\(Z = 4.63\\), \\(p = 0\\)), and Emp on SelfEst (\\(\\beta = 0.234\\), \\(Z = 2.57\\), \\(p = 0.005\\)). We can see that the direction of the effects seems to support of our hypotheses, but without taking the next steps to investigate the indirect effect, we should be hesitant to say more. Remember that an indirect effect (IE) is the product of multiple regression slopes. Therefore, to estimate an IE, we must define this product in our model syntax. In lavaan, we define the new IE parameter in two steps. Label the relevant regression paths. Use the labels to define a new parameter that represent the desired IE. We can define new parameters in lavaan model syntax via the := operator. The lavaan website contains a tutorial on this procedure: http://lavaan.ugent.be/tutorial/mediation.html 3.3.1.6 Use the procedure described above to modify the model syntax from 3.3.1.3 by adding the definition of the hypothesized IE from PeerAtt to SelfEst. Click to show code mod <- ' ## Equation for outcome: SelfEst ~ b * Emp + PeerAtt ## Equation for mediator: Emp ~ a * PeerAtt ## Indirect effect: ie := a * b ' Click for explanation Notice that I only label the parameters that I will use to define the IE. You are free to label any parameter that you like, but I choose the to label only the minimally sufficient set to avoid cluttering the code/output. 3.3.1.7 Use lavaan::sem() to estimate the model with the IEs defined. Use the default settings for sem(). Is the hypothesized IE significant according to the default tests? Hint: Refer to the Mediation Testing lecture Click to show code out <- sem(mod, data = seData) summary(out) ## lavaan 0.6.16 ended normally after 1 iteration ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 5 ## ## Number of observations 143 ## ## Model Test User Model: ## ## Test statistic 0.000 ## Degrees of freedom 0 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## SelfEst ~ ## Emp (b) 0.234 0.091 2.568 0.010 ## PeerAtt 0.174 0.088 1.968 0.049 ## Emp ~ ## PeerAtt (a) 0.349 0.076 4.628 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .SelfEst 0.934 0.110 8.456 0.000 ## .Emp 0.785 0.093 8.456 0.000 ## ## Defined Parameters: ## Estimate Std.Err z-value P(>|z|) ## ie 0.082 0.036 2.245 0.025 Click for explanation The IE of Peer Attachment on Self Esteem through Empathy is statistically significant (\\(\\hat{\\textit{IE}} = 0.082\\), \\(Z = 2.25\\), \\(p = 0.012\\)). Note: The p-value above doesn’t match the output because we’re testing a directional hypothesis, but lavaan conducts two-tailed tests for the model parameters. As we learned in the lecture, the above test of the indirect effect is equivalent to Sobel’s Z test (which we don’t really want). An appropriate, robust test of the indirect effect requires bootstrapping, which we will do later this week as part of the in-class exercises. For now, we’ll add another input variable to our model: parental attachment. We will use this model to evaluate the following research questions: Is there a direct effect of parental attachment on self-esteem, after controlling for peer attachment and empathy? Is there a direct effect of peer attachment on self-esteem, after controlling for parental attachment and empathy? Is the effect of parental attachment on self-esteem mediated by empathy, after controlling for peer attachment? Is the effect of peer attachment on self-esteem mediated by empathy, after controlling for parental attachment? 3.3.1.8 Run the path model needed to test the research questions listed above. Specify the lavaan model syntax implied by the research questions. Allow peer attachment and parental attachment to covary. Define two new parameters to represent the hypothesized indirect effects. Estimate the model using lavaan::sem(). Use the default settings in sem(). Investigate the model summary. Click to show code mod <- ' ## Equation for outcome: SelfEst ~ b * Emp + ParAtt + PeerAtt ## Equation for mediator: Emp ~ a1 * ParAtt + a2 * PeerAtt ## Covariance: ParAtt ~~ PeerAtt ie_ParAtt := a1 * b ie_PeerAtt := a2 * b ' out <- sem(mod, data = seData) summary(out) ## lavaan 0.6.16 ended normally after 16 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 10 ## ## Number of observations 143 ## ## Model Test User Model: ## ## Test statistic 0.000 ## Degrees of freedom 0 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## SelfEst ~ ## Emp (b) 0.206 0.088 2.357 0.018 ## ParAtt 0.287 0.078 3.650 0.000 ## PeerAtt 0.024 0.094 0.252 0.801 ## Emp ~ ## ParAtt (a1) 0.078 0.075 1.045 0.296 ## PeerAtt (a2) 0.306 0.086 3.557 0.000 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ## ParAtt ~~ ## PeerAtt 0.537 0.103 5.215 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .SelfEst 0.854 0.101 8.456 0.000 ## .Emp 0.779 0.092 8.456 0.000 ## ParAtt 1.277 0.151 8.456 0.000 ## PeerAtt 0.963 0.114 8.456 0.000 ## ## Defined Parameters: ## Estimate Std.Err z-value P(>|z|) ## ie_ParAtt 0.016 0.017 0.956 0.339 ## ie_PeerAtt 0.063 0.032 1.965 0.049 3.3.1.9 What can we say about the two indirect effects? Can we say that empathy mediates both paths? Click to show explanation According to the Sobel-style test, after controlling for parental attachment, the indirect effect of peer attachment on self-esteem was statistically significant (\\(\\hat{IE} = 0.063\\), \\(Z = 1.96\\), \\(p = 0.049\\)), as was the analogous direct effect (\\(\\hat{\\beta} = 0.306\\), \\(Z = 3.56\\), \\(p < 0.001\\)). After controlling for peer attachment, neither the indirect effect (\\(\\hat{IE} = 0.016\\), \\(Z = 0.96\\), \\(p = 0.339\\)) nor the direct effect (\\(\\hat{\\beta} = 0.078\\), \\(Z = 1.05\\), \\(p = 0.296\\)) of parental attachment on self-esteem was significant, though. 3.3.2 Moderation Remember that moderation attempts to describe when one variable influences another. For the home exercise, we’ll go back to the Sesame Street data we worked with for the in-class exercises last week. 3.3.2.1 Load the Sesam2.sav data.2 NOTE: Unless otherwise specified, all analyses in Section 3.3.2 use these data. Click to show code # Read the data into an object called 'sesam2': sesam2 <- read_sav("Sesam2.sav") VIEWCAT is a nominal grouping variable, but it is represented as a numeric variable in the sesam2 data. The levels represent the following frequencies of Sesame Street viewership of the children in the data: VIEWCAT = 1: Rarely/Never VIEWCAT = 2: 2–3 times a week VIEWCAT = 3: 4–5 times a week VIEWCAT = 4: > 5 times a week 3.3.2.2 Convert VIEWCAT into a factor. Make sure that VIEWCAT = 1 is the reference group. Hints: You can identify the reference group with the levels() or contrasts() functions. The reference group is the group labelled with the first level printed by levels(). When you run contrasts(), you will see a pattern matrix that defines a certain dummy coding scheme. The reference group is the group that has zeros in each column of this matrix. If you need to change the reference group, you can use the relevel() function. Click to show code library(forcats) ## Convert 'VIEWCAT' to a factor: sesam2 <- sesam2 %>% mutate(VIEWCAT = factor(VIEWCAT)) ## Optionally specify the labels # sesam2 <- # sesam2 %>% # mutate(VIEWCAT = factor(VIEWCAT, # levels = c(1, 2, 3, 4), # labels = c("Rarely/never", # "2-3 times per week", # "4-5 times per week", # "> 5 times per week"))) ## Check the reference group: levels(sesam2$VIEWCAT) ## [1] "1" "2" "3" "4" contrasts(sesam2$VIEWCAT) ## 2 3 4 ## 1 0 0 0 ## 2 1 0 0 ## 3 0 1 0 ## 4 0 0 1 ## If necessary, relevel # sesam <- # sesam2 %>% # mutate(VIEWCAT = relevel(VIEWCAT, 1)) 3.3.2.3 Use lm() to estimate a multiple regression model wherein VIEWCAT predicts POSTNUMB. Summarize the model. Interpret the estimates. Click to show code lmOut <- lm(POSTNUMB ~ VIEWCAT, data = sesam2) summary(lmOut) ## ## Call: ## lm(formula = POSTNUMB ~ VIEWCAT, data = sesam2) ## ## Residuals: ## Min 1Q Median 3Q Max ## -25.474 -7.942 0.240 8.526 25.240 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) 18.760 2.316 8.102 8.95e-14 *** ## VIEWCAT2 9.331 2.900 3.218 0.00154 ** ## VIEWCAT3 14.714 2.777 5.298 3.49e-07 *** ## VIEWCAT4 18.032 2.809 6.419 1.24e-09 *** ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 11.58 on 175 degrees of freedom ## Multiple R-squared: 0.2102, Adjusted R-squared: 0.1967 ## F-statistic: 15.53 on 3 and 175 DF, p-value: 5.337e-09 Click for explanation Viewing category explains a statistically significant proportion of the variance in the post-test score of numbers learned (\\(R^2 = 0.21\\), \\(F(3, 175) = 15.53\\), \\(p < 0.001\\)). Kids who never or rarely watched Sesame Street had an average score of 18.76 on the post-test. Kids with weekly viewing habits of 2–3, 4–5, or 5+ times per week all had significantly higher scores on the post-test than kids who never or rarely watched Sesame Street (2–3: \\(\\hat{\\beta} = 9.33\\), \\(t = 3.22\\), \\(p = 0.002\\); 4–5: \\(\\hat{\\beta} = 14.71\\), \\(t = 5.3\\), \\(p < 0.001\\); 5+: \\(\\hat{\\beta} = 18.03\\), \\(t = 6.42\\), \\(p < 0.001\\)). If we compare the box plot, kernel density plot, and model output below, the relationships between the regression coefficient estimates for the viewing categories and the group means should be evident. 3.3.2.4 Use ggplot() to make a scatterplot with AGE on the x-axis and POSTNUMB on the y-axis. Color the points according to the their VIEWCAT level. Save the plot object to a variable in your environment. Hint: You can map color to the levels of a variable on your dataset by assigning the variable names to the color argument of the aes() function in ggplot(). Click to show code library(ggplot2) ## Add aes(..., color = VIEWCAT) to get different colors for each group: p <- ggplot(sesam2, aes(x = AGE, y = POSTNUMB, color = VIEWCAT)) + geom_point() # Add points for scatterplot ## Print the plot stored as 'p': p We assigned the global color aesthetic to the VIEWCAT variable, so the points are colored based on their group. 3.3.2.5 Add linear regression lines for each group to the above scatterplot. Hints: You can add regression lines with ggplot2::geom_smooth() To get linear regression lines, set the argument method = \"lm\" To omit error envelopes, set the argument se = FALSE Click to show code ## Add OLS best-fit lines: p + geom_smooth(method = "lm", se = FALSE) The global color aesthetic assignment from above carries through to any additional plot elements that we add, including the regression lines. So, we also get a separate regression line for each VIEWCAT group. 3.3.2.6 How would you interpret the pattern of regression lines above? Click for explanation All the lines show a positive slope, so post-test number recognition appears to increase along with increasing age. The lines are not parallel, though. So VIEWCAT may be moderating the effect of AGE on POSTNUMB. Based on the figure we just created, we may want to test for moderation in our regression model. To do so, we need to add an interaction between AGE and VIEWCAT. The VIEWCAT factor is represented by 3 dummy codes in our model, though. So when we interact AGE and VIEWCAT, we will create 3 interaction terms. To test the overall moderating influence of VIEWCAT, we need to conduct a multiparameter hypothesis test of all 3 interaction terms. One way that we can go about implementing such a test is through a hierarchical regression analysis entailing three steps: Estimate the additive model wherein we regress POSTNUMB onto AGE and VIEWCAT without any interaction. Estimate the moderated model by adding the interaction between AGE and VIEWCAT into the additive model. Conduct a \\(\\Delta R^2\\) test to compare the fit of the two models. 3.3.2.7 Conduct the hierarchical regression analysis described above. Does VIEWCAT significantly moderate the effect of AGE on POSTNUMB? Provide statistical justification for your conclusion. Click to show code ## Estimate the additive model a view the results: results_add <- lm(POSTNUMB ~ VIEWCAT + AGE, data = sesam2) summary(results_add) ## ## Call: ## lm(formula = POSTNUMB ~ VIEWCAT + AGE, data = sesam2) ## ## Residuals: ## Min 1Q Median 3Q Max ## -23.680 -8.003 -0.070 8.464 22.635 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) -10.1056 6.5091 -1.553 0.12235 ## VIEWCAT2 9.1453 2.7390 3.339 0.00103 ** ## VIEWCAT3 13.8602 2.6294 5.271 3.98e-07 *** ## VIEWCAT4 16.9215 2.6636 6.353 1.79e-09 *** ## AGE 0.5750 0.1221 4.708 5.08e-06 *** ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 10.94 on 174 degrees of freedom ## Multiple R-squared: 0.2995, Adjusted R-squared: 0.2834 ## F-statistic: 18.6 on 4 and 174 DF, p-value: 9.642e-13 ## Estimate the moderated model and view the results: results_mod <- lm(POSTNUMB ~ VIEWCAT * AGE, data = sesam2) summary(results_mod) ## ## Call: ## lm(formula = POSTNUMB ~ VIEWCAT * AGE, data = sesam2) ## ## Residuals: ## Min 1Q Median 3Q Max ## -23.8371 -8.2387 0.6158 8.7988 22.5611 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) -18.7211 15.5883 -1.201 0.2314 ## VIEWCAT2 9.9741 20.6227 0.484 0.6293 ## VIEWCAT3 23.5825 19.3591 1.218 0.2248 ## VIEWCAT4 34.3969 19.3600 1.777 0.0774 . ## AGE 0.7466 0.3074 2.429 0.0162 * ## VIEWCAT2:AGE -0.0175 0.4060 -0.043 0.9657 ## VIEWCAT3:AGE -0.1930 0.3782 -0.510 0.6104 ## VIEWCAT4:AGE -0.3416 0.3770 -0.906 0.3663 ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 10.99 on 171 degrees of freedom ## Multiple R-squared: 0.3046, Adjusted R-squared: 0.2762 ## F-statistic: 10.7 on 7 and 171 DF, p-value: 3.79e-11 ## Test for moderation: anova(results_add, results_mod) Click for explanation VIEWCAT does not significantly moderate the effect of AGE on POSTNUMB (\\(F[3, 171] = 0.422\\), \\(p = 0.738\\)). 3.3.2.8 Sketch the analytic path diagrams for the additive and moderated models you estimated in 3.3.2.7 (on paper). Click for explanation Additive Model Moderated Model End of At-Home Exercises 3 These data were simulated from the covariance matrix provided in Laible, D. J., Carlo, G., & Roesch, S. C. (2004). Pathways to self-esteem in late adolescence: The role of parent and peer attachment, empathy, and social behaviours. Journal of adolescence, 27(6), 703-716.↩︎ These data are from the very interesting study: Ball, S., & Bogatz, G. A. (1970). A Summary of the Major Findings in” The First Year of Sesame Street: An Evaluation”.↩︎ "],["in-class-exercises-2.html", "3.4 In-Class Exercises", " 3.4 In-Class Exercises 3.4.1 Mediation In this practical, we’ll go back to the data from the at-home exercises, SelfEsteem.sav. Recall that these data comprise 143 observations of the following variables. case: Participant ID number ParAtt: Parental Attachment PeerAtt: Peer Attachment Emp: Empathy ProSoc: Prosocial behavior Aggr: Aggression SelfEst: Self-esteem When we last worked with the data, we built a model with one mediator (Emp), creating indirect effects between our predictors ParAtt and PeerAtt, and our outcome variable SelfEst. Below, you will estimate a more complex, multiple-mediator model. 3.4.1.1 Load the data into the object seData using haven::read_sav() Click to show code library(haven) seData <- read_sav("SelfEsteem.sav") For this analysis, we are interested in the (indirect) effects of parental and peer attachment on self-esteem. Furthermore, we want to evaluate the mediating roles of empathy and social behavior (i.e., prosocial behavior and aggression). Specifically, we have the following hypotheses. Better peer relationships will promote higher self-esteem via a three-step indirect process. Better peer relationships will increase empathy levels. Higher empathy will increase prosocial behavior and decrease aggressive behavior. More prosocial behaviors and less aggressive behavior will both produce higher self-esteem. Better relationships with parents directly increase self-esteem. To evaluate these hypotheses, we will use lavaan to estimate the following multiple mediator model as a path model. 3.4.1.2 Specify the lavaan model syntax implied by the path diagram shown above. Save the resulting character string as an object in your environment. Click to show code mod0 <- ' ## Equation for outcome: SelfEst ~ ProSoc + Aggr + Emp + ParAtt + PeerAtt ## Equations for stage 2 mediators: ProSoc ~ PeerAtt + ParAtt + Emp Aggr ~ PeerAtt + ParAtt + Emp ## Equation for stage 1 mediator: Emp ~ ParAtt + PeerAtt ## Covariances: ProSoc ~~ Aggr ParAtt ~~ PeerAtt ' 3.4.1.3 Use the lavaan::sem() function to estimate the model defined in 3.4.1.2. Use the default settings in sem(). Summarize the fitted model. Click to show code library(lavaan) out <- sem(mod0, data = seData) summary(out) ## lavaan 0.6.16 ended normally after 16 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 21 ## ## Number of observations 143 ## ## Model Test User Model: ## ## Test statistic 0.000 ## Degrees of freedom 0 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## SelfEst ~ ## ProSoc 0.252 0.096 2.634 0.008 ## Aggr 0.185 0.085 2.172 0.030 ## Emp 0.143 0.098 1.460 0.144 ## ParAtt 0.244 0.078 3.133 0.002 ## PeerAtt 0.051 0.091 0.555 0.579 ## ProSoc ~ ## PeerAtt -0.037 0.080 -0.469 0.639 ## ParAtt 0.193 0.067 2.886 0.004 ## Emp 0.477 0.074 6.411 0.000 ## Aggr ~ ## PeerAtt -0.095 0.090 -1.055 0.291 ## ParAtt -0.034 0.075 -0.454 0.650 ## Emp -0.309 0.084 -3.697 0.000 ## Emp ~ ## ParAtt 0.078 0.075 1.045 0.296 ## PeerAtt 0.306 0.086 3.557 0.000 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ## .ProSoc ~~ ## .Aggr -0.086 0.058 -1.476 0.140 ## ParAtt ~~ ## PeerAtt 0.537 0.103 5.215 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .SelfEst 0.796 0.094 8.456 0.000 ## .ProSoc 0.618 0.073 8.456 0.000 ## .Aggr 0.777 0.092 8.456 0.000 ## .Emp 0.779 0.092 8.456 0.000 ## ParAtt 1.277 0.151 8.456 0.000 ## PeerAtt 0.963 0.114 8.456 0.000 3.4.1.4 Considering the parameter estimates from 3.4.1.3, what can you say about the hypotheses? Click for explanation Notice that all of the hypotheses stated above are explicitly directional. Hence, when evaluating the significance of the structural paths that speak to these hypotheses, we should use one-tailed tests. We cannot ask lavaan to return one-tailed p-values, but we have no need to do so. We can simply divide the two-tailed p-values in half. The significant direct effect of ParAtt on SelfEst (\\(\\beta = 0.244\\), \\(Z = 3.13\\), \\(p = 0.001\\)) and the lack of a significant direct effect of PeerAtt on SelfEst (\\(\\beta = 0.051\\), \\(Z = 0.555\\), \\(p = 0.29\\)) align with our hypotheses. The remaining patterns of individual estimates also seem to conform to the hypotheses (e.g., all of the individual paths comprising the indirect effects of PeerAtt on SelfEst are significant). We cannot make any firm conclusions until we actually estimate and test the indirect effects, though. 3.4.1.5 Modify the model syntax from 3.4.1.2 by adding definitions of the two hypothesized IEs from PeerAtt to SelfEst. Click to show code You can use any labeling scheme that makes sense to you, but I recommend adopting some kind of systematic rule. Here, I will label the individual estimates in terms of the short variable names used in the path diagram above. mod <- ' ## Equation for outcome: SelfEst ~ y_m21 * ProSoc + y_m22 * Aggr + Emp + ParAtt + PeerAtt ## Equations for stage 2 mediators: ProSoc ~ m21_x2 * PeerAtt + ParAtt + m21_m1 * Emp Aggr ~ m22_x2 * PeerAtt + ParAtt + m22_m1 * Emp ## Equation for stage 1 mediator: Emp ~ ParAtt + m1_x2 * PeerAtt ## Covariances: ProSoc ~~ Aggr ParAtt ~~ PeerAtt ## Indirect effects: ie_pro := m1_x2 * m21_m1 * y_m21 ie_agg := m1_x2 * m22_m1 * y_m22 ' 3.4.1.6 Use lavaan::sem() to estimate the model with the IEs defined. Use the default settings for sem(). Are the hypothesized IEs significant according to the default tests? Click to show code out <- sem(mod, data = seData) summary(out) ## lavaan 0.6.16 ended normally after 16 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 21 ## ## Number of observations 143 ## ## Model Test User Model: ## ## Test statistic 0.000 ## Degrees of freedom 0 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## SelfEst ~ ## ProSoc (y_21) 0.252 0.096 2.634 0.008 ## Aggr (y_22) 0.185 0.085 2.172 0.030 ## Emp 0.143 0.098 1.460 0.144 ## ParAtt 0.244 0.078 3.133 0.002 ## PerAtt 0.051 0.091 0.555 0.579 ## ProSoc ~ ## PerAtt (m21_2) -0.037 0.080 -0.469 0.639 ## ParAtt 0.193 0.067 2.886 0.004 ## Emp (m21_1) 0.477 0.074 6.411 0.000 ## Aggr ~ ## PerAtt (m22_2) -0.095 0.090 -1.055 0.291 ## ParAtt -0.034 0.075 -0.454 0.650 ## Emp (m22_1) -0.309 0.084 -3.697 0.000 ## Emp ~ ## ParAtt 0.078 0.075 1.045 0.296 ## PerAtt (m1_2) 0.306 0.086 3.557 0.000 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ## .ProSoc ~~ ## .Aggr -0.086 0.058 -1.476 0.140 ## ParAtt ~~ ## PeerAtt 0.537 0.103 5.215 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .SelfEst 0.796 0.094 8.456 0.000 ## .ProSoc 0.618 0.073 8.456 0.000 ## .Aggr 0.777 0.092 8.456 0.000 ## .Emp 0.779 0.092 8.456 0.000 ## ParAtt 1.277 0.151 8.456 0.000 ## PeerAtt 0.963 0.114 8.456 0.000 ## ## Defined Parameters: ## Estimate Std.Err z-value P(>|z|) ## ie_pro 0.037 0.018 2.010 0.044 ## ie_agg -0.017 0.011 -1.657 0.098 Click for explanation The IE of Peer Attachment on Self Esteem through Empathy and Prosocial Behavior is significant (\\(\\hat{\\textit{IE}} = 0.037\\), \\(Z = 2.01\\), \\(p = 0.022\\)), as is the analogous IE through Aggressive Behavior (\\(\\hat{\\textit{IE}} = -0.017\\), \\(Z = -1.66\\), \\(p = 0.049\\)). Though, this latter effect is just barely significant at the \\(\\alpha = 0.05\\) level. The tests we used to evaluate the significance of the IEs in 3.4.1.6 are flawed because they assume normal sampling distributions for the IEs. However the IEs are defined as products of multiple, normally distributed, regression slopes. So the IEs themselves cannot be normally distributed (at least in finite samples), and the results of the normal-theory significance tests may be misleading. To get an accurate test of the IEs, we should use bootstrapping to generate an empirical sampling distribution for each IE. In lavaan, we implement bootstrapping by specifying the se = \"bootstrap\" option in the fitting function (i.e., the cfa() or sem() function) and specifying the number of bootstrap samples via the bootstrap option. Workflow Tip To draw reliable conclusions from bootstrapped results, we need many bootstrap samples (i.e., B > 1000), but we must estimate the full model for each of these samples, so the estimation can take a long time. To avoid too much frustration, you should first estimate the model without bootstrapping to make sure everything is specified correctly. Only after you are certain that your code is correct do you want to run the full bootstrapped version. 3.4.1.7 Re-estimate the model from 3.4.1.6 using 1000 bootstrap samples. Other than the se and bootstrap options, use the defaults. Are the hypothesized IEs significant according to the bootstrap-based test statistics? Click to show code ## Set a seed to get replicable bootstrap samples: set.seed(235711) ## Estimate the model with bootstrapping: out_boot <- sem(mod, data = seData, se = "bootstrap", bootstrap = 1000) ## Summarize the model: summary(out_boot) ## lavaan 0.6.16 ended normally after 16 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 21 ## ## Number of observations 143 ## ## Model Test User Model: ## ## Test statistic 0.000 ## Degrees of freedom 0 ## ## Parameter Estimates: ## ## Standard errors Bootstrap ## Number of requested bootstrap draws 1000 ## Number of successful bootstrap draws 1000 ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## SelfEst ~ ## ProSoc (y_21) 0.252 0.100 2.529 0.011 ## Aggr (y_22) 0.185 0.085 2.174 0.030 ## Emp 0.143 0.095 1.507 0.132 ## ParAtt 0.244 0.079 3.089 0.002 ## PerAtt 0.051 0.095 0.530 0.596 ## ProSoc ~ ## PerAtt (m21_2) -0.037 0.082 -0.456 0.648 ## ParAtt 0.193 0.068 2.831 0.005 ## Emp (m21_1) 0.477 0.078 6.092 0.000 ## Aggr ~ ## PerAtt (m22_2) -0.095 0.087 -1.093 0.275 ## ParAtt -0.034 0.076 -0.448 0.654 ## Emp (m22_1) -0.309 0.092 -3.356 0.001 ## Emp ~ ## ParAtt 0.078 0.072 1.092 0.275 ## PerAtt (m1_2) 0.306 0.079 3.896 0.000 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ## .ProSoc ~~ ## .Aggr -0.086 0.058 -1.493 0.135 ## ParAtt ~~ ## PeerAtt 0.537 0.128 4.195 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .SelfEst 0.796 0.082 9.698 0.000 ## .ProSoc 0.618 0.068 9.114 0.000 ## .Aggr 0.777 0.104 7.476 0.000 ## .Emp 0.779 0.090 8.651 0.000 ## ParAtt 1.277 0.197 6.473 0.000 ## PeerAtt 0.963 0.105 9.203 0.000 ## ## Defined Parameters: ## Estimate Std.Err z-value P(>|z|) ## ie_pro 0.037 0.019 1.891 0.059 ## ie_agg -0.017 0.011 -1.638 0.101 Click for explanation As with the normal-theory tests, the hypothesized IE of Peer Attachment on Self Esteem was significant (\\(\\hat{\\textit{IE}} = 0.037\\), \\(Z = 1.89\\), \\(p = 0.029\\)), but the IE of Aggressive Behavior has crossed into nonsignificant territory (\\(\\hat{\\textit{IE}} = -0.017\\), \\(Z = -1.64\\), \\(p = 0.051\\)). Note: Bootstrapping is a stochastic method, so each run can provide different results. Since the indirect effect of aggressive behavior is so close to the critical value, you may come to a different conclusions vis-á-vis statistical significance if you run this analysis with a different random number seed or a different number of bootstrap samples. When you use the summary() function to summarize the bootstrapped model from 3.4.1.7, the output will probably look pretty much the same as it did in 3.4.1.6, but it’s not. The standard errors and test statistics in the bootstrapped summary are derived from empirical sampling distributions, whereas these values are based on an assumed normal sampling distribution in 3.4.1.6. The standard method of testing IEs with bootstrapping is to compute confidence intervals (CIs) from the empirical sampling distribution of the IEs. In lavaan, we can compute basic (percentile, 95%) CIs by adding the ci = TRUE option to the summary() function. To evaluate our directional hypotheses at an \\(\\alpha = 0.05\\) level, however, we need to compute 90% CIs. We can get more control over the summary statistics (include the CIs) with the parameterEstimates() function. 3.4.1.8 Check the documentation for lavaan::parameterEstimates(). Click to show code ?parameterEstimates 3.4.1.9 Use the parameterEstimates() function to compute bootstrapped CIs for the hypothesized IEs. Compute percentile CIs. Are the IEs significant according to the bootstrapped CIs? Click to show code parameterEstimates(out_boot, ci = TRUE, level = 0.9) Click for explanation When evaluating a directional hypothesis with a CI, we only consider one of the interval’s boundaries. For a hypothesized positive effect, we check only if the lower boundary is greater than zero. For a hypothesized negative effect, we check if the upper boundary is less than zero. As with the previous tests, the IE of Peer Attachment on Self Esteem through Empathy and Prosocial Behavior is significant (\\(\\hat{\\textit{IE}} = 0.037\\), \\(95\\% ~ CI = [0.009; \\infty]\\)), but the analogous IE through Aggressive Behavior is not quite significant (\\(\\hat{\\textit{IE}} = -0.017\\), \\(95\\% ~ CI = [-\\infty; -0.003]\\)). 3.4.1.10 Based on the analyses you’ve conducted here, what do you conclude vis-à-vis the original hypotheses? Click for explanation When using normal-theory tests, both hypothesized indirect effects between Peer Attachment and Self Esteem were supported in that the IE through Empathy and Prosocial Behavior as well as the IE through Empathy and Aggressive Behavior were both significant. The hypothesized direct effect of Parent Attachment on Self Esteem was also born out via a significant direct effect in the model. When testing the indirect effects with bootstrapping, however, the effect through Aggressive Behavior was nonsignificant. Since bootstrapping gives a more accurate test of the indirect effect, we should probably trust these results more than the normal-theory results. We should not infer a significant indirect effect of Peer Attachment on Self Esteem transmitted through Empathy and Aggressive Behavior. These results may not tell the whole story, though. We have not tested for indirect effects between Parent Attachment and Self Esteem, and we have not evaluated simpler indirect effects between Peer Attachment and Self Esteem (e.g., PeerAtt \\(\\rightarrow\\) Emp \\(\\rightarrow\\) SelfEst). 3.4.2 Moderation We will first analyze a synthetic version of the Outlook on Life Survey data. The original data were collected in the United States in 2012 to measure, among other things, attitudes about racial issues, opinions of the Federal government, and beliefs about the future. We will work with a synthesized subset of the original data. You can access these synthetic data as outlook.rds. This dataset comprises 2288 observations of the following 13 variables. d1:d3: Three observed indicators of a construct measuring disillusionment with the US Federal government. Higher scores indicate more disillusionment s1:s4: Four observed indicators of a construct measuring the perceived achievability of material success. Higher scores indicate greater perceived achievability progress: A single item assessing perceived progress toward achieving the “American Dream” Higher scores indicate greater perceived progress merit: A single item assessing endorsement of the meritocratic ideal that hard work leads to success. Higher scores indicate stronger endorsement of the meritocratic ideal lib2Con: A single item assessing liberal-to-conservative orientation Lower scores are more liberal, higher scores are more conservative party: A four-level factor indicating self-reported political party affiliation disillusion: A scale score representing disillusionment with the US Federal government Created as the mean of d1:d3 success: A scale score representing the perceived achievability of material success Created as the mean of s1:s4 To satisfy the access and licensing conditions under which the original data are distributed, the data contained in outlook.rds were synthesized from the original variables using the methods described by Volker and Vink (2021). You can access the original data here, and you can access the code used to process the data here. 3.4.2.1 Read in the outlook.rds dataset. Hint: An RDS file is an R object that’s been saved to a file. To read in this type of file, we use readRDS() from base R. Click to show code outlook <- readRDS("outlook.rds") 3.4.2.2 Summarize the outlook data to get a sense of their characteristics. Click to show code head(outlook) summary(outlook) ## d1 d2 d3 s1 ## Min. :1.000 Min. :1.000 Min. :1.000 Min. :1.000 ## 1st Qu.:3.000 1st Qu.:2.000 1st Qu.:3.000 1st Qu.:2.000 ## Median :4.000 Median :3.000 Median :4.000 Median :2.000 ## Mean :3.642 Mean :3.218 Mean :3.629 Mean :2.288 ## 3rd Qu.:4.000 3rd Qu.:4.000 3rd Qu.:4.000 3rd Qu.:3.000 ## Max. :5.000 Max. :5.000 Max. :5.000 Max. :4.000 ## s2 s3 s4 progress ## Min. :1.000 Min. :1.000 Min. :1.000 Min. : 1.000 ## 1st Qu.:1.000 1st Qu.:1.000 1st Qu.:2.000 1st Qu.: 5.000 ## Median :2.000 Median :2.000 Median :2.000 Median : 7.000 ## Mean :1.922 Mean :2.012 Mean :2.469 Mean : 6.432 ## 3rd Qu.:2.000 3rd Qu.:3.000 3rd Qu.:3.000 3rd Qu.: 8.000 ## Max. :4.000 Max. :4.000 Max. :4.000 Max. :10.000 ## merit lib2Con party disillusion ## Min. :1.000 Min. :1.000 republican : 332 Min. :1.000 ## 1st Qu.:4.000 1st Qu.:3.000 democrat :1264 1st Qu.:3.000 ## Median :5.000 Median :4.000 independent: 576 Median :3.667 ## Mean :4.826 Mean :3.998 other : 116 Mean :3.497 ## 3rd Qu.:6.000 3rd Qu.:5.000 3rd Qu.:4.000 ## Max. :7.000 Max. :7.000 Max. :5.000 ## success ## Min. :1.000 ## 1st Qu.:1.750 ## Median :2.000 ## Mean :2.173 ## 3rd Qu.:2.500 ## Max. :4.000 str(outlook) ## 'data.frame': 2288 obs. of 13 variables: ## $ d1 : num 4 4 4 5 5 4 5 4 4 4 ... ## $ d2 : num 4 2 4 4 3 5 4 2 4 5 ... ## $ d3 : num 4 4 4 5 4 4 4 3 3 4 ... ## $ s1 : num 3 3 4 2 2 2 2 1 3 3 ... ## $ s2 : num 2 2 2 1 1 2 1 1 2 2 ... ## $ s3 : num 3 2 4 1 2 1 1 1 3 2 ... ## $ s4 : num 3 3 3 1 2 3 3 2 2 2 ... ## $ progress : num 8 4 6 1 6 5 7 6 9 7 ... ## $ merit : num 6 5 5 4 3 4 2 5 5 5 ... ## $ lib2Con : num 5 6 4 1 4 4 4 4 4 5 ... ## $ party : Factor w/ 4 levels "republican","democrat",..: 1 3 3 2 2 2 2 2 4 1 ... ## $ disillusion: num 4 3.33 4 4.67 4 ... ## $ success : num 2.75 2.5 3.25 1.25 1.75 2 1.75 1.25 2.5 2.25 ... We will first use OLS regression to estimate a model encoding the following relations: Belief in the achievability of success, success, predicts perceived progress toward the American Dream, progress, as the focal effect. Disillusionment with the US Federal government, disillusion moderates the success \\(\\rightarrow\\) progress effect. Placement on the liberal-to-conservative continuum, lib2Con is partialed out as a covariate. 3.4.2.3 Draw the conceptual path diagram for the model described above. Click for explanation 3.4.2.4 Write out the regression equation necessary to evaluate the moderation hypothesis described above. Click for explanation \\[ Y_{progress} = \\beta_0 + \\beta_1 W_{lib2Con} + \\beta_2 X_{success} + \\beta_3 Z_{disillusion} + \\beta_4 XZ + \\varepsilon \\] 3.4.2.5 Use lm() to estimate the moderated regression model via OLS regression. Click to show code olsFit <- lm(progress ~ lib2Con + success * disillusion, data = outlook) 3.4.2.6 Summarize the fitted model and interpret the results. Is the moderation hypothesis supported? How does disillusionment level affect the focal effect? Click to show code summary(olsFit) ## ## Call: ## lm(formula = progress ~ lib2Con + success * disillusion, data = outlook) ## ## Residuals: ## Min 1Q Median 3Q Max ## -7.4315 -1.2525 0.1307 1.4369 5.6717 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) 6.81128 0.62073 10.973 < 2e-16 *** ## lib2Con 0.03052 0.03040 1.004 0.3155 ## success 0.42360 0.25853 1.638 0.1015 ## disillusion -0.78002 0.16864 -4.625 3.95e-06 *** ## success:disillusion 0.17429 0.07273 2.396 0.0166 * ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 2.041 on 2283 degrees of freedom ## Multiple R-squared: 0.1385, Adjusted R-squared: 0.137 ## F-statistic: 91.74 on 4 and 2283 DF, p-value: < 2.2e-16 Click for explanation Yes, disillusion significantly moderates the relation between success and progress (\\(\\beta = 0.174\\), \\(t[2283] = 2.396\\), \\(p = 0.017\\)) such that the effect of success on progress increases as levels of disillusion increase, after controlling for lib2Con. The rockchalk package contains some useful routines for probing interactions estimated via lm(). Specifically, the plotslopes() function will estimate and plot simple slopes, and the testSlopes() function tests the simple slopes estimated by plotSlopes(). 3.4.2.7 Probe the interaction. Use the plotSlopes() and testSlopes() functions from the rockchalk package to conduct a simple slopes analysis for the model from 3.4.2.5. Click to show code library(rockchalk) ## Estimate and plot simple slopes: psOut <- plotSlopes(olsFit, plotx = "success", modx = "disillusion", modxVals = "std.dev") ## Test the simple slopes: tsOut <- testSlopes(psOut) ## Values of disillusion OUTSIDE this interval: ## lo hi ## -28.9332857 0.2672244 ## cause the slope of (b1 + b2*disillusion)success to be statistically significant ## View the results: tsOut$hypotests Note: The message printed by testSlopes() gives the boundaries of the Johnson-Neyman Region of Significance (Johnson & Neyman, 1936). Johnson-Neyman analysis is an alternative method of probing interactions that we have not covered in this course. For more information, check out Preacher, et al. (2006). We will now use lavaan to estimate the moderated regression model from above as a path analysis. 3.4.2.8 Define the model syntax for the path analytic version of the model described above. Parameterize the model as in the OLS regression. Use only observed items and scale scores. Click to show code pathMod <- ' progress ~ 1 + lib2Con + success + disillusion + success:disillusion ' 3.4.2.9 Estimate the path model on the outlook data. Click to show code pathFit <- sem(pathMod, data = outlook) 3.4.2.10 Summarize the fitted path model and interpret the results. Do the results match the OLS regression results? What proportion of the variability in progress is explained by this model? Hint: the function lavInspect() can be used to extract information from models Click to show code summary(pathFit) ## lavaan 0.6.16 ended normally after 1 iteration ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 6 ## ## Number of observations 2288 ## ## Model Test User Model: ## ## Test statistic 0.000 ## Degrees of freedom 0 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## progress ~ ## lib2Con 0.031 0.030 1.005 0.315 ## success 0.424 0.258 1.640 0.101 ## disillusion -0.780 0.168 -4.630 0.000 ## success:dsllsn 0.174 0.073 2.399 0.016 ## ## Intercepts: ## Estimate Std.Err z-value P(>|z|) ## .progress 6.811 0.620 10.985 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .progress 4.157 0.123 33.823 0.000 lavInspect(pathFit, "r2") ## progress ## 0.138 Click for explanation Yes, the estimates and inferential conclusions are all the same as in the OLS regression model. The model explains 13.85% of the variability in progress. The semTools package contains some helpful routines for probing interactions estimated via the lavaan() function (or one of it’s wrappers). Specifically, the probe2WayMC() and plotProbe() functions will estimate/test simple slopes and plot the estimated simple slopes, respectively. 3.4.2.11 Probe the interaction from 3.4.2.9 using semTools utilities. Use probe2WayMC() to estimate and test the simple slopes. Use plotProbe() to visualize the simple slopes. Define the simple slopes with the same conditional values of disillusion that you used in 3.4.2.7. Which simple slopes are significant? Do these results match the results from 3.4.2.7? Click to show code library(semTools) ## Define the conditional values at which to calculate simple slopes: condVals <- summarise(outlook, "m-sd" = mean(disillusion) - sd(disillusion), mean = mean(disillusion), "m+sd" = mean(disillusion) + sd(disillusion) ) %>% unlist() ## Compute simple slopes and intercepts: ssOut <- probe2WayMC(pathFit, nameX = c("success", "disillusion", "success:disillusion"), nameY = "progress", modVar = "disillusion", valProbe = condVals) ## Check the results: ssOut ## $SimpleIntcept ## disillusion est se z pvalue ## m-sd 2.719 4.690 0.231 20.271 0 ## mean 3.497 4.084 0.190 21.508 0 ## m+sd 4.274 3.477 0.230 15.122 0 ## ## $SimpleSlope ## disillusion est se z pvalue ## m-sd 2.719 0.897 0.083 10.792 0 ## mean 3.497 1.033 0.065 15.994 0 ## m+sd 4.274 1.169 0.088 13.223 0 ## Visualize the simple slopes: plotProbe(ssOut, xlim = range(outlook$success), xlab = "Ease of Personal Success", ylab = "Progress toward American Dream", legendArgs = list(legend = names(condVals)) ) Click for explanation Each of the simple slopes is significant. As level of disillusionment increases, the effect of success on progress also increases, and this effect is significant for all levels of disillusion considered here. These results match the simple slopes from the OLS regression analysis. End of In-Class Exercises 3 "],["efa.html", "4 EFA", " 4 EFA This week will be a general introduction to latent variables and scaling procedures. We will discuss several different aspects of exploratory factor analysis (EFA). Most notably: The differences between Principal Component Analyses (PCA) and Factor Analysis Model estimation and factor extraction methods Factor rotations You will have to make decisions regarding each of these aspects when conducting a factor analysis. We will also discuss reliability and factor scores as means of evaluating the properties of a scale. Homework before the lecture Watch the Lecture Recording for this week. Complete the Reading for this week, and answer the associated reading questions. Homework before the practical Complete the At-Home Exercises. Practical content During the practical you will work on the In-Class Exercises. "],["lecture-3.html", "4.1 Lecture", " 4.1 Lecture How do you know if you have measured the putative hypothetical construct that you intend to measure? The methods introduced in this lecture (namely, latent variables, factor analysis, and reliability analysis) can shed empirical light on this issue. In the social and behavioral sciences we’re often forced to measure key concepts indirectly. For example, we have no way of directly quantifying a person’s current level of depression, or their innate motivation, or their risk-aversion, or any of the other myriad psychological features that comprise the human mental state. In truth, we cannot really measure these hypothetical constructs at all, we must estimate latent representations thereof (though, psychometricians still use the language of physical measurement to describe this process). Furthermore, we can rarely estimate an adequate representation with only a single observed variable (e.g., question on a survey, score on a test, reading from a sensor). We generally need several observed variables to reliably represent a single hypothetical construct. For example, we cannot accurately determine someone’s IQ or socio-economic status based on their response to a single question; we need several questions that each tap into slightly different aspects of IQ or SES. Given multiple items measuring the same construct, we can use the methods discussed in this lecture (i.e., factor analysis and reliability analysis) to evaluate the quality of our measurement (i.e., how well we have estimated the underlying hypothetical construct). If we do well enough in this estimation task, we will be able to combine these estimated latent variables with the path analysis methods discussed in previous two weeks to produce the full structural equation models that we will cover at the end of this course. 4.1.1 Recording Notes: This week (and next), we’ll be re-using Caspar van Lissa’s old slides and lecture recording. So, you’ll see Caspar in the following video, and the slides will have a notably different flavor than our usual materials. Don’t be confused by any mention of “model fit” in the lecture. We haven’t covered model fit yet, but we will do so next week. 4.1.2 Slides You can download the lecture slides here. "],["reading-3.html", "4.2 Reading", " 4.2 Reading This week, you will read two papers. Reference 1 Preacher, K. J., & MacCullum, R. C. (2003). Repairing Tom Swift’s electric factor analysis machine, Understanding Statistics 2(1) 13–43. Questions 1 What is a latent variable? Give an example of a latent variable. What is factor analysis, and what can you investigate using this method? In the introduction, Preacher and Maccallum describe a “little jiffy” method of doing factor analysis. Briefly describe this little jiffy—or bad practice—method. Briefly explain the key differences between Principal Component Analyses (PCA) and Exploratory Factor Analyses (EFA). What is the purpose of factor rotation? Reference 2 Kestilä, E. (2006). Is there demand for radical right populism in the Finnish electorate? Scandinavian Political Studies 29(3), 169–191. Questions 2 What is the research question that the author tries to answer? Briefly describe the characteristics of the Radical Right Parties (RRP) in Europe. What are the two main explanations of support for RRP upon which this paper focuses? Does the empirical part of the paper reflect the theoretical framework well? Why or why not? According to the author, is Finland very different from other European countries on the main dependent variables? What is the author’s conclusion (i.e., how does the author answer the research question)? "],["at-home-exercises-3.html", "4.3 At-Home Exercises", " 4.3 At-Home Exercises In these exercises, you will attempt to replicate some of the analyses from the second reading for this week: Kestilä, E. (2006). Is there demand for radical right populism in the Finnish electorate? Scandinavian Political Studies 29(3), 169–191. The data for this practical were collected during the first round of the European Social Survey (ESS). The ESS is a repeated cross-sectional survey administered in 32 European countries. The first wave was collected in 2002, and two new waves have been collected each year since. You can find more info and access the data at https://www.europeansocialsurvey.org. The data we will analyze for this practical are contained in the file named ESSround1-a.sav. This file contains data for all respondents, but only includes those variables that you will need to complete the following exercises. 4.3.1 Load the ESSround1-a.sav dataset into R. Inspect the data after loading to make sure everything went well. Click to show code ## Load the 'haven' package: library(haven) library(tidySEM) ## Read the 'ESSround1-a.sav' data into a data frame called 'ess': ess <- read_spss("ESSround1-a.sav") ## Inspect the result: dim(ess) head(ess) descriptives(ess) ## [1] 42359 50 Click here for a description of the variables. Variable Description name Title of dataset essround ESS round edition Edition proddate Production date cntry Country idno Respondent’s identification number trstlgl Trust in the legal system trstplc Trust in the police trstun Trust in the United Nations trstep Trust in the European Parliament trstprl Trust in country’s parliament stfhlth State of health services in country nowadays stfedu State of education in country nowadays stfeco How satisfied with present state of economy in country stfgov How satisfied with the national government stfdem How satisfied with the way democracy works in country pltinvt Politicians interested in votes rather than peoples opinions pltcare Politicians in general care what people like respondent think trstplt Trust in politicians imsmetn Allow many/few immigrants of same race/ethnic group as majority imdfetn Allow many/few immigrants of different race/ethnic group from majority eimrcnt Allow many/few immigrants from richer countries in Europe eimpcnt Allow many/few immigrants from poorer countries in Europe imrcntr Allow many/few immigrants from richer countries outside Europe impcntr Allow many/few immigrants from poorer countries outside Europe qfimchr Qualification for immigration: christian background qfimwht Qualification for immigration: be white imwgdwn Average wages/salaries generally brought down by immigrants imhecop Immigrants harm economic prospects of the poor more than the rich imtcjob Immigrants take jobs away in country or create new jobs imbleco Taxes and services: immigrants take out more than they put in or less imbgeco Immigration bad or good for country’s economy imueclt Country’s cultural life undermined or enriched by immigrants imwbcnt Immigrants make country worse or better place to live imwbcrm Immigrants make country’s crime problems worse or better imrsprc Richer countries should be responsible for accepting people from poorer countries pplstrd Better for a country if almost everyone share customs and traditions vrtrlg Better for a country if a variety of different religions shrrfg Country has more than its fair share of people applying refugee status rfgawrk People applying refugee status allowed to work while cases considered gvrfgap Government should be generous judging applications for refugee status rfgfrpc Most refugee applicants not in real fear of persecution own countries rfggvfn Financial support to refugee applicants while cases considered rfgbfml Granted refugees should be entitled to bring close family members gndr Gender yrbrn Year of birth edulvl Highest level of education eduyrs Years of full-time education completed polintr How interested in politics lrscale Placement on left right scale One thing you might notice when inspecting the ess data is that most of the variables are stored as labelled vectors. When loading SPSS data, haven will use these labelled vectors to preserve the metadata associated with SPSS scale variables (i.e., variable labels and value labels). While it’s good to have this metadata available, we want to analyze these items as numeric variables and factors, so the value labels are only going to make our lives harder. Thankfully, the labelled package contains many routines for manipulating labelled vectors. We’ll deal with the numeric variables in just a bit, but our first task will be to covert grouping variables to factors. 4.3.2 Convert the cntry, gndr, edulvl, and polintr variables into factors. Use the as_factor() function to do the conversion. Convert edulvl and polintr to ordered factors. Click to see code library(dplyr) ess <- mutate(ess, country = as_factor(cntry), sex = as_factor(gndr), edulvl = as_factor(edulvl, ordered = TRUE), polintr = as_factor(polintr, ordered = TRUE) ) The ess dataset contains much more information than Kestilä (2006) used. Kestilä only analyzed data from the following ten countries: Austria Belgium Denmark Finland France Germany Italy Netherlands Norway Sweden So, our next task is to subset the data to only the relevant population. When we apply logical subsetting, we can select rows from a dataset based on logical conditions. In this case, we want to select only rows from the 10 countries listed above. 4.3.3 Subset the data to include only the 10 countries analyzed by Kestilä (2006). Inspect the subsetted data to check that everything went well. Hints: Use the %in% operator to create a logical vector that indicates which elements of the cntry variable are in the set of target counties. Use the droplevels() levels function to clean up empty factor levels. Click to show code ## Create a character vector naming the target countries: targets <- c("Austria", "Belgium", "Denmark", "Finland", "France", "Germany", "Italy", "Netherlands", "Norway", "Sweden") ## Select only those rows that come from a target country: ess <- filter(ess, country %in% targets) %>% # Subset rows droplevels() # Drop empty factor levels ## Inspect the result: dim(ess) ## [1] 19690 52 table(ess$country) ## ## Austria Belgium Germany Denmark Finland France ## 2257 1899 2919 1506 2000 1503 ## Italy Netherlands Norway Sweden ## 1207 2364 2036 1999 In keeping with common practice, we will treat ordinal Likert-type rating scales with five or more levels as continuous. Since some R routines will treat labelled vectors as discrete variables, we can make things easier for ourselves by converting all the labelled vectors in our data to numeric vectors. We can use the labelled::remove_val_labels() function to strip the value labels and convert all of the labelled vectors to numeric vectors. 4.3.4 Convert the remaining labelled vectors to numeric vectors. Click to see code ## If necessary, install the labelled package: # install.packages("labelled", repos = "https://cloud.r-project.org") ## Load the labelled package: library(labelled) ## Strip the value labels: ess <- remove_val_labels(ess) ## Check the effects: str(ess) ## tibble [19,690 × 52] (S3: tbl_df/tbl/data.frame) ## $ name : chr [1:19690] "ESS1e06_1" "ESS1e06_1" "ESS1e06_1" "ESS1e06_1" ... ## ..- attr(*, "label")= chr "Title of dataset" ## ..- attr(*, "format.spss")= chr "A9" ## ..- attr(*, "display_width")= int 14 ## $ essround: num [1:19690] 1 1 1 1 1 1 1 1 1 1 ... ## ..- attr(*, "label")= chr "ESS round" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 10 ## $ edition : chr [1:19690] "6.1" "6.1" "6.1" "6.1" ... ## ..- attr(*, "label")= chr "Edition" ## ..- attr(*, "format.spss")= chr "A3" ## ..- attr(*, "display_width")= int 9 ## $ proddate: chr [1:19690] "03.10.2008" "03.10.2008" "03.10.2008" "03.10.2008" ... ## ..- attr(*, "label")= chr "Production date" ## ..- attr(*, "format.spss")= chr "A10" ## ..- attr(*, "display_width")= int 12 ## $ cntry : num [1:19690] 1 18 1 1 18 1 2 18 1 18 ... ## ..- attr(*, "label")= chr "Country" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 7 ## $ idno : num [1:19690] 1 1 2 3 3 4 4 4 6 6 ... ## ..- attr(*, "label")= chr "Respondent's identification number" ## ..- attr(*, "format.spss")= chr "F9.0" ## ..- attr(*, "display_width")= int 11 ## $ trstlgl : num [1:19690] 10 6 8 4 8 10 9 7 7 7 ... ## ..- attr(*, "label")= chr "Trust in the legal system" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ trstplc : num [1:19690] 10 8 5 8 8 9 8 9 4 9 ... ## ..- attr(*, "label")= chr "Trust in the police" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ trstun : num [1:19690] 9 8 6 NA 5 8 NA 7 5 7 ... ## ..- attr(*, "label")= chr "Trust in the United Nations" ## ..- attr(*, "format.spss")= chr "F2.0" ## $ trstep : num [1:19690] NA 3 0 7 3 7 0 3 4 6 ... ## ..- attr(*, "label")= chr "Trust in the European Parliament" ## ..- attr(*, "format.spss")= chr "F2.0" ## $ trstprl : num [1:19690] 9 7 0 6 8 8 10 2 6 8 ... ## ..- attr(*, "label")= chr "Trust in country's parliament" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ stfhlth : num [1:19690] 10 4 0 7 6 8 NA 6 3 5 ... ## ..- attr(*, "label")= chr "State of health services in country nowadays" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ stfedu : num [1:19690] 8 7 7 5 8 7 NA 7 6 7 ... ## ..- attr(*, "label")= chr "State of education in country nowadays" ## ..- attr(*, "format.spss")= chr "F2.0" ## $ stfeco : num [1:19690] 7 6 0 7 8 6 NA 9 8 9 ... ## ..- attr(*, "label")= chr "How satisfied with present state of economy in country" ## ..- attr(*, "format.spss")= chr "F2.0" ## $ stfgov : num [1:19690] 7 7 0 7 6 3 NA 5 5 7 ... ## ..- attr(*, "label")= chr "How satisfied with the national government" ## ..- attr(*, "format.spss")= chr "F2.0" ## $ stfdem : num [1:19690] 8 5 5 5 7 7 NA 7 7 9 ... ## ..- attr(*, "label")= chr "How satisfied with the way democracy works in country" ## ..- attr(*, "format.spss")= chr "F2.0" ## $ pltinvt : num [1:19690] 1 3 1 1 4 1 1 3 2 3 ... ## ..- attr(*, "label")= chr "Politicians interested in votes rather than peoples opinions" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ pltcare : num [1:19690] 1 4 1 1 4 3 2 5 2 3 ... ## ..- attr(*, "label")= chr "Politicians in general care what people like respondent think" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ trstplt : num [1:19690] 0 5 0 2 5 4 8 2 4 6 ... ## ..- attr(*, "label")= chr "Trust in politicians" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ imsmetn : num [1:19690] 4 3 2 3 2 1 NA 2 NA 1 ... ## ..- attr(*, "label")= chr "Allow many/few immigrants of same race/ethnic group as majority" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ imdfetn : num [1:19690] 3 3 2 3 2 2 NA 2 NA 1 ... ## ..- attr(*, "label")= chr "Allow many/few immigrants of different race/ethnic group from majority" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ eimrcnt : num [1:19690] 4 2 2 2 3 1 NA 2 NA 1 ... ## ..- attr(*, "label")= chr "Allow many/few immigrants from richer countries in Europe" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ eimpcnt : num [1:19690] 3 2 2 2 2 2 NA 2 NA 1 ... ## ..- attr(*, "label")= chr "Allow many/few immigrants from poorer countries in Europe" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ imrcntr : num [1:19690] 3 3 2 2 2 1 NA 2 NA 2 ... ## ..- attr(*, "label")= chr "Allow many/few immigrants from richer countries outside Europe" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ impcntr : num [1:19690] 3 2 2 3 2 1 NA 2 NA 2 ... ## ..- attr(*, "label")= chr "Allow many/few immigrants from poorer countries outside Europe" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ qfimchr : num [1:19690] 4 2 0 6 2 0 99 0 1 2 ... ## ..- attr(*, "label")= chr "Qualification for immigration: christian background" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ qfimwht : num [1:19690] 1 0 0 0 0 0 99 0 0 1 ... ## ..- attr(*, "label")= chr "Qualification for immigration: be white" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ imwgdwn : num [1:19690] 3 4 2 2 3 3 NA 4 NA 4 ... ## ..- attr(*, "label")= chr "Average wages/salaries generally brought down by immigrants" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ imhecop : num [1:19690] 2 2 1 4 3 2 NA 3 NA 2 ... ## ..- attr(*, "label")= chr "Immigrants harm economic prospects of the poor more than the rich" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ imtcjob : num [1:19690] 7 5 6 5 7 10 NA 8 NA 4 ... ## ..- attr(*, "label")= chr "Immigrants take jobs away in country or create new jobs" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ imbleco : num [1:19690] 9 4 2 NA 3 10 NA 9 NA 6 ... ## ..- attr(*, "label")= chr "Taxes and services: immigrants take out more than they put in or less" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ imbgeco : num [1:19690] 4 3 10 7 5 10 NA 8 NA 5 ... ## ..- attr(*, "label")= chr "Immigration bad or good for country's economy" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ imueclt : num [1:19690] 9 4 10 5 4 10 NA 9 NA 3 ... ## ..- attr(*, "label")= chr "Country's cultural life undermined or enriched by immigrants" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ imwbcnt : num [1:19690] 7 3 5 5 5 10 NA 8 NA 5 ... ## ..- attr(*, "label")= chr "Immigrants make country worse or better place to live" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ imwbcrm : num [1:19690] 3 3 5 2 3 5 NA 5 NA 3 ... ## ..- attr(*, "label")= chr "Immigrants make country's crime problems worse or better" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ imrsprc : num [1:19690] 2 2 1 4 1 2 NA 1 1 3 ... ## ..- attr(*, "label")= chr "Richer countries should be responsible for accepting people from poorer countries" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ pplstrd : num [1:19690] 2 4 2 2 3 4 NA 4 4 2 ... ## ..- attr(*, "label")= chr "Better for a country if almost everyone share customs and traditions" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ vrtrlg : num [1:19690] 3 5 3 2 4 1 NA 4 2 3 ... ## ..- attr(*, "label")= chr "Better for a country if a variety of different religions" ## ..- attr(*, "format.spss")= chr "F1.0" ## $ shrrfg : num [1:19690] 3 2 1 1 3 3 NA 3 4 3 ... ## ..- attr(*, "label")= chr "Country has more than its fair share of people applying refugee status" ## ..- attr(*, "format.spss")= chr "F1.0" ## $ rfgawrk : num [1:19690] 2 2 1 2 2 2 NA 2 1 2 ... ## ..- attr(*, "label")= chr "People applying refugee status allowed to work while cases considered" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ gvrfgap : num [1:19690] 4 3 2 4 2 2 NA 3 2 4 ... ## ..- attr(*, "label")= chr "Government should be generous judging applications for refugee status" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ rfgfrpc : num [1:19690] 4 3 2 4 4 4 NA 4 3 4 ... ## ..- attr(*, "label")= chr "Most refugee applicants not in real fear of persecution own countries" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ rfggvfn : num [1:19690] 2 3 2 4 3 2 NA 2 2 2 ... ## ..- attr(*, "label")= chr "Financial support to refugee applicants while cases considered" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ rfgbfml : num [1:19690] 2 3 1 2 2 1 NA 4 2 3 ... ## ..- attr(*, "label")= chr "Granted refugees should be entitled to bring close family members" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 9 ## $ gndr : num [1:19690] 1 2 1 2 2 1 NA 2 2 1 ... ## ..- attr(*, "label")= chr "Gender" ## ..- attr(*, "format.spss")= chr "F1.0" ## ..- attr(*, "display_width")= int 6 ## $ yrbrn : num [1:19690] 1949 1978 1953 1940 1964 ... ## ..- attr(*, "label")= chr "Year of birth" ## ..- attr(*, "format.spss")= chr "F4.0" ## ..- attr(*, "display_width")= int 7 ## $ edulvl : Ord.factor w/ 7 levels "Not completed primary education"<..: NA 4 NA NA 4 NA NA 7 NA 6 ... ## $ eduyrs : num [1:19690] 11 16 14 9 12 18 NA 17 15 17 ... ## ..- attr(*, "label")= chr "Years of full-time education completed" ## ..- attr(*, "format.spss")= chr "F2.0" ## $ polintr : Ord.factor w/ 4 levels "Very interested"<..: 3 3 1 2 3 2 1 4 3 3 ... ## $ lrscale : num [1:19690] 6 7 6 5 8 5 NA 8 5 7 ... ## ..- attr(*, "label")= chr "Placement on left right scale" ## ..- attr(*, "format.spss")= chr "F2.0" ## ..- attr(*, "display_width")= int 9 ## $ country : Factor w/ 10 levels "Austria","Belgium",..: 1 9 1 1 9 1 2 9 1 9 ... ## $ sex : Factor w/ 2 levels "Male","Female": 1 2 1 2 2 1 NA 2 2 1 ... descriptives(ess) Click for explanation Note that the numeric variables are now simple numeric vectors, but the variable labels have been retained as column attributes (which is probably useful). If we want to completely nuke the labelling information, we can use the labelled::remove_labels() function to do so. In addition to screening with summary statistics, we can also visualize the variables’ distributions. You have already created a few such visualizations for single variables. Now, we will use a few tricks to efficiently plot each of our target variables. The first step in this process will be to convert the interesting part of our data from “wide format” (one column per variable) into “long format” (one column of variable names, one column of data values). The pivot_longer() function from the tidyr package provides a convenient way to execute this conversion. 4.3.5 Use tidyr::pivot_longer() to create a long-formatted data frame from the target variables in ess. The target variables are all columns from trstlgl to rfgbfml. Click to show code ## Load the tidyr package: library(tidyr) ## Convert the target variables into a long-formatted data frame: ess_plot <- pivot_longer(ess, cols = trstlgl:rfgbfml, # Which columns to convert names_to = "variable", # Name for the new grouping variable values_to = "value") # Name for the column of stacked values The next step in the process will be to plot the variables using ggplot(). In the above code, I’ve named the new grouping variable variable and the new stacked data variable value. So, to create one plot for each (original, wide-format) variable, we will use the facet_wrap() function to facet the plots of value on the variable column (i.e., create a separate conditional plot of value for each unique value in variable). 4.3.6 Use ggplot() with an appropriate geom (e.g., geom_histogram(), geom_density(), geom_boxplot()) and facet_wrap() to visualize each of the target variables. Hint: To implement the faceting, simply add facet_wrap(~ variable, scales = \"free_x\") to the end of your ggplot() call (obviously, replacing “variable” with whatever you named the grouping variable in your pivot_longer() call). Click to show code library(ggplot2) ggplot(ess_plot, aes(x = value)) + geom_histogram() + # Create a histogram facet_wrap(~ variable, scales = "free_x") # Facet on 'variable' Click for explanation Notice that the variables are actually discrete (i.e., each variable takes only a few integer values). However, most variables look relatively normal despite being categorical. So, we’ll bend the rules a bit and analyze these variables as continuous. It also looks like there’s something weird going on with qfimchr and qfimwht. More on that below. 4.3.7 Check the descriptives for the target variables again. Do you see any remaining issues? Click to show code select(ess, trstlgl:rfgbfml) %>% descriptives() Click for explanation The variables qfimchr and qfimwht both contain values that fall outside the expected range for our survey responses: 77, 88, and 99. In SPSS, these were labeled as “Refusal” “Don’t know” and “No answer” respectively, and would not have contributed to the analysis. 4.3.8 Correct any remaining issues you found above. Click to show code ess <- ess %>% mutate(across(c(qfimchr, qfimwht), na_if, 77)) %>% mutate(across(c(qfimchr, qfimwht), na_if, 88)) %>% mutate(across(c(qfimchr, qfimwht), na_if, 99)) ## Check the results: select(ess, trstlgl:rfgbfml) %>% descriptives() Click to show explanation Here, we need to tell R that these values should be considered missing, or NA. Otherwise they will contribute the numeric value to the analysis, as though someone had provided an answer of 77 on a 10-point scale. We’ve done quite a bit of data processing, and we’ll continue to use these data for several future practicals, so it would be a good idea to save the processed dataset for later use. When saving data that you plan to analyze in R, you will usually want to use the R Data Set (RDS) format. Datasets saved in RDS format retain all of their attributes and formatting (e.g., factor are still factors, missing values are coded as NA, etc.). So, you don’t have to redo any data processing before future analyses. 4.3.9 Use the saveRDS() function to save the processed dataset. Click to show code ## Save the processed data: saveRDS(ess, "ess_round1.rds") Now, we’re ready to run the analyses and see if we can replicate the Kestilä (2006) results. 4.3.10 Run two principal component analyses (PCA): one for trust in politics, one for attitudes towards immigration. Use the principal() function from the psych package. Use exactly the same specifications as Kestilä (2006) concerning the estimation method, rotation, number of components extracted, etc. Hints: Remember that you can view the help file for psych::principal() by running ?psych::principal or, if the psych package already loaded, simply running ?principal. When you print the output from psych::principal(), you can use the cut option to hide any factor loadings smaller than a given threshold. You could consider hiding any loadings smaller than those reported by Kestilä (2006) to make the output easier to interpret. Click to show code Trust in politics Kestilä extracted three components with VARIMAX rotation. ## Load the psych package: library(psych) ## Run the PCA: pca_trust <- select(ess, trstlgl:trstplt) %>% principal(nfactors = 3, rotate = "varimax") ## Print the results: print(pca_trust, cut = 0.3, digits = 3) ## Principal Components Analysis ## Call: principal(r = ., nfactors = 3, rotate = "varimax") ## Standardized loadings (pattern matrix) based upon correlation matrix ## RC3 RC2 RC1 h2 u2 com ## trstlgl 0.779 0.669 0.331 1.21 ## trstplc 0.761 0.633 0.367 1.18 ## trstun 0.675 0.556 0.444 1.44 ## trstep 0.651 0.332 0.549 0.451 1.57 ## trstprl 0.569 0.489 0.650 0.350 2.49 ## stfhlth 0.745 0.567 0.433 1.04 ## stfedu 0.750 0.603 0.397 1.14 ## stfeco 0.711 0.300 0.616 0.384 1.44 ## stfgov 0.634 0.377 0.587 0.413 1.88 ## stfdem 0.369 0.568 0.325 0.564 0.436 2.38 ## pltinvt 0.817 0.695 0.305 1.08 ## pltcare 0.811 0.695 0.305 1.11 ## trstplt 0.510 0.611 0.716 0.284 2.40 ## ## RC3 RC2 RC1 ## SS loadings 2.942 2.668 2.490 ## Proportion Var 0.226 0.205 0.192 ## Cumulative Var 0.226 0.432 0.623 ## Proportion Explained 0.363 0.329 0.307 ## Cumulative Proportion 0.363 0.693 1.000 ## ## Mean item complexity = 1.6 ## Test of the hypothesis that 3 components are sufficient. ## ## The root mean square of the residuals (RMSR) is 0.07 ## with the empirical chi square 15240.94 with prob < 0 ## ## Fit based upon off diagonal values = 0.967 Attitudes toward immigration Kestilä extracted five components with VARIMAX rotation. pca_att <- select(ess, imsmetn:rfgbfml) %>% principal(nfactors = 5, rotate = "varimax") print(pca_att, cut = 0.3, digits = 3) ## Principal Components Analysis ## Call: principal(r = ., nfactors = 5, rotate = "varimax") ## Standardized loadings (pattern matrix) based upon correlation matrix ## RC2 RC1 RC5 RC3 RC4 h2 u2 com ## imsmetn 0.797 0.725 0.275 1.30 ## imdfetn 0.775 0.794 0.206 1.70 ## eimrcnt 0.827 0.715 0.285 1.09 ## eimpcnt 0.800 0.789 0.211 1.49 ## imrcntr 0.835 0.747 0.253 1.15 ## impcntr 0.777 0.782 0.218 1.63 ## qfimchr 0.813 0.688 0.312 1.08 ## qfimwht 0.752 0.637 0.363 1.26 ## imwgdwn 0.807 0.712 0.288 1.19 ## imhecop 0.747 0.669 0.331 1.42 ## imtcjob 0.569 0.334 0.484 0.516 1.99 ## imbleco 0.703 0.554 0.446 1.25 ## imbgeco 0.698 0.605 0.395 1.52 ## imueclt 0.568 -0.340 0.545 0.455 2.43 ## imwbcnt 0.673 0.633 0.367 1.87 ## imwbcrm 0.655 0.478 0.522 1.23 ## imrsprc 0.614 0.440 0.560 1.34 ## pplstrd 0.324 -0.551 0.468 0.532 2.11 ## vrtrlg -0.345 0.471 0.419 0.581 2.67 ## shrrfg 0.365 -0.352 0.418 0.582 4.16 ## rfgawrk 0.614 0.396 0.604 1.10 ## gvrfgap 0.691 0.559 0.441 1.35 ## rfgfrpc -0.387 0.327 0.673 3.34 ## rfggvfn 0.585 0.417 0.583 1.46 ## rfgbfml 0.596 0.460 0.540 1.61 ## ## RC2 RC1 RC5 RC3 RC4 ## SS loadings 4.374 3.393 2.774 2.199 1.723 ## Proportion Var 0.175 0.136 0.111 0.088 0.069 ## Cumulative Var 0.175 0.311 0.422 0.510 0.579 ## Proportion Explained 0.302 0.235 0.192 0.152 0.119 ## Cumulative Proportion 0.302 0.537 0.729 0.881 1.000 ## ## Mean item complexity = 1.7 ## Test of the hypothesis that 5 components are sufficient. ## ## The root mean square of the residuals (RMSR) is 0.05 ## with the empirical chi square 29496.06 with prob < 0 ## ## Fit based upon off diagonal values = 0.976 Feature engineering (i.e., creating new variables by combining and/or transforming existing variables) is one of the most common applications of PCA. PCA is a dimension reduction technique that distills the most salient information from a set of variables into a (smaller) set of component scores. Hence, PCA can be a good way of creating aggregate items (analogous to weighted scale scores) when the data are not collected with validated scales. Principal component scores are automatically generated when we run the PCA. If we want to use these scores in subsequent analyses (e.g., as predictors in a regression model), we usually add them to our dataset as additional columns. 4.3.11 Add the component scores produced by the analyses you ran above to the ess data frame. Give each component score an informative name, based on your interpretation of the factor loading matrix I.e., What hypothetical construct do you think each component represents given the items that load onto it? Hints: You can use the data.frame() function to join multiple objects into a single data frame. You can use the colnames() function to assign column names to a matrix or data frame. 1. Extract the component scores Click to show code ## Save the component scores in stand-alone matrices: trust_scores <- pca_trust$scores att_scores <- pca_att$scores ## Inspect the result: head(trust_scores) ## RC3 RC2 RC1 ## [1,] NA NA NA ## [2,] 0.09755193 -0.01552183 0.994954 ## [3,] 0.23069626 -1.53162604 -2.022642 ## [4,] NA NA NA ## [5,] -0.21112678 0.84370377 1.200007 ## [6,] 1.86596955 0.31083233 -1.062603 summary(trust_scores) ## RC3 RC2 RC1 ## Min. :-4.035 Min. :-3.706 Min. :-3.139 ## 1st Qu.:-0.527 1st Qu.:-0.652 1st Qu.:-0.649 ## Median : 0.155 Median : 0.094 Median : 0.092 ## Mean : 0.055 Mean : 0.015 Mean : 0.049 ## 3rd Qu.: 0.727 3rd Qu.: 0.742 3rd Qu.: 0.742 ## Max. : 3.302 Max. : 3.452 Max. : 3.539 ## NA's :4912 NA's :4912 NA's :4912 head(att_scores) ## RC2 RC1 RC5 RC3 RC4 ## [1,] 1.9873715 1.3233586 -0.8382499 -0.02172765 -0.0908143 ## [2,] 0.1692841 -1.2178436 -0.5016936 -0.21749066 0.6758844 ## [3,] -0.3630480 0.3260383 -1.5133423 -0.51405480 -2.2071787 ## [4,] NA NA NA NA NA ## [5,] -0.1137484 -0.7891232 -1.4732563 -0.05843873 0.4110692 ## [6,] -0.9195530 2.8231404 -0.3480398 -0.75699796 -1.3230602 summary(att_scores) ## RC2 RC1 RC5 RC3 ## Min. :-3.660 Min. :-3.929 Min. :-3.824 Min. :-2.764 ## 1st Qu.:-0.616 1st Qu.:-0.585 1st Qu.:-0.656 1st Qu.:-0.748 ## Median :-0.085 Median : 0.062 Median :-0.008 Median :-0.121 ## Mean :-0.013 Mean : 0.012 Mean : 0.021 Mean : 0.014 ## 3rd Qu.: 0.680 3rd Qu.: 0.654 3rd Qu.: 0.652 3rd Qu.: 0.698 ## Max. : 3.743 Max. : 4.584 Max. : 4.108 Max. : 4.084 ## NA's :5447 NA's :5447 NA's :5447 NA's :5447 ## RC4 ## Min. :-3.784 ## 1st Qu.:-0.683 ## Median : 0.046 ## Mean : 0.003 ## 3rd Qu.: 0.717 ## Max. : 3.254 ## NA's :5447 Click for explanation The object produced by psych::principal() is simply list, and the component scores are already stored therein. So, to extract the component scores, we simply use the $ operator to extract them. 2. Name the component scores Click to show code ## Check names (note the order): colnames(trust_scores) ## [1] "RC3" "RC2" "RC1" colnames(att_scores) ## [1] "RC2" "RC1" "RC5" "RC3" "RC4" ## Give informative names: colnames(trust_scores) <- c("Trust_Institutions", "Satisfaction", "Trust_Politicians") colnames(att_scores) <- c("Quantity", "Effects", "Refugees", "Diversity", "Economic") 3. Add the component scores to the dataset Click to show code # Add the component scores to the 'ess' data: ess <- data.frame(ess, trust_scores, att_scores) 4.3.12 Were you able to replicate the results of Kestilä (2006)? Click for explanation Yes, more-or-less. Although the exact estimates differ somewhat, the general pattern of factor loadings in Kestilä (2006) matches what we found here. End of At-Home Exercises "],["in-class-exercises-3.html", "4.4 In-Class Exercises", " 4.4 In-Class Exercises In these exercises, we will continue with our re-analysis/replication of the Kestilä (2006) results. Rather than attempting a direct replication, we will now redo the analysis using exploratory factor analysis (EFA). 4.4.1 Load the ess_round1.rds dataset. These are the data that we saved after the data processing in the At-Home Exercises. Click to show code ess <- readRDS("ess_round1.rds") 4.4.2 Kestilä (2006) claimed that running a PCA is a good way to test if the questions in the ESS measure attitudes towards immigration and trust in politics. Based on what you’ve learned from the readings and lectures, do you agree with this position? Click for explanation Hopefully not. PCA is not a method for estimating latent measurement structure; PCA is a dimension reduction technique that tries to summarize a set of data with a smaller set of component scores. If we really want to estimate the factor structure underlying a set of observed variables, we should use EFA. 4.4.3 Suppose you had to construct the trust in politics and attitude towards immigration scales described by Kestilä (2006) based on the theory and background information presented in that article. What type of analysis would you choose? What key factors would influence your decision? Click for explanation We are trying to estimate meaningful latent factors, so EFA would be an appropriate method. The theory presented by Kestilä (2006) did not hypothesize a particular number of factors, so we would need to use appropriate techniques to estimate the best number. In particular, combining information from: Scree plots Parallel analysis Substantive interpretability of the (rotated) factor loadings Since the factors are almost certainly correlated, we should apply an oblique rotation. We will now rerun the two PCAs that you conducted for the At-Home Exercises using EFA. We will estimate the EFA models using the psych::fa() function, but we need to know how many factors to extract. We could simply estimate a range of solutions and compare the results. We can restrict the range of plausible solutions and save some time by first checking/plotting the eigenvalues and running parallel analysis. 4.4.4 Estimate the number of latent factors underlying the Trust items based on the eigenvalues, the scree plot, and parallel analysis. How many factors are suggested by each method? 1. Eigenvalue estimation Click to show code ## Load the psych package: library(psych) ## Run a trivial EFA on the 'trust' items efa_trust0 <- select(ess, trstlgl:trstplt) %>% fa(nfactors = 1, rotate = "none") Click for explanation (EFA) First, we run a trivial EFA using the psych::fa() function to estimate the eigenvalues. We don’t care about the factors yet, so we can extract a single factor. We also don’t care about interpretable solutions, so we don’t need rotation. ## View the estimated eigenvalues: round(efa_trust0$values, digits = 3) ## [1] 4.980 0.716 0.482 0.165 0.069 0.014 -0.066 -0.092 -0.182 -0.207 ## [11] -0.284 -0.296 -0.319 Click for explanation (eigenvalue extraction) We can check the eigenvalues to see what proportion of the observed variance is accounted for by each additional factor we may extract. Since only one eigenvalue is greater than one, the so-called “Kaiser Criterion” would suggest extracting a single factor. The Kaiser Criterion is not a valid way to select the number of factors in EFA. So, we don’t want to rely on this information alone. We can still use the eigenvalues to help us with factor enumeration, though. One way to do so is by plotting the eigenvalues in a scree plot. 2. Scree plot Click to show code Given a vector of estimated eigenvalues, we can create a scree plot using ggplot() and the geom_line() or geom_path() geometry. library(ggplot2) library(magrittr) efa_trust0 %$% data.frame(y = values, x = 1:length(values)) %>% ggplot(aes(x, y)) + geom_line() + xlab("No. of Factors") + ylab("Eigenvalues") We can also use the psych::scree() function to create a scree plot directly from the data. select(ess, trstlgl:trstplt) %>% scree(pc = FALSE) Click for explanation (scree plot) Although the scree plot provides useful information, we need to interpret that information subjectively, and the conclusions are sometimes ambiguous, in this case. In this case, the plot seems to suggest either one or three components, depending on where we consider the “elbow” to lie. As recommended in the lecture, we can also use “parallel analysis” (Horn, 1965) to provide more objective information about the number of factors. We’ll use the psych::fa.parallel() function to implement parallel analysis. Parallel analysis relies on randomly simulated/permuted data, so we should set a seed to make sure our results are reproducible. We can set the fa = \"fa\" option to get only the results for EFA. 3. Parallel Analysis Click to show code ## Set the random number seed: set.seed(235711) ## Run the parallel analysis: pa_trust <- select(ess, trstlgl:trstplt) %>% fa.parallel(fa = "fa") ## Parallel analysis suggests that the number of factors = 6 and the number of components = NA Click for explanation The results of the parallel analysis suggest 6 factors. If you’ve been paying close attention, you may have noticed that we need to compute the eigenvalues from the original data to run parallel analysis. Hence, we don’t actually need to run a separate EFA to estimate the eigenvalues. ## View the eigenvalues estimated during the parallel analysis: pa_trust$fa.values ## [1] 4.97995262 0.71644127 0.48201040 0.16517645 0.06885820 0.01422241 ## [7] -0.06606777 -0.09225113 -0.18231333 -0.20740917 -0.28415857 -0.29573407 ## [13] -0.31877470 ## Compare to the version from the EFA: pa_trust$fa.values - efa_trust0$values ## [1] 0 0 0 0 0 0 0 0 0 0 0 0 0 ## Recreate the scree plot from above: pa_trust %$% data.frame(y = fa.values, x = 1:length(fa.values)) %>% ggplot(aes(x, y)) + geom_line() + xlab("No. of Factors") + ylab("Eigenvalues") Of course, we also see the same scree plot printed as part of the parallel analysis. So, there’s really no reason to create a separate scree plot, at all, if we’re doing parallel analysis. 4. Conclusion Click for explanation The different criteria disagree on how many factors we should extract, but we have narrowed the range. Based on the scree plot and parallel analysis, we should consider solutions for 3 to 6 factors. We need to examine the factor loadings to see which solution makes the most substantive sense. 4.4.5 Do the same analysis for the attitudes toward immigration items. Click to show code This time, we’ll start by running the parallel analysis and get the eigenvalues and scree plot from psych::fa.parallel(). ## Set the seed: set.seed(235711) ## Run parallel analysis on the 'attitudes' items: pa_att <- select(ess, imsmetn:rfgbfml) %>% fa.parallel(fa = "fa") ## Parallel analysis suggests that the number of factors = 7 and the number of components = NA ## Check the eigenvalues: round(pa_att$fa.values, digits = 3) ## [1] 7.895 1.449 0.734 0.533 0.313 0.156 0.121 0.019 -0.001 -0.064 ## [11] -0.083 -0.103 -0.119 -0.131 -0.150 -0.175 -0.185 -0.200 -0.212 -0.233 ## [21] -0.239 -0.247 -0.334 -0.422 -0.427 Click for explanation For the attitudes toward immigration analysis, the results are even more ambiguous than they were for the trust items. The Kaiser Criterion suggests 2 factors. The scree plot is hopelessly ambiguous. At least 3 factors? No more than 9 factors? Parallel analysis suggests 7 factors Based on the scree plot and parallel analysis, it seems reasonable to consider solutions for 3 to 7 factors. Again, we need to check the substantive interpretation to choose the most reasonable solution. To evaluate the substantive interpretability of the different solutions, we need to estimate the full EFA models for each candidate number of factors. We then compare the factor loadings across solutions to see which set of loadings define the most reasonable set of latent variables. 4.4.6 For the trust items, estimate the EFA models for each plausible number of components that you identified above. Use the psych::fa() function to estimate the models. You will need to specify a few key options. The data (including only the variables you want to analyze) The number of factors that you want to extract The rotation method The estimation method The method of estimating factor scores Hint: You can save yourself a lot of typing/copy-pasting (and the attendant chances of errors) by using a for() loop to iterate through numbers of factors. Click to show code ## Define an empty list to hold all of our fitted EFA objects: efa_trust <- list() ## Loop through the interesting numbers of factors and estimate an EFA for each: for(i in 3:6) efa_trust[[as.character(i)]] <- ess %>% select(trstlgl:trstplt) %>% fa(nfactors = i, # Number of factors = Loop index rotate = "promax", # Oblique rotation scores = "Bartlett") # Estimate factor scores with WLS 4.4.7 Repeat the above analysis for the attitudes items. Click to show code efa_att <- list() for(i in 3:7) efa_att[[as.character(i)]] <- ess %>% select(imsmetn:rfgbfml) %>% fa(nfactors = i, rotate = "promax", scores = "Bartlett") 4.4.8 Compare the factor loading matrices from the models estimated from the Trust items, and select the best solution. Hints: The factor loadings are stored in the loadings slot of the object returned by psych::fa(). Looping can also be useful here. Click to show code for(x in efa_trust) print(x$loadings) ## ## Loadings: ## MR3 MR2 MR1 ## trstlgl 0.839 -0.115 ## trstplc 0.763 -0.218 ## trstun 0.579 0.161 ## trstep 0.554 0.198 ## trstprl 0.444 0.342 ## stfhlth 0.656 -0.125 ## stfedu 0.695 -0.157 ## stfeco -0.102 0.704 0.146 ## stfgov 0.593 0.226 ## stfdem 0.183 0.476 0.150 ## pltinvt 0.813 ## pltcare 0.808 ## trstplt 0.330 0.526 ## ## MR3 MR2 MR1 ## SS loadings 2.299 2.016 1.970 ## Proportion Var 0.177 0.155 0.152 ## Cumulative Var 0.177 0.332 0.483 ## ## Loadings: ## MR2 MR1 MR4 MR3 ## trstlgl 0.797 ## trstplc 0.725 ## trstun 0.656 0.113 ## trstep 1.003 -0.175 ## trstprl 0.121 0.455 0.200 0.112 ## stfhlth 0.663 -0.106 ## stfedu 0.704 -0.110 0.100 ## stfeco 0.729 ## stfgov 0.631 0.175 -0.149 ## stfdem 0.501 0.107 0.115 ## pltinvt 0.855 ## pltcare -0.103 0.863 ## trstplt 0.479 0.340 ## ## MR2 MR1 MR4 MR3 ## SS loadings 2.161 1.952 1.722 1.239 ## Proportion Var 0.166 0.150 0.132 0.095 ## Cumulative Var 0.166 0.316 0.449 0.544 ## ## Loadings: ## MR1 MR4 MR5 MR3 MR2 ## trstlgl 0.935 ## trstplc 0.810 ## trstun 0.505 0.168 ## trstep -0.138 1.128 -0.108 -0.154 ## trstprl 0.359 0.250 0.140 0.201 -0.104 ## stfhlth 0.557 ## stfedu 0.752 ## stfeco 0.710 -0.118 0.172 ## stfgov 0.973 -0.132 ## stfdem 0.556 0.153 ## pltinvt 0.882 ## pltcare 0.855 ## trstplt 0.288 0.308 0.313 ## ## MR1 MR4 MR5 MR3 MR2 ## SS loadings 2.019 1.716 1.655 1.674 0.936 ## Proportion Var 0.155 0.132 0.127 0.129 0.072 ## Cumulative Var 0.155 0.287 0.415 0.543 0.615 ## ## Loadings: ## MR5 MR1 MR4 MR3 MR2 MR6 ## trstlgl 0.980 ## trstplc 0.655 ## trstun 0.911 ## trstep -0.116 0.739 0.163 ## trstprl 0.197 0.577 0.138 ## stfhlth 0.614 ## stfedu 0.771 ## stfeco 0.689 -0.123 0.144 ## stfgov 0.891 ## stfdem 0.513 0.144 ## pltinvt 0.816 ## pltcare 0.778 ## trstplt 0.706 0.193 ## ## MR5 MR1 MR4 MR3 MR2 MR6 ## SS loadings 1.606 1.417 1.442 1.327 1.014 0.879 ## Proportion Var 0.124 0.109 0.111 0.102 0.078 0.068 ## Cumulative Var 0.124 0.233 0.343 0.446 0.524 0.591 Click for explanation Note: Any factor loadings with magnitude lower than 0.1 are suppressed in above output. The factor loadings matrix indicates how strongly each latent factor (columns) associates with the observed items (rows). We can interpret these factor loadings in the same way that we would interpret regression coefficients (indeed, a factor analytic model can be viewed as a multivariate regression model wherein the latent factors are the predictors and the observed items are the outcomes). A higher factor loading indicates a stronger association between the item and factor linked by that loading. Items with high factor loadings are “good” indicators of the respective factors. Items with only very low loadings do not provide much information about any factor. You may want to exclude such items from your analysis. Note that the size of the factor loadings depends on the number of factors. So, you should only consider excluding an observed item after you have chosen the number of latent factors. When we print the loading matrix, we see additional information printed below the factor loadings. Proportion Var: What proportion of the items’ variance is explained by each of the factors. Cumulative Var: How much variance the factors explain, in total. If you estimated as many factors as items, then the Cumulative Var for the final factor would be 1.00 (i.e., 100%). 4.4.9 Compare the factor loading matrices from the models estimated from the Attitudes items, and select the best solution. Click to show code for(x in efa_att) print(x$loadings) ## ## Loadings: ## MR1 MR2 MR3 ## imsmetn 0.802 ## imdfetn 0.754 0.106 ## eimrcnt 0.843 ## eimpcnt 0.814 ## imrcntr 0.857 ## impcntr 0.769 ## qfimchr 0.235 0.858 ## qfimwht 0.132 0.719 ## imwgdwn 0.293 -0.181 ## imhecop 0.371 -0.162 ## imtcjob 0.619 ## imbleco 0.702 ## imbgeco 0.687 ## imueclt 0.561 -0.207 ## imwbcnt 0.732 ## imwbcrm 0.637 ## imrsprc -0.494 -0.125 ## pplstrd 0.249 -0.413 ## vrtrlg -0.275 0.240 ## shrrfg 0.514 -0.111 ## rfgawrk -0.386 ## gvrfgap -0.601 -0.148 ## rfgfrpc 0.432 ## rfggvfn -0.489 ## rfgbfml -0.545 ## ## MR1 MR2 MR3 ## SS loadings 4.819 3.950 1.683 ## Proportion Var 0.193 0.158 0.067 ## Cumulative Var 0.193 0.351 0.418 ## ## Loadings: ## MR2 MR4 MR1 MR3 ## imsmetn 0.788 ## imdfetn 0.731 0.153 0.110 ## eimrcnt 0.855 -0.143 ## eimpcnt 0.790 0.165 ## imrcntr 0.860 ## impcntr 0.743 0.182 ## qfimchr -0.122 0.853 ## qfimwht 0.723 ## imwgdwn 0.638 0.264 ## imhecop 0.680 0.217 ## imtcjob 0.633 0.136 ## imbleco 0.563 -0.212 0.153 ## imbgeco 0.604 -0.168 ## imueclt 0.392 -0.236 -0.168 ## imwbcnt 0.526 -0.282 ## imwbcrm 0.397 -0.292 ## imrsprc 0.616 ## pplstrd 0.231 -0.378 ## vrtrlg 0.279 0.264 ## shrrfg 0.299 -0.271 ## rfgawrk 0.452 ## gvrfgap 0.123 0.774 ## rfgfrpc 0.193 -0.281 ## rfggvfn 0.467 ## rfgbfml 0.619 ## ## MR2 MR4 MR1 MR3 ## SS loadings 3.828 2.778 2.570 1.602 ## Proportion Var 0.153 0.111 0.103 0.064 ## Cumulative Var 0.153 0.264 0.367 0.431 ## ## Loadings: ## MR2 MR1 MR5 MR3 MR4 ## imsmetn 0.792 ## imdfetn 0.728 0.169 0.113 ## eimrcnt 0.910 -0.150 -0.237 ## eimpcnt 0.779 0.126 0.213 ## imrcntr 0.910 -0.128 -0.187 ## impcntr 0.731 0.131 0.236 ## qfimchr 0.109 -0.156 0.882 ## qfimwht 0.139 0.736 ## imwgdwn 0.740 ## imhecop 0.700 ## imtcjob 0.543 0.124 0.182 ## imbleco 0.682 0.135 ## imbgeco 0.799 ## imueclt 0.572 -0.202 ## imwbcnt 0.712 ## imwbcrm 0.545 -0.124 ## imrsprc 0.620 ## pplstrd 0.207 -0.396 ## vrtrlg -0.198 0.151 0.285 0.116 ## shrrfg 0.208 -0.263 0.139 ## rfgawrk 0.457 ## gvrfgap 0.783 ## rfgfrpc -0.338 0.156 ## rfggvfn 0.477 ## rfgbfml -0.125 0.538 ## ## MR2 MR1 MR5 MR3 MR4 ## SS loadings 3.970 2.790 2.215 1.693 1.166 ## Proportion Var 0.159 0.112 0.089 0.068 0.047 ## Cumulative Var 0.159 0.270 0.359 0.427 0.473 ## ## Loadings: ## MR2 MR1 MR6 MR3 MR5 MR4 ## imsmetn 0.705 0.166 ## imdfetn 0.833 ## eimrcnt 0.249 0.859 ## eimpcnt 0.946 ## imrcntr 0.456 0.517 ## impcntr 0.951 ## qfimchr 0.134 -0.122 0.875 ## qfimwht 0.151 0.725 ## imwgdwn 0.748 ## imhecop 0.678 ## imtcjob 0.566 0.123 0.175 ## imbleco 0.753 0.144 ## imbgeco 0.822 ## imueclt 0.580 -0.201 ## imwbcnt 0.751 ## imwbcrm 0.597 ## imrsprc 0.146 0.527 ## pplstrd 0.204 -0.392 ## vrtrlg -0.204 0.143 0.281 0.115 ## shrrfg 0.198 -0.275 0.141 ## rfgawrk 0.517 ## gvrfgap 0.784 ## rfgfrpc -0.294 0.144 ## rfggvfn 0.512 ## rfgbfml 0.596 ## ## MR2 MR1 MR6 MR3 MR5 MR4 ## SS loadings 3.304 3.013 1.994 1.649 1.065 1.133 ## Proportion Var 0.132 0.121 0.080 0.066 0.043 0.045 ## Cumulative Var 0.132 0.253 0.332 0.398 0.441 0.486 ## ## Loadings: ## MR2 MR1 MR6 MR3 MR5 MR7 MR4 ## imsmetn 0.700 0.162 ## imdfetn 0.821 ## eimrcnt 0.245 0.879 ## eimpcnt 0.935 ## imrcntr 0.452 0.523 ## impcntr 0.938 ## qfimchr 0.751 ## qfimwht 0.720 ## imwgdwn 0.700 ## imhecop 0.172 0.624 ## imtcjob 0.574 -0.120 0.174 ## imbleco 0.679 0.108 ## imbgeco 0.832 -0.145 ## imueclt 0.531 -0.191 ## imwbcnt 0.649 0.138 ## imwbcrm 0.464 0.131 0.290 ## imrsprc 0.146 0.440 -0.100 ## pplstrd -0.274 0.392 ## vrtrlg -0.121 0.190 -0.297 0.115 ## shrrfg -0.124 0.437 0.131 ## rfgawrk 0.538 ## gvrfgap 0.616 -0.237 ## rfgfrpc -0.131 0.437 0.135 ## rfggvfn 0.504 ## rfgbfml 0.526 ## ## MR2 MR1 MR6 MR3 MR5 MR7 MR4 ## SS loadings 3.224 2.467 1.456 1.305 1.105 0.901 0.984 ## Proportion Var 0.129 0.099 0.058 0.052 0.044 0.036 0.039 ## Cumulative Var 0.129 0.228 0.286 0.338 0.382 0.418 0.458 It is very possible that you selected a different numbers of factors than Kestilä (2006). We need to keep these exercises consistent, though. So, the remaining questions will all assume you have extract three factors from the Trust items and five factors from the Attitudes items, to parallel the Kestilä (2006) results. ## Select the three-factor solution for 'trust': efa_trust <- efa_trust[["3"]] ## Select the five-factor solution for 'attitudes': efa_att <- efa_att[["5"]] 4.4.10 Give the factor scores meaningful names, and add the scores to the ess dataset as new columns. Hint: If you’re not sure of what do to, check 4.3.11. Click to show code ## Rename the factor scores: colnames(efa_trust$scores) <- c("trust_inst", "satisfy", "trust_pol") colnames(efa_att$scores) <- c("effects", "allowance", "refugees", "ethnic", "europe") ## Add factor scores to the dataset as new columns: ess <- data.frame(ess, efa_trust$scores, efa_att$scores) Kestilä (2006) used the component scores to descriptively evaluate country-level differences in Attitudes toward Immigration and Political Trust. So, now it’s time to replicate those analyses. 4.4.11 Repeat the Kestilä (2006) between-country comparison using the factor scores you created in 4.4.10 and an appropriate statistical test. Click to show code Here, we’ll only demonstrate a possible approach to analyzing one of the Trust dimensions. We can use a linear model to test whether the countries differ in average levels of Trust in Institutions (as quantified by the relevant factor score). ## Estimate the model: out <- lm(trust_inst ~ country, data = ess) ## View the regression-style summary: summary(out) ## ## Call: ## lm(formula = trust_inst ~ country, data = ess) ## ## Residuals: ## Min 1Q Median 3Q Max ## -4.2295 -0.6226 0.1171 0.7194 3.3061 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) -0.09028 0.02445 -3.692 0.000224 *** ## countryBelgium -0.28923 0.03642 -7.942 2.12e-15 *** ## countryGermany -0.05966 0.03211 -1.858 0.063205 . ## countryDenmark 0.75509 0.03882 19.452 < 2e-16 *** ## countryFinland 0.59235 0.03439 17.224 < 2e-16 *** ## countryItaly 0.10991 0.04071 2.700 0.006939 ** ## countryNetherlands -0.05357 0.03379 -1.585 0.112893 ## countryNorway 0.36922 0.03493 10.570 < 2e-16 *** ## countrySweden 0.28560 0.03613 7.904 2.89e-15 *** ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 1.029 on 14769 degrees of freedom ## (4912 observations deleted due to missingness) ## Multiple R-squared: 0.082, Adjusted R-squared: 0.0815 ## F-statistic: 164.9 on 8 and 14769 DF, p-value: < 2.2e-16 ## View the results as an ANOVA table: anova(out) ## Post-hoc tests out %>% aov() %>% TukeyHSD() ## Tukey multiple comparisons of means ## 95% family-wise confidence level ## ## Fit: aov(formula = .) ## ## $country ## diff lwr upr p adj ## Belgium-Austria -0.289225482 -0.40219224 -0.17625873 0.0000000 ## Germany-Austria -0.059655996 -0.15926604 0.03995405 0.6429963 ## Denmark-Austria 0.755089552 0.63466911 0.87551000 0.0000000 ## Finland-Austria 0.592348290 0.48565882 0.69903776 0.0000000 ## Italy-Austria 0.109910185 -0.01636587 0.23618624 0.1476635 ## Netherlands-Austria -0.053567808 -0.15838407 0.05124846 0.8131104 ## Norway-Austria 0.369224250 0.26085692 0.47759158 0.0000000 ## Sweden-Austria 0.285601197 0.17350905 0.39769334 0.0000000 ## Germany-Belgium 0.229569486 0.12386351 0.33527546 0.0000000 ## Denmark-Belgium 1.044315033 0.91880537 1.16982470 0.0000000 ## Finland-Belgium 0.881573772 0.76917165 0.99397589 0.0000000 ## Italy-Belgium 0.399135667 0.26799745 0.53027389 0.0000000 ## Netherlands-Belgium 0.235657673 0.12503199 0.34628336 0.0000000 ## Norway-Belgium 0.658449732 0.54445381 0.77244566 0.0000000 ## Sweden-Belgium 0.574826679 0.45728417 0.69236918 0.0000000 ## Denmark-Germany 0.814745547 0.70110863 0.92838247 0.0000000 ## Finland-Germany 0.652004286 0.55303505 0.75097352 0.0000000 ## Italy-Germany 0.169566181 0.04974170 0.28939066 0.0003895 ## Netherlands-Germany 0.006088188 -0.09085878 0.10303516 0.9999999 ## Norway-Germany 0.428880246 0.32810453 0.52965596 0.0000000 ## Sweden-Germany 0.345257193 0.24048642 0.45002796 0.0000000 ## Finland-Denmark -0.162741262 -0.28263218 -0.04285034 0.0008579 ## Italy-Denmark -0.645179366 -0.78279052 -0.50756821 0.0000000 ## Netherlands-Denmark -0.808657360 -0.92688442 -0.69043030 0.0000000 ## Norway-Denmark -0.385865301 -0.50725174 -0.26447886 0.0000000 ## Sweden-Denmark -0.469488354 -0.59421139 -0.34476531 0.0000000 ## Italy-Finland -0.482438105 -0.60820928 -0.35666693 0.0000000 ## Netherlands-Finland -0.645916098 -0.75012357 -0.54170862 0.0000000 ## Norway-Finland -0.223124040 -0.33090264 -0.11534544 0.0000000 ## Sweden-Finland -0.306747093 -0.41827017 -0.19522402 0.0000000 ## Netherlands-Italy -0.163477993 -0.28766412 -0.03929186 0.0014719 ## Norway-Italy 0.259314065 0.13211649 0.38651164 0.0000000 ## Sweden-Italy 0.175691012 0.04530545 0.30607657 0.0009794 ## Norway-Netherlands 0.422792059 0.31686740 0.52871671 0.0000000 ## Sweden-Netherlands 0.339169005 0.22943659 0.44890142 0.0000000 ## Sweden-Norway -0.083623053 -0.19675232 0.02950622 0.3462227 Click for explanation According to the omnibus F-test, average levels of Trust in Institutions significantly differ between countries, but this test cannot tell us between which countries the differences lie. Similarly, the t statistics associated with each dummy code in the regression-style summary only tell us if that country differs significantly from the reference country (i.e., Austria), but we cannot see, for example, if there is a significant difference in average trust levels between Belgium and the Netherlands. One way to test for differences between the individual countries would be a post hoc test of all pairwise comparisons. Since we’ll be doing 45 tests, we need to apply a correction for repeated testing. Above, we use the TukeyHSD() function to conduct all pairwise comparisons while applying Tukey’s HSD correction. The TukeyHSD() function only accepts models estimated with the aov() function, so we first pass our fitted lm object through aov(). The second part of the Kestilä (2006) analysis was to evaluate how socio-demographic characteristics affected attitudes towards immigrants and trust in politics among the Finnish electorate. Before we can replicate this part of the analysis, we need to subset the data to only the Finnish cases. 4.4.12 Create a new data frame that contains only the Finnish cases from ess. Hint: You can use logical indexing based on the country variable. Click to show code ess_finland <- filter(ess, country == "Finland") We still have one more step before we can estimate any models. We must prepare our variables for analysis. Our dependent variables will be the factor scores generated above. So, we do not need to apply any further processing. We have not yet used any of the independent variables, though. So, we should inspect those variables to see if they require any processing. In our processed ess data, the relevant variables have the following names: sex yrbrn eduyrs polintr lrscale 4.4.13 Inspect the independent variables listed above. Click to show code library(tidySEM) select(ess_finland, sex, yrbrn, eduyrs, polintr, lrscale) %>% descriptives() Click for explanation It looks like we still need some recoding. 4.4.14 Apply any necessary recoding/transformations. 1. Age Click to show code ess_finland <- mutate(ess_finland, age = 2002 - yrbrn) Click for explanation The data contain the participants’ years of birth instead of their age, but Kestilä analyzed age. Fortunately, we know that the data were collected in 2002, so we can simply subtract each participant’s value of yrbrn from the 2002 to compute their age. 2. Political Interest Click to show code First, we’ll transform polintr. ## Recode the four factor levels into two factor levels: ess_finland <- mutate(ess_finland, polintr_bin = recode_factor(polintr, "Not at all interested" = "Low Interest", "Hardly interested" = "Low Interest", "Quite interested" = "High Interest", "Very interested" = "High Interest") ) ## Check the conversion: with(ess_finland, table(old = polintr, new = polintr_bin, useNA = "always")) ## new ## old Low Interest High Interest <NA> ## Very interested 0 144 0 ## Quite interested 0 785 0 ## Hardly interested 842 0 0 ## Not at all interested 228 0 0 ## <NA> 0 0 1 Click for explanation Kestilä (2006) dichotomized polintr by combining the lowest two and highest two categories. So, we don’t actually want to convert the polint variable into a numeric, Likert-type variable. We want polint to be a binary factor. The recode_factor() function from dplyr() will automatically convert our result into a factor. As with the ess_round1.rds data, we will be coming back to this Finnish subsample data in future practical exercises. So, we should save our work by writing the processed dataset to disk. 4.4.15 Use the saveRDS() function to save the processed Finnish subsample data. Click to see code ## Save the processed Finnish data: saveRDS(ess_finland, "ess_finland.rds") Now, we’re finally ready to replicate the regression analysis from Kestilä (2006). Creating a single aggregate score by summing the individual component scores is a pretty silly thing to do, though. So, we won’t reproduce that aspect of the analysis. 4.4.16 Run a series of multiple linear regression analyses with the factor scores you created in 4.4.10 as the dependent variables and the same predictors used by Kestilä (2006). Do your results agree with those reported by Kestilä (2006)? Click to show code ## Predicting 'Trust in Institutions': out_trust_inst <- lm(trust_inst ~ sex + age + eduyrs + polintr_bin + lrscale, data = ess_finland) summary(out_trust_inst) ## ## Call: ## lm(formula = trust_inst ~ sex + age + eduyrs + polintr_bin + ## lrscale, data = ess_finland) ## ## Residuals: ## Min 1Q Median 3Q Max ## -3.9499 -0.5102 0.1337 0.6638 2.5919 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) -0.057518 0.124294 -0.463 0.643595 ## sexFemale 0.004091 0.045170 0.091 0.927849 ## age -0.003071 0.001380 -2.225 0.026219 * ## eduyrs 0.023223 0.006388 3.635 0.000286 *** ## polintr_binHigh Interest 0.166860 0.046448 3.592 0.000337 *** ## lrscale 0.058951 0.011232 5.249 1.72e-07 *** ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 0.9321 on 1734 degrees of freedom ## (260 observations deleted due to missingness) ## Multiple R-squared: 0.04155, Adjusted R-squared: 0.03879 ## F-statistic: 15.03 on 5 and 1734 DF, p-value: 1.78e-14 ## Predicting 'Trust in Politicians': out_trust_pol <- lm(trust_pol ~ sex + age + eduyrs + polintr_bin + lrscale, data = ess_finland) summary(out_trust_pol) ## ## Call: ## lm(formula = trust_pol ~ sex + age + eduyrs + polintr_bin + lrscale, ## data = ess_finland) ## ## Residuals: ## Min 1Q Median 3Q Max ## -3.03673 -0.67306 0.05346 0.69666 2.38771 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) -0.165989 0.126840 -1.309 0.19083 ## sexFemale 0.015572 0.046095 0.338 0.73554 ## age -0.009112 0.001409 -6.469 1.28e-10 *** ## eduyrs 0.018476 0.006519 2.834 0.00465 ** ## polintr_binHigh Interest 0.463763 0.047399 9.784 < 2e-16 *** ## lrscale 0.054932 0.011462 4.793 1.79e-06 *** ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 0.9512 on 1734 degrees of freedom ## (260 observations deleted due to missingness) ## Multiple R-squared: 0.09806, Adjusted R-squared: 0.09546 ## F-statistic: 37.71 on 5 and 1734 DF, p-value: < 2.2e-16 ## Predicting 'Attitudes toward Refugees': out_refugees <- lm(refugees ~ sex + age + eduyrs + polintr_bin + lrscale, data = ess_finland) summary(out_refugees) ## ## Call: ## lm(formula = refugees ~ sex + age + eduyrs + polintr_bin + lrscale, ## data = ess_finland) ## ## Residuals: ## Min 1Q Median 3Q Max ## -2.9118 -0.6860 -0.0594 0.6904 4.1044 ## ## Coefficients: ## Estimate Std. Error t value Pr(>|t|) ## (Intercept) -1.690e-01 1.438e-01 -1.175 0.240080 ## sexFemale -4.828e-01 5.181e-02 -9.318 < 2e-16 *** ## age 2.903e-05 1.604e-03 0.018 0.985561 ## eduyrs -2.537e-02 7.459e-03 -3.401 0.000688 *** ## polintr_binHigh Interest -2.131e-01 5.345e-02 -3.986 6.99e-05 *** ## lrscale 9.359e-02 1.296e-02 7.223 7.65e-13 *** ## --- ## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1 ## ## Residual standard error: 1.06 on 1699 degrees of freedom ## (295 observations deleted due to missingness) ## Multiple R-squared: 0.09535, Adjusted R-squared: 0.09269 ## F-statistic: 35.81 on 5 and 1699 DF, p-value: < 2.2e-16 That does it for our replication of the Kesilä (2006) analyses, but we still have one more topic to consider in this practical. One of the most common applications of EFA is scale development. Given a pool of items without a known factor structure, we try to estimate the underlying latent factors that define the (sub)scales represented by our items. In such applications, we use the factor loading matrix for our optimal solution to make “bright-line” assignments of items to putative factors according to the simple structure represented by the estimated factor loading matrix. In other words, we disregard small factor loadings and assign observed items to only the single latent factor upon which they load most strongly. We then hypothesize that those items are true indicators of that latent factor. We can use confirmatory factor analysis (which you will learn about next week) to test rigorously this hypothesis, but we can already get started by estimating the internal consistency (a type of reliability) of the hypothesized subscales. 4.4.17 Estimate the internal consistency of the three Trust subscales and five Attitudes subscales implied by your EFA solutions from above. Use Cronbach’s Alpha to quantify internal consistency. Use the alpha() function from the psych package to conduct the analysis. Run your analysis on the full ess dataset, not the Finnish subset. Are the subscales implied by your EFA reliable, in the sense of good internal consistency? Note that \\(\\alpha > 0.7\\) is generally considered acceptable, and \\(\\alpha > 0.8\\) is usually considered good. Click to show code ## Run the reliability analysis on the subscale data: ( out <- select(ess, starts_with("stf")) %>% psych::alpha() ) ## ## Reliability analysis ## Call: psych::alpha(x = .) ## ## raw_alpha std.alpha G6(smc) average_r S/N ase mean sd median_r ## 0.79 0.79 0.77 0.44 3.9 0.0023 5.4 1.7 0.41 ## ## 95% confidence boundaries ## lower alpha upper ## Feldt 0.79 0.79 0.8 ## Duhachek 0.79 0.79 0.8 ## ## Reliability if an item is dropped: ## raw_alpha std.alpha G6(smc) average_r S/N alpha se var.r med.r ## stfhlth 0.78 0.78 0.73 0.47 3.5 0.0026 0.0064 0.46 ## stfedu 0.76 0.76 0.72 0.45 3.2 0.0028 0.0109 0.44 ## stfeco 0.74 0.74 0.70 0.41 2.8 0.0031 0.0069 0.39 ## stfgov 0.74 0.74 0.69 0.42 2.9 0.0030 0.0035 0.41 ## stfdem 0.75 0.75 0.71 0.43 3.0 0.0029 0.0074 0.40 ## ## Item statistics ## n raw.r std.r r.cor r.drop mean sd ## stfhlth 19481 0.69 0.69 0.56 0.50 5.8 2.3 ## stfedu 18844 0.73 0.73 0.62 0.55 5.9 2.3 ## stfeco 19211 0.78 0.78 0.70 0.63 5.0 2.4 ## stfgov 19106 0.77 0.76 0.69 0.61 4.5 2.3 ## stfdem 19106 0.75 0.75 0.67 0.59 5.7 2.3 Click for explanation Here, we estimate the reliability of the Satisfaction subscale from the Trust analysis. According to our EFA, the Satisfaction subscale should be indicated by the following five variables: stfeco stfgov stfdem stfedu stfhlth We select these variables using the tidy-select function starts_with() to extract all variables beginning with the three characters “stf”. To estimate the internal consistency of this subscale, we simply provide a data frame containing only the subscale data to the alpha() function. The raw_alpha value is the estimate of Cronbach’s Alpha. In this case \\(\\alpha = 0.794\\), so the subscale is pretty reliable. The table labeled “Reliability if an item is dropped” shows what Cronbach’s Alpha would be if each item were excluded from the scale. If this value is notably higher than the raw_alpha value, it could indicate a bad item. Note that reliability is only one aspect of scale quality, though. So, you shouldn’t throw out items just because they perform poorly in reliability analysis. End of In-Class Exercises "],["cfa.html", "5 CFA", " 5 CFA This week, we will introduce confirmatory factor analysis (CFA) and discuss how it differs from EFA. Furthermore, we will revisit the idea of model fit and introduce into the R-package lavaan. Homework before the lecture Watch the Lecture Recording for this week. Complete the Reading for this week, and answer the associated reading questions. Homework before the practical Complete the At-Home Exercises. Practical content During the practical you will work on the In-Class Exercises. "],["lecture-4.html", "5.1 Lecture", " 5.1 Lecture Often, we work with scales that have a validated or hypothesized factor structure. In the former case, the scale structure has been validated through previous psychometric studies. In the latter case, we may have conducted an EFA to estimate the factor structure on prior data, or theory/intuition may suggest a plausible structure. Regardless of how we come to expect a given factor structure, such situations represent confirmatory modeling problems, because we are attempting to empirically confirm an a priori expectation. Hence, exploratory methods like EFA are not appropriate, and we should employ confirmatory modeling techniques. This week we consider one such technique: confirmatory factor analysis (CFA). As the name suggests, CFA is related to the EFA methods we discussed last week in that both methods are flavors of factor analysis. However, the two methods address fundamentally different research questions. Rather than attempting to estimate an unknown factor structure (as in EFA), we now want to compare a hypothesized measurement model (i.e., factor structure) to observed data in order to evaluate the model’s plausibility. 5.1.1 Recording Note: When Caspar discusses the complexity of the second-order CFA model, it’s easy to misunderstand his statements. We need to be careful not to over-generalize. In general, a second-order CFA is not more complex than a first-order CFA. Actually, in most practical applications, the opposite is true. A second-order CFA is more complex than a first-order CFA, when the factors in the first-order CFA are uncorrelated. This is the situation Caspar references in the recording when claiming that the second-order model is more complex. We hardly ever want to fit such first-order CFA, though. The default CFA fully saturates the latent covariance structure. If the factors in the first-order CFA are fully correlated (according to standard practice), and we include a single second-order factor, the following statements hold. If the first-order CFA has more than three factors, the first-order model is more complex than the second-order model. If the first-order model has three or fewer factors, the first- and second-order models are equivalent (due to scaling constraints we need to impose to identify the second-order model). The second-order model cannot be more complex than the first-order model (assuming both models are correctly identified and no extra constraints are imposed). The above statements may not hold in more complex situations (e.g., more than one second-order factor, partially saturated first-order correlation structure, etc.). You can always identify the more complex model by calculating the degrees of freedom for both models. The model with fewer degrees of freedom is more complex. 5.1.2 Slides You can download the lecture slides here "],["reading-4.html", "5.2 Reading", " 5.2 Reading Reference Byrne, B. (2005). Factor analytic models: Viewing the structure of an assessment instrument from three perspectives, Journal of Personality Assessment, 85(1), 17–32. Questions What are the main differences between exploratory factor analysis (EFA) and confirmatory factor analysis (CFA)? In which circumstances should a researcher use EFA, and in which should they use CFA? What are the five main limitations of EFA that CFA overcomes? In which circumstances can a second order CFA model be useful? Consider the following four techniques: PCA, EFA, CFA, second order CFA. For each of the following three research situations, which of the above techniques would you use and why? A researcher has developed a new questionnaire that should measure personality and wants to know how many factors underlie the items in their new measure. A researcher is modeling data collected with a seven-item scale that has been used since the 1960s to measure authoritarianism. A researcher has recorded highest completed level of education, years of education, and highest level of education attempted for all respondents in a survey. The researcher wants to include some operationalization of the concept of ‘education’ in their model but is unsure of which observed variable to use. "],["at-home-exercises-4.html", "5.3 At-Home Exercises", " 5.3 At-Home Exercises This week, we will wrap up our re-analysis of the Kestilä (2006) results. During this practical, you will conduct a CFA of the Trust in Politics items and compare the results to those obtained from your previous EFA- and PCA-based replications of Kestilä (2006). 5.3.1 Load the ESS data. The relevant data are contained in the ess_round1.rds file. This file is in R Data Set (RDS) format. The dataset is already stored as a data frame with the processing and cleaning that you should have done for previous practicals completed. Click to show code ess <- readRDS("ess_round1.rds") Although you may have settled on any number of EFA solutions during the Week 4 In-Class Exercises, we are going to base the following CFA on a three-factor model of Trust in Politics similar to the original PCA results from Kestilä (2006). Note: Unless otherwise specified, all following questions refer to the Trust in Politics items. We will not consider the Attitudes toward Immigration items in these exercises. 5.3.2 Define the lavaan model syntax for the CFA implied by the three-factor EFA solution you found in the Week 4 In-Class Exercises. Covary the three latent factors. Do not specify any mean structure. Save this model syntax as an object in your environment. Click to show code mod_3f <- ' institutions =~ trstlgl + trstplc + trstun + trstep + trstprl satisfaction =~ stfhlth + stfedu + stfeco + stfgov + stfdem politicians =~ pltinvt + pltcare + trstplt ' Click for explanation We don’t have to specify the latent covariances in the model syntax, we can tell lavaan to estimate all latent covariances when we fit the model. 5.3.3 Estimate the CFA model you defined above, and summarize the results. Use the lavaan::cfa() function to estimate the model. Use the default settings for the cfa() function. Request the model fit statistics with the summary by supplying the fit.measures = TRUE argument to summary(). Request the standardized parameter estimates with the summary by supplying the standardized = TRUE argument to summary(). Check the results, and answer the following questions: Does the model fit the data well? How are the latent variances and covariances specified when using the default settings? How is the model identified when using the default settings? Click the code ## Load the lavaan package: library(lavaan) ## Estimate the CFA model: fit_3f <- cfa(mod_3f, data = ess) ## Summarize the fitted model: summary(fit_3f, fit.measures = TRUE, standardized = TRUE) ## lavaan 0.6.16 ended normally after 46 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 29 ## ## Used Total ## Number of observations 14778 19690 ## ## Model Test User Model: ## ## Test statistic 10652.207 ## Degrees of freedom 62 ## P-value (Chi-square) 0.000 ## ## Model Test Baseline Model: ## ## Test statistic 81699.096 ## Degrees of freedom 78 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.870 ## Tucker-Lewis Index (TLI) 0.837 ## ## Loglikelihood and Information Criteria: ## ## Loglikelihood user model (H0) -371404.658 ## Loglikelihood unrestricted model (H1) -366078.555 ## ## Akaike (AIC) 742867.317 ## Bayesian (BIC) 743087.743 ## Sample-size adjusted Bayesian (SABIC) 742995.583 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.108 ## 90 Percent confidence interval - lower 0.106 ## 90 Percent confidence interval - upper 0.109 ## P-value H_0: RMSEA <= 0.050 0.000 ## P-value H_0: RMSEA >= 0.080 1.000 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.059 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## institutions =~ ## trstlgl 1.000 1.613 0.677 ## trstplc 0.770 0.012 61.866 0.000 1.241 0.567 ## trstun 0.929 0.013 69.227 0.000 1.498 0.642 ## trstep 0.908 0.013 70.929 0.000 1.464 0.660 ## trstprl 1.139 0.014 84.084 0.000 1.837 0.809 ## satisfaction =~ ## stfhlth 1.000 1.173 0.521 ## stfedu 1.106 0.022 50.840 0.000 1.297 0.577 ## stfeco 1.415 0.025 57.214 0.000 1.659 0.713 ## stfgov 1.480 0.025 58.764 0.000 1.736 0.756 ## stfdem 1.384 0.024 57.904 0.000 1.623 0.731 ## politicians =~ ## pltinvt 1.000 0.646 0.613 ## pltcare 1.021 0.016 62.862 0.000 0.660 0.628 ## trstplt 3.012 0.039 76.838 0.000 1.946 0.891 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## institutions ~~ ## satisfaction 1.391 0.032 43.206 0.000 0.736 0.736 ## politicians 0.909 0.018 49.934 0.000 0.872 0.872 ## satisfaction ~~ ## politicians 0.539 0.013 41.053 0.000 0.711 0.711 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## .trstlgl 3.068 0.041 75.262 0.000 3.068 0.541 ## .trstplc 3.248 0.041 80.037 0.000 3.248 0.678 ## .trstun 3.197 0.041 77.141 0.000 3.197 0.588 ## .trstep 2.776 0.036 76.243 0.000 2.776 0.564 ## .trstprl 1.776 0.029 61.361 0.000 1.776 0.345 ## .stfhlth 3.695 0.046 79.989 0.000 3.695 0.729 ## .stfedu 3.368 0.043 77.916 0.000 3.368 0.667 ## .stfeco 2.656 0.038 69.070 0.000 2.656 0.491 ## .stfgov 2.264 0.035 64.201 0.000 2.264 0.429 ## .stfdem 2.289 0.034 67.172 0.000 2.289 0.465 ## .pltinvt 0.694 0.009 78.255 0.000 0.694 0.624 ## .pltcare 0.668 0.009 77.562 0.000 0.668 0.605 ## .trstplt 0.978 0.028 34.461 0.000 0.978 0.205 ## institutions 2.601 0.059 44.198 0.000 1.000 1.000 ## satisfaction 1.375 0.044 31.407 0.000 1.000 1.000 ## politicians 0.417 0.011 38.843 0.000 1.000 1.000 Click for explanation No, the model does not seem to fit the data well. The SRMR looks good, but one good looking fit statistic is not enough. The RMSEA, TLI, and CFI are all in the “unacceptable” range. The \\(\\chi^2\\) is highly significant, but we don’t care. The cfa() function is just a wrapper for the lavaan() function with several options set at the defaults you would want for a standard CFA. By default: All latent variances and covariances are freely estimated (due to the argument auto.cov.lv.x = TRUE) The model is identified by fixing the first factor loading of each factor to 1 (due to the argument auto.fix.first = TRUE) To see a full list of the (many) options you can specify to tweak the behavior of lavaan estimation functions run ?lavOptions. Now, we will consider a couple of alternative factor structures for the Trust in Politics CFA. First, we will go extremely simple by estimating a one-factor model wherein all Trust items are explained by a single latent variable. 5.3.4 Define the lavaan model syntax for a one-factor model of the Trust items. Save this syntax as an object in your environment. Click to show code mod_1f <- ' political_trust =~ trstlgl + trstplc + trstun + trstep + trstprl + stfhlth + stfedu + stfeco + stfgov + stfdem + pltinvt + pltcare + trstplt ' 5.3.5 Estimate the one-factor model, and summarize the results. Does this model appear to fit better or worse than the three-factor model? Note: You can use the lavaan::fitMeasures() function to extract only the model fit information from a fitted lavaan object. Click to show code ## Estimate the one factor model: fit_1f <- cfa(mod_1f, data = ess) ## Summarize the results: summary(fit_1f, fit.measures = TRUE) ## lavaan 0.6.16 ended normally after 33 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 26 ## ## Used Total ## Number of observations 14778 19690 ## ## Model Test User Model: ## ## Test statistic 17667.304 ## Degrees of freedom 65 ## P-value (Chi-square) 0.000 ## ## Model Test Baseline Model: ## ## Test statistic 81699.096 ## Degrees of freedom 78 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.784 ## Tucker-Lewis Index (TLI) 0.741 ## ## Loglikelihood and Information Criteria: ## ## Loglikelihood user model (H0) -374912.206 ## Loglikelihood unrestricted model (H1) -366078.555 ## ## Akaike (AIC) 749876.413 ## Bayesian (BIC) 750074.036 ## Sample-size adjusted Bayesian (SABIC) 749991.410 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.135 ## 90 Percent confidence interval - lower 0.134 ## 90 Percent confidence interval - upper 0.137 ## P-value H_0: RMSEA <= 0.050 0.000 ## P-value H_0: RMSEA >= 0.080 1.000 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.080 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) ## political_trust =~ ## trstlgl 1.000 ## trstplc 0.774 0.013 57.949 0.000 ## trstun 0.930 0.014 64.200 0.000 ## trstep 0.909 0.014 65.679 0.000 ## trstprl 1.182 0.015 79.401 0.000 ## stfhlth 0.615 0.013 45.947 0.000 ## stfedu 0.695 0.014 51.424 0.000 ## stfeco 0.895 0.014 62.316 0.000 ## stfgov 0.985 0.014 68.200 0.000 ## stfdem 0.998 0.014 70.899 0.000 ## pltinvt 0.382 0.006 59.215 0.000 ## pltcare 0.396 0.006 61.195 0.000 ## trstplt 1.183 0.014 81.716 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .trstlgl 3.370 0.042 79.787 0.000 ## .trstplc 3.410 0.041 82.311 0.000 ## .trstun 3.451 0.043 80.749 0.000 ## .trstep 3.019 0.038 80.272 0.000 ## .trstprl 1.938 0.027 70.878 0.000 ## .stfhlth 4.201 0.050 84.093 0.000 ## .stfedu 3.941 0.047 83.419 0.000 ## .stfeco 3.565 0.044 81.289 0.000 ## .stfgov 3.044 0.038 79.326 0.000 ## .stfdem 2.631 0.034 78.072 0.000 ## .pltinvt 0.775 0.009 82.043 0.000 ## .pltcare 0.743 0.009 81.579 0.000 ## .trstplt 1.548 0.023 67.052 0.000 ## political_trst 2.299 0.055 41.569 0.000 ## Compare fit statistics: fitMeasures(fit_3f) ## npar fmin chisq ## 29.000 0.360 10652.207 ## df pvalue baseline.chisq ## 62.000 0.000 81699.096 ## baseline.df baseline.pvalue cfi ## 78.000 0.000 0.870 ## tli nnfi rfi ## 0.837 0.837 0.836 ## nfi pnfi ifi ## 0.870 0.691 0.870 ## rni logl unrestricted.logl ## 0.870 -371404.658 -366078.555 ## aic bic ntotal ## 742867.317 743087.743 14778.000 ## bic2 rmsea rmsea.ci.lower ## 742995.583 0.108 0.106 ## rmsea.ci.upper rmsea.ci.level rmsea.pvalue ## 0.109 0.900 0.000 ## rmsea.close.h0 rmsea.notclose.pvalue rmsea.notclose.h0 ## 0.050 1.000 0.080 ## rmr rmr_nomean srmr ## 0.255 0.255 0.059 ## srmr_bentler srmr_bentler_nomean crmr ## 0.059 0.059 0.064 ## crmr_nomean srmr_mplus srmr_mplus_nomean ## 0.064 0.059 0.059 ## cn_05 cn_01 gfi ## 113.901 126.971 0.897 ## agfi pgfi mfi ## 0.849 0.611 0.699 ## ecvi ## 0.725 fitMeasures(fit_1f) ## npar fmin chisq ## 26.000 0.598 17667.304 ## df pvalue baseline.chisq ## 65.000 0.000 81699.096 ## baseline.df baseline.pvalue cfi ## 78.000 0.000 0.784 ## tli nnfi rfi ## 0.741 0.741 0.741 ## nfi pnfi ifi ## 0.784 0.653 0.784 ## rni logl unrestricted.logl ## 0.784 -374912.206 -366078.555 ## aic bic ntotal ## 749876.413 750074.036 14778.000 ## bic2 rmsea rmsea.ci.lower ## 749991.410 0.135 0.134 ## rmsea.ci.upper rmsea.ci.level rmsea.pvalue ## 0.137 0.900 0.000 ## rmsea.close.h0 rmsea.notclose.pvalue rmsea.notclose.h0 ## 0.050 1.000 0.080 ## rmr rmr_nomean srmr ## 0.364 0.364 0.080 ## srmr_bentler srmr_bentler_nomean crmr ## 0.080 0.080 0.087 ## crmr_nomean srmr_mplus srmr_mplus_nomean ## 0.087 0.080 0.080 ## cn_05 cn_01 gfi ## 71.949 79.980 0.825 ## agfi pgfi mfi ## 0.756 0.590 0.551 ## ecvi ## 1.199 Click for explanation The one-factor model definitely seems to fit worse than the three-factor model. A second order CFA model is another way of representing the latent structure underlying a set of items. As you read in Byrne (2005), however, the second order CFA is only appropriate in certain circumstances. 5.3.6 Given the CFA results above, would a second order CFA be appropriate for the Trust data? Why or why not? Click for explanation Yes, a second order CFA model is a theoretically appropriate representation of the Trust items. The first order latent variables in the three-factor model are all significantly correlated. The first order latent variables in the three-factor model seem to tap different aspects of some single underlying construct. 5.3.7 Define the lavaan model syntax for a second-order CFA model of the Trust items. Use the three factors defined in 5.3.2 as the first order factors. Click to show code mod_2nd <- ' institutions =~ trstlgl + trstplc + trstun + trstep + trstprl satisfaction =~ stfhlth + stfedu + stfeco + stfgov + stfdem politicians =~ pltinvt + pltcare + trstplt trust =~ politicians + satisfaction + institutions ' Click for explanation To define the second order factor, we use the same syntactic conventions that we employ to define a first order factor. The only differences is that the “indicators” of the second order factor (i.e., the variables listed on the RHS of the =~ operator) are previously defined first order latent variables. 5.3.8 Estimate the second order CFA model, and summarize the results. Does this model fit better or worse than the three-factor model? Is this model more or less complex than the three-factor model? What information can you use to quantify this difference in complexity? Click to show code fit_2nd <- cfa(mod_2nd, data = ess) summary(fit_2nd, fit.measures = TRUE, standardized = TRUE) ## lavaan 0.6.16 ended normally after 44 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 29 ## ## Used Total ## Number of observations 14778 19690 ## ## Model Test User Model: ## ## Test statistic 10652.207 ## Degrees of freedom 62 ## P-value (Chi-square) 0.000 ## ## Model Test Baseline Model: ## ## Test statistic 81699.096 ## Degrees of freedom 78 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.870 ## Tucker-Lewis Index (TLI) 0.837 ## ## Loglikelihood and Information Criteria: ## ## Loglikelihood user model (H0) -371404.658 ## Loglikelihood unrestricted model (H1) -366078.555 ## ## Akaike (AIC) 742867.317 ## Bayesian (BIC) 743087.743 ## Sample-size adjusted Bayesian (SABIC) 742995.583 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.108 ## 90 Percent confidence interval - lower 0.106 ## 90 Percent confidence interval - upper 0.109 ## P-value H_0: RMSEA <= 0.050 0.000 ## P-value H_0: RMSEA >= 0.080 1.000 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.059 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## institutions =~ ## trstlgl 1.000 1.613 0.677 ## trstplc 0.770 0.012 61.866 0.000 1.241 0.567 ## trstun 0.929 0.013 69.227 0.000 1.498 0.642 ## trstep 0.908 0.013 70.929 0.000 1.464 0.660 ## trstprl 1.139 0.014 84.084 0.000 1.837 0.809 ## satisfaction =~ ## stfhlth 1.000 1.173 0.521 ## stfedu 1.106 0.022 50.840 0.000 1.297 0.577 ## stfeco 1.415 0.025 57.214 0.000 1.659 0.713 ## stfgov 1.480 0.025 58.764 0.000 1.736 0.756 ## stfdem 1.384 0.024 57.904 0.000 1.623 0.731 ## politicians =~ ## pltinvt 1.000 0.646 0.613 ## pltcare 1.021 0.016 62.862 0.000 0.660 0.628 ## trstplt 3.012 0.039 76.838 0.000 1.946 0.891 ## trust =~ ## politicians 1.000 0.918 0.918 ## satisfaction 1.531 0.033 46.494 0.000 0.774 0.774 ## institutions 2.583 0.045 56.796 0.000 0.950 0.950 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## .trstlgl 3.068 0.041 75.262 0.000 3.068 0.541 ## .trstplc 3.248 0.041 80.037 0.000 3.248 0.678 ## .trstun 3.197 0.041 77.141 0.000 3.197 0.588 ## .trstep 2.776 0.036 76.243 0.000 2.776 0.564 ## .trstprl 1.776 0.029 61.361 0.000 1.776 0.345 ## .stfhlth 3.695 0.046 79.989 0.000 3.695 0.729 ## .stfedu 3.368 0.043 77.916 0.000 3.368 0.667 ## .stfeco 2.656 0.038 69.070 0.000 2.656 0.491 ## .stfgov 2.264 0.035 64.201 0.000 2.264 0.429 ## .stfdem 2.289 0.034 67.172 0.000 2.289 0.465 ## .pltinvt 0.694 0.009 78.255 0.000 0.694 0.624 ## .pltcare 0.668 0.009 77.562 0.000 0.668 0.605 ## .trstplt 0.978 0.028 34.461 0.000 0.978 0.205 ## .institutions 0.255 0.022 11.691 0.000 0.098 0.098 ## .satisfaction 0.551 0.020 27.846 0.000 0.400 0.400 ## .politicians 0.065 0.004 17.091 0.000 0.157 0.157 ## trust 0.352 0.010 35.005 0.000 1.000 1.000 ## Compare fit between the first and second order models: fitMeasures(fit_3f) ## npar fmin chisq ## 29.000 0.360 10652.207 ## df pvalue baseline.chisq ## 62.000 0.000 81699.096 ## baseline.df baseline.pvalue cfi ## 78.000 0.000 0.870 ## tli nnfi rfi ## 0.837 0.837 0.836 ## nfi pnfi ifi ## 0.870 0.691 0.870 ## rni logl unrestricted.logl ## 0.870 -371404.658 -366078.555 ## aic bic ntotal ## 742867.317 743087.743 14778.000 ## bic2 rmsea rmsea.ci.lower ## 742995.583 0.108 0.106 ## rmsea.ci.upper rmsea.ci.level rmsea.pvalue ## 0.109 0.900 0.000 ## rmsea.close.h0 rmsea.notclose.pvalue rmsea.notclose.h0 ## 0.050 1.000 0.080 ## rmr rmr_nomean srmr ## 0.255 0.255 0.059 ## srmr_bentler srmr_bentler_nomean crmr ## 0.059 0.059 0.064 ## crmr_nomean srmr_mplus srmr_mplus_nomean ## 0.064 0.059 0.059 ## cn_05 cn_01 gfi ## 113.901 126.971 0.897 ## agfi pgfi mfi ## 0.849 0.611 0.699 ## ecvi ## 0.725 fitMeasures(fit_2nd) ## npar fmin chisq ## 29.000 0.360 10652.207 ## df pvalue baseline.chisq ## 62.000 0.000 81699.096 ## baseline.df baseline.pvalue cfi ## 78.000 0.000 0.870 ## tli nnfi rfi ## 0.837 0.837 0.836 ## nfi pnfi ifi ## 0.870 0.691 0.870 ## rni logl unrestricted.logl ## 0.870 -371404.658 -366078.555 ## aic bic ntotal ## 742867.317 743087.743 14778.000 ## bic2 rmsea rmsea.ci.lower ## 742995.583 0.108 0.106 ## rmsea.ci.upper rmsea.ci.level rmsea.pvalue ## 0.109 0.900 0.000 ## rmsea.close.h0 rmsea.notclose.pvalue rmsea.notclose.h0 ## 0.050 1.000 0.080 ## rmr rmr_nomean srmr ## 0.255 0.255 0.059 ## srmr_bentler srmr_bentler_nomean crmr ## 0.059 0.059 0.064 ## crmr_nomean srmr_mplus srmr_mplus_nomean ## 0.064 0.059 0.059 ## cn_05 cn_01 gfi ## 113.901 126.971 0.897 ## agfi pgfi mfi ## 0.849 0.611 0.699 ## ecvi ## 0.725 Click for explanation We don’t have to do anything special here. We can estimate and summarize the second order CFA exactly as we did the first order CFA. You should quickly notice something strange about the model fit statistics compared above. If you don’t see it, consider the following: fitMeasures(fit_3f) - fitMeasures(fit_2nd) ## npar fmin chisq ## 0 0 0 ## df pvalue baseline.chisq ## 0 0 0 ## baseline.df baseline.pvalue cfi ## 0 0 0 ## tli nnfi rfi ## 0 0 0 ## nfi pnfi ifi ## 0 0 0 ## rni logl unrestricted.logl ## 0 0 0 ## aic bic ntotal ## 0 0 0 ## bic2 rmsea rmsea.ci.lower ## 0 0 0 ## rmsea.ci.upper rmsea.ci.level rmsea.pvalue ## 0 0 0 ## rmsea.close.h0 rmsea.notclose.pvalue rmsea.notclose.h0 ## 0 0 0 ## rmr rmr_nomean srmr ## 0 0 0 ## srmr_bentler srmr_bentler_nomean crmr ## 0 0 0 ## crmr_nomean srmr_mplus srmr_mplus_nomean ## 0 0 0 ## cn_05 cn_01 gfi ## 0 0 0 ## agfi pgfi mfi ## 0 0 0 ## ecvi ## 0 The two models produce identical fit statistics! We also see that the degrees of freedom are identical between the two models. Hence, the two models have equal complexity. This result taps into a critical idea in statistical modeling, namely, model equivalency. It turns out the two models we’re comparing here are equivalent in the sense that they are statistically indistinguishable representations of the data. Since this is a very important idea, I want to spend some time discussing it in person. So, spend some time between now and the Week 6 lecture session thinking about the implications of this model equivalence. Specifically, consider the following questions: What do we mean when we say that these two models are equivalent? How is it possible for these two models to be equivalent when one contains an additional latent variable? Why are the degrees of freedom equal for these two models? Why are the fit statistics equal for these two models? We’ll take some time to discuss these ideas in the Week 6 lecture session. End of At-Home Exercises "],["in-class-exercises-4.html", "5.4 In-Class Exercises", " 5.4 In-Class Exercises This week, we will wrap up our re-analysis of the Kestilä (2006) results. During this practical, you will conduct a CFA of the Attitudes toward Immigration items and compare the results to those obtained from your previous EFA- and PCA-based replications of Kestilä (2006). 5.4.1 Load the ESS data. The relevant data are contained in the ess_round1.rds file. Click to show code ess <- readRDS("ess_round1.rds") We are going to conduct a CFA to evaluate the measurement model implied by the five-factor representation of the Attitudes toward Immigration items that you should have found via the EFA you conducted in the Week 4 In-Class Exercises. Caveat: Technically, the following CFA result have no confirmatory value because we’ll be estimating our CFA models from the same data that we used for our EFA. Practicing the techniques will still be useful, though. 5.4.2 Define the lavaan model syntax for the CFA implied by the five-factor solution from 4.4.9. Enforce a simple structure; do not allow any cross-loadings. Covary the five latent factors. Do not specify any mean structure. Save this model syntax as an object in your environment. Hints: You can algorithmically enforce a simple structure by assigning each item to the factor upon which it loads most strongly. You can download the fitted psych::efa() object for the five-factor solution here. The pattern matrix for the five-factor EFA solution in our Week 4 exercises is equivalent to the solution presented in Table 3 of Kestilä (2006). Click to show code mod_5f <- ' ## Immigration Policy: ip =~ imrcntr + eimrcnt + eimpcnt + imsmetn + impcntr + imdfetn ## Social Threat: st =~ imbgeco + imbleco + imwbcnt + imwbcrm + imtcjob + imueclt ## Refugee Policy: rp =~ gvrfgap + imrsprc + rfgbfml + rfggvfn + rfgawrk + rfgfrpc + shrrfg ## Cultural Threat: ct =~ qfimchr + qfimwht + pplstrd + vrtrlg ## Economic Threat: et =~ imwgdwn + imhecop ' Note: We don’t have to specify the latent covariances in the model syntax, we can tell lavaan to estimate all latent covariances when we fit the model. 5.4.3 Estimate the CFA model you defined above, and summarize the results. Use the lavaan::cfa() function to estimate the model. Use the default settings for the cfa() function. Request the model fit statistics with the summary by supplying the fit.measures = TRUE argument to summary(). Request the standardized parameter estimates with the summary by supplying the standardized = TRUE argument to summary(). Check the results, and answer the following questions: Does the model fit the data well? How are the latent variances and covariances specified when using the default settings? How is the model identified when using the default settings? Click to show code ## Load the lavaan package: library(lavaan) ## Estimate the CFA model: fit_5f <- cfa(mod_5f, data = ess) ## Summarize the fitted model: summary(fit_5f, fit.measures = TRUE, standardized = TRUE) ## lavaan 0.6.16 ended normally after 72 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 60 ## ## Used Total ## Number of observations 14243 19690 ## ## Model Test User Model: ## ## Test statistic 18631.556 ## Degrees of freedom 265 ## P-value (Chi-square) 0.000 ## ## Model Test Baseline Model: ## ## Test statistic 159619.058 ## Degrees of freedom 300 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.885 ## Tucker-Lewis Index (TLI) 0.869 ## ## Loglikelihood and Information Criteria: ## ## Loglikelihood user model (H0) -520035.133 ## Loglikelihood unrestricted model (H1) -510719.354 ## ## Akaike (AIC) 1040190.265 ## Bayesian (BIC) 1040644.106 ## Sample-size adjusted Bayesian (SABIC) 1040453.432 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.070 ## 90 Percent confidence interval - lower 0.069 ## 90 Percent confidence interval - upper 0.071 ## P-value H_0: RMSEA <= 0.050 0.000 ## P-value H_0: RMSEA >= 0.080 0.000 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.048 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## ip =~ ## imrcntr 1.000 0.617 0.748 ## eimrcnt 0.942 0.011 84.943 0.000 0.582 0.696 ## eimpcnt 1.127 0.010 113.413 0.000 0.695 0.898 ## imsmetn 0.982 0.010 98.753 0.000 0.606 0.796 ## impcntr 1.150 0.010 113.623 0.000 0.710 0.900 ## imdfetn 1.132 0.010 111.802 0.000 0.698 0.887 ## st =~ ## imbgeco 1.000 1.608 0.728 ## imbleco 0.826 0.012 69.222 0.000 1.327 0.619 ## imwbcnt 1.046 0.012 88.056 0.000 1.682 0.792 ## imwbcrm 0.713 0.011 63.102 0.000 1.146 0.564 ## imtcjob 0.751 0.011 66.787 0.000 1.207 0.597 ## imueclt 1.008 0.013 78.043 0.000 1.621 0.698 ## rp =~ ## gvrfgap 1.000 0.659 0.610 ## imrsprc 0.855 0.016 51.881 0.000 0.563 0.535 ## rfgbfml 1.047 0.019 56.174 0.000 0.690 0.593 ## rfggvfn 0.849 0.016 51.714 0.000 0.559 0.533 ## rfgawrk 0.653 0.016 41.044 0.000 0.430 0.405 ## rfgfrpc -0.810 0.016 -51.095 0.000 -0.534 -0.525 ## shrrfg -0.999 0.017 -58.381 0.000 -0.658 -0.625 ## ct =~ ## qfimchr 1.000 1.836 0.629 ## qfimwht 0.941 0.017 54.250 0.000 1.728 0.659 ## pplstrd -0.366 0.007 -51.585 0.000 -0.673 -0.600 ## vrtrlg 0.252 0.006 41.294 0.000 0.462 0.443 ## et =~ ## imwgdwn 1.000 0.723 0.667 ## imhecop 1.151 0.023 49.736 0.000 0.832 0.771 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## ip ~~ ## st -0.605 0.012 -48.693 0.000 -0.610 -0.610 ## rp 0.264 0.006 45.566 0.000 0.648 0.648 ## ct 0.634 0.015 41.007 0.000 0.560 0.560 ## et -0.206 0.006 -35.411 0.000 -0.462 -0.462 ## st ~~ ## rp -0.838 0.017 -48.329 0.000 -0.792 -0.792 ## ct -1.622 0.041 -39.091 0.000 -0.550 -0.550 ## et 0.675 0.017 39.083 0.000 0.580 0.580 ## rp ~~ ## ct 0.626 0.018 34.950 0.000 0.518 0.518 ## et -0.233 0.007 -33.007 0.000 -0.490 -0.490 ## ct ~~ ## et -0.592 0.020 -30.127 0.000 -0.446 -0.446 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## .imrcntr 0.299 0.004 77.941 0.000 0.299 0.440 ## .eimrcnt 0.359 0.005 79.638 0.000 0.359 0.515 ## .eimpcnt 0.116 0.002 62.821 0.000 0.116 0.193 ## .imsmetn 0.212 0.003 75.580 0.000 0.212 0.366 ## .impcntr 0.119 0.002 62.454 0.000 0.119 0.191 ## .imdfetn 0.132 0.002 65.344 0.000 0.132 0.213 ## .imbgeco 2.288 0.033 70.261 0.000 2.288 0.470 ## .imbleco 2.837 0.037 76.688 0.000 2.837 0.617 ## .imwbcnt 1.677 0.027 63.198 0.000 1.677 0.372 ## .imwbcrm 2.810 0.036 78.612 0.000 2.810 0.682 ## .imtcjob 2.630 0.034 77.524 0.000 2.630 0.643 ## .imueclt 2.761 0.038 72.515 0.000 2.761 0.512 ## .gvrfgap 0.733 0.010 73.584 0.000 0.733 0.628 ## .imrsprc 0.791 0.010 77.119 0.000 0.791 0.714 ## .rfgbfml 0.877 0.012 74.508 0.000 0.877 0.648 ## .rfggvfn 0.788 0.010 77.203 0.000 0.788 0.716 ## .rfgawrk 0.945 0.012 80.870 0.000 0.945 0.836 ## .rfgfrpc 0.749 0.010 77.501 0.000 0.749 0.724 ## .shrrfg 0.676 0.009 72.682 0.000 0.676 0.609 ## .qfimchr 5.142 0.080 64.113 0.000 5.142 0.604 ## .qfimwht 3.891 0.064 60.623 0.000 3.891 0.566 ## .pplstrd 0.804 0.012 67.054 0.000 0.804 0.640 ## .vrtrlg 0.872 0.011 76.990 0.000 0.872 0.804 ## .imwgdwn 0.652 0.012 53.300 0.000 0.652 0.555 ## .imhecop 0.472 0.014 34.353 0.000 0.472 0.405 ## ip 0.381 0.007 51.578 0.000 1.000 1.000 ## st 2.584 0.054 47.795 0.000 1.000 1.000 ## rp 0.434 0.012 36.748 0.000 1.000 1.000 ## ct 3.371 0.096 35.174 0.000 1.000 1.000 ## et 0.523 0.015 34.944 0.000 1.000 1.000 Click for explanation No, the model does not seem to fit the data well. The SRMR looks good, but one good looking fit statistic is not enough. The TLI and CFI are in the “unacceptable” range. RMSEA is in the “questionable” range. The \\(\\chi^2\\) is highly significant, but we don’t care. The cfa() function is just a wrapper for the lavaan() function with several options set at the defaults you would want for a standard CFA. By default: All latent variances and covariances are freely estimated (due to the argument auto.cov.lv.x = TRUE) The model is identified by fixing the first factor loading of each factor to 1 (due to the argument auto.fix.first = TRUE) To see a full list of the (many) options you can specify to tweak the behavior of lavaan estimation functions run ?lavOptions. Now, we will consider a couple of alternative factor structures for the Attitudes toward Immigration CFA. First, we will go extremely simple by estimating a one-factor model wherein all Attitude items are explained by a single latent variable. 5.4.4 Define the lavaan model syntax for a one-factor model of the Immigration items. Save this syntax as an object in your environment. Click to show code mod_1f <- ' ati =~ imrcntr + eimrcnt + eimpcnt + imsmetn + impcntr + imdfetn + imbgeco + imbleco + imwbcnt + imwbcrm + imtcjob + imueclt + gvrfgap + imrsprc + rfgbfml + rfggvfn + rfgawrk + rfgfrpc + shrrfg + qfimchr + qfimwht + pplstrd + vrtrlg + imwgdwn + imhecop ' 5.4.5 Estimate the one-factor model, and summarize the results. Compare the fit measures for the one-factor and five-factor models Which model better fits the data? Note: Remember, you can use the lavaan::fitMeasures() function to extract only the model fit information from a fitted lavaan object. Click to show code ## Estimate the one factor model: fit_1f <- cfa(mod_1f, data = ess) ## Summarize the results: summary(fit_1f) ## lavaan 0.6.16 ended normally after 47 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 50 ## ## Used Total ## Number of observations 14243 19690 ## ## Model Test User Model: ## ## Test statistic 49510.917 ## Degrees of freedom 275 ## P-value (Chi-square) 0.000 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) ## ati =~ ## imrcntr 1.000 ## eimrcnt 0.937 0.012 78.324 0.000 ## eimpcnt 1.114 0.011 101.263 0.000 ## imsmetn 0.987 0.011 90.990 0.000 ## impcntr 1.147 0.011 102.371 0.000 ## imdfetn 1.153 0.011 103.148 0.000 ## imbgeco -2.055 0.032 -64.749 0.000 ## imbleco -1.625 0.031 -52.533 0.000 ## imwbcnt -2.173 0.030 -71.324 0.000 ## imwbcrm -1.432 0.029 -48.849 0.000 ## imtcjob -1.532 0.029 -52.519 0.000 ## imueclt -2.198 0.033 -65.876 0.000 ## gvrfgap 0.807 0.016 51.746 0.000 ## imrsprc 0.757 0.015 49.790 0.000 ## rfgbfml 0.861 0.017 51.272 0.000 ## rfggvfn 0.722 0.015 47.671 0.000 ## rfgawrk 0.530 0.015 34.448 0.000 ## rfgfrpc -0.755 0.015 -51.462 0.000 ## shrrfg -0.931 0.015 -61.438 0.000 ## qfimchr 1.597 0.042 37.835 0.000 ## qfimwht 1.769 0.038 46.697 0.000 ## pplstrd -0.873 0.016 -53.994 0.000 ## vrtrlg 0.602 0.015 39.940 0.000 ## imwgdwn -0.682 0.016 -43.576 0.000 ## imhecop -0.773 0.016 -49.611 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .imrcntr 0.327 0.004 79.021 0.000 ## .eimrcnt 0.388 0.005 80.422 0.000 ## .eimpcnt 0.161 0.002 70.832 0.000 ## .imsmetn 0.235 0.003 77.101 0.000 ## .impcntr 0.158 0.002 69.688 0.000 ## .imdfetn 0.150 0.002 68.791 0.000 ## .imbgeco 3.381 0.041 82.203 0.000 ## .imbleco 3.666 0.044 83.130 0.000 ## .imwbcnt 2.839 0.035 81.477 0.000 ## .imwbcrm 3.399 0.041 83.334 0.000 ## .imtcjob 3.260 0.039 83.130 0.000 ## .imueclt 3.683 0.045 82.092 0.000 ## .gvrfgap 0.938 0.011 83.176 0.000 ## .imrsprc 0.906 0.011 83.285 0.000 ## .rfgbfml 1.092 0.013 83.203 0.000 ## .rfggvfn 0.917 0.011 83.394 0.000 ## .rfgawrk 1.031 0.012 83.913 0.000 ## .rfgfrpc 0.832 0.010 83.192 0.000 ## .shrrfg 0.803 0.010 82.499 0.000 ## .qfimchr 7.613 0.091 83.803 0.000 ## .qfimwht 5.772 0.069 83.442 0.000 ## .pplstrd 0.988 0.012 83.040 0.000 ## .vrtrlg 0.958 0.011 83.728 0.000 ## .imwgdwn 1.010 0.012 83.583 0.000 ## .imhecop 0.954 0.011 83.294 0.000 ## ati 0.353 0.007 48.941 0.000 ## Compare fit statistics: fitMeasures(fit_5f, fit.measures = c("npar", # Estimated parameters "chisq", "df", "pvalue", # Model fit vs. saturated "cfi", "tli", # Model fit vs. baseline "rmsea", "srmr"), # Model fit vs. saturated output = "text") ## ## Model Test User Model: ## ## Test statistic 18631.556 ## Degrees of freedom 265 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.885 ## Tucker-Lewis Index (TLI) 0.869 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.070 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.048 fitMeasures(fit_1f, fit.measures = c("npar", # Estimated parameters "chisq", "df", "pvalue", # Model fit vs. saturated "cfi", "tli", # Model fit vs. baseline "rmsea", "srmr"), # Model fit vs. saturated output = "text") ## ## Model Test User Model: ## ## Test statistic 49510.917 ## Degrees of freedom 275 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.691 ## Tucker-Lewis Index (TLI) 0.663 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.112 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.087 Click for explanation The one-factor model definitely seems to fit worse than the five-factor model. 5.4.6 Given the CFA results from the five factor model, would a second-order CFA be appropriate for the Attitudes towards Immigration data? Why or why not? Click for explanation Yes, a second-order CFA model is a theoretically appropriate representation of the Attitudes towards Immigration items. The first order latent variables in the five-factor model are all significantly correlated. The first order latent variables in the five-factor model seem to tap different aspects of some single underlying construct. 5.4.7 Define the lavaan model syntax for a second-order CFA model of the Attitudes towards Immigration items, estimate it, and inspect the results. Use the five factors defined in 5.4.2 as the first order factors. Click to show code mod_2o <- paste(mod_5f, 'ati =~ ip + rp + st + ct + et', sep = '\\n') fit_2o <- cfa(mod_2o, data = ess) summary(fit_2o, fit.measures = TRUE) ## lavaan 0.6.16 ended normally after 94 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 55 ## ## Used Total ## Number of observations 14243 19690 ## ## Model Test User Model: ## ## Test statistic 19121.111 ## Degrees of freedom 270 ## P-value (Chi-square) 0.000 ## ## Model Test Baseline Model: ## ## Test statistic 159619.058 ## Degrees of freedom 300 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.882 ## Tucker-Lewis Index (TLI) 0.869 ## ## Loglikelihood and Information Criteria: ## ## Loglikelihood user model (H0) -520279.910 ## Loglikelihood unrestricted model (H1) -510719.354 ## ## Akaike (AIC) 1040669.820 ## Bayesian (BIC) 1041085.841 ## Sample-size adjusted Bayesian (SABIC) 1040911.056 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.070 ## 90 Percent confidence interval - lower 0.069 ## 90 Percent confidence interval - upper 0.071 ## P-value H_0: RMSEA <= 0.050 0.000 ## P-value H_0: RMSEA >= 0.080 0.000 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.048 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) ## ip =~ ## imrcntr 1.000 ## eimrcnt 0.943 0.011 85.095 0.000 ## eimpcnt 1.126 0.010 113.523 0.000 ## imsmetn 0.982 0.010 98.910 0.000 ## impcntr 1.149 0.010 113.651 0.000 ## imdfetn 1.130 0.010 111.789 0.000 ## st =~ ## imbgeco 1.000 ## imbleco 0.822 0.012 68.916 0.000 ## imwbcnt 1.047 0.012 88.172 0.000 ## imwbcrm 0.709 0.011 62.846 0.000 ## imtcjob 0.747 0.011 66.424 0.000 ## imueclt 1.013 0.013 78.434 0.000 ## rp =~ ## gvrfgap 1.000 ## imrsprc 0.854 0.017 51.127 0.000 ## rfgbfml 1.048 0.019 55.377 0.000 ## rfggvfn 0.853 0.017 51.170 0.000 ## rfgawrk 0.657 0.016 40.785 0.000 ## rfgfrpc -0.828 0.016 -51.249 0.000 ## shrrfg -1.020 0.017 -58.369 0.000 ## ct =~ ## qfimchr 1.000 ## qfimwht 0.939 0.018 51.902 0.000 ## pplstrd -0.389 0.008 -51.072 0.000 ## vrtrlg 0.271 0.006 41.908 0.000 ## et =~ ## imwgdwn 1.000 ## imhecop 1.158 0.024 48.877 0.000 ## ati =~ ## ip 1.000 ## rp 1.264 0.024 53.732 0.000 ## st -3.123 0.051 -61.058 0.000 ## ct 2.638 0.058 45.467 0.000 ## et -1.000 0.024 -42.490 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .imrcntr 0.299 0.004 77.900 0.000 ## .eimrcnt 0.359 0.005 79.597 0.000 ## .eimpcnt 0.116 0.002 62.698 0.000 ## .imsmetn 0.211 0.003 75.502 0.000 ## .impcntr 0.119 0.002 62.476 0.000 ## .imdfetn 0.133 0.002 65.406 0.000 ## .imbgeco 2.285 0.033 70.158 0.000 ## .imbleco 2.852 0.037 76.762 0.000 ## .imwbcnt 1.668 0.027 62.920 0.000 ## .imwbcrm 2.821 0.036 78.653 0.000 ## .imtcjob 2.646 0.034 77.607 0.000 ## .imueclt 2.734 0.038 72.213 0.000 ## .gvrfgap 0.740 0.010 73.738 0.000 ## .imrsprc 0.797 0.010 77.211 0.000 ## .rfgbfml 0.885 0.012 74.621 0.000 ## .rfggvfn 0.791 0.010 77.189 0.000 ## .rfgawrk 0.946 0.012 80.833 0.000 ## .rfgfrpc 0.741 0.010 77.149 0.000 ## .shrrfg 0.665 0.009 72.020 0.000 ## .qfimchr 5.347 0.081 65.623 0.000 ## .qfimwht 4.084 0.065 62.673 0.000 ## .pplstrd 0.778 0.012 64.838 0.000 ## .vrtrlg 0.854 0.011 75.931 0.000 ## .imwgdwn 0.655 0.012 52.977 0.000 ## .imhecop 0.468 0.014 33.353 0.000 ## .ip 0.177 0.004 44.418 0.000 ## .st 0.596 0.023 26.030 0.000 ## .rp 0.101 0.005 21.784 0.000 ## .ct 1.745 0.060 29.185 0.000 ## .et 0.316 0.010 31.813 0.000 ## ati 0.204 0.005 37.371 0.000 5.4.8 Compare the model fit of the first- and second-order five-factor models using the fitMeasures() function. Which model offers the better fit? Which model is more complex? Click to show code fitMeasures(fit_5f, fit.measures = c("npar", # Estimated parameters "chisq", "df", "pvalue", # Model fit vs. saturated "cfi", "tli", # Model fit vs. baseline "rmsea", "srmr"), # Model fit vs. saturated output = "text") ## ## Model Test User Model: ## ## Test statistic 18631.556 ## Degrees of freedom 265 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.885 ## Tucker-Lewis Index (TLI) 0.869 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.070 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.048 fitMeasures(fit_2o, fit.measures = c("npar", # Estimated parameters "chisq", "df", "pvalue", # Model fit vs. saturated "cfi", "tli", # Model fit vs. baseline "rmsea", "srmr"), # Model fit vs. saturated output = "text") ## ## Model Test User Model: ## ## Test statistic 19121.111 ## Degrees of freedom 270 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.882 ## Tucker-Lewis Index (TLI) 0.869 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.070 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.048 Click for explanation The CFI and TLI are both slightly better in the original five factor model, but the RMSEA and SRMR of both models don’t differ out to three decimal places. As usual, both models have a significant \\(\\chi^2\\), but that doesn’t tell us much. Qualitative comparisons of model fit are fine, but we’d like to have an actual statistical test for these fit differences. As it happens, we have just such a test: a nested model \\(\\Delta \\chi^2\\) test (AKA, chi-squared difference test, change in chi-squared test, likelihood ratio test). In the Week 7 lecture, we’ll cover nested models and tests thereof, but it will be useful to start thinking about these concepts now. Two models are said to be nested if you can define one model by placing constraints on the other model. By way of example, consider the following two CFA models. The second model is nested within the first model, because we can define the second model by fixing the latent covariance to zero in the first model. Notice that the data contain \\(6(6 + 1) / 2 = 21\\) unique pieces of information. The first model estimates 13 parameters, and the second model estimates 12 parameters. Hence the first model has 8 degrees of freedom, and the second model has 9 degrees of freedom. In general, the following must hold whenever Model B is nested within Model A. Model B will have fewer estimated parameters than Model A. Model B will have more degrees of freedom than Model A. Model A will be more complex than model B. Model A will fit the data better than model B. Saturated Model All models are nested within the saturated model, because the saturated model estimates all possible relations among the variables. Regardless of what model we may be considering, we can always convert that model to a saturated model by estimating all possible associations. Hence, all models are nested within the saturated model. Baseline Model Similarly, the baseline model (AKA, independence model) is nested within all other models. In the baseline model, we only estimate the variances of the observed items; all associations are constrained to zero. We can always convert our model to the baseline model by fixing all associations to zero. Hence, the baseline model is nested within all other models. When two models are nested, we can use a \\(\\Delta \\chi^2\\) test to check if the nested model fits significantly worse than its parent model. Whenever we place constraints on the model, the fit will deteriorate, but we want to know if the constraints we imposed to define the nested model have produced too much loss of fit. We can use the anova() function to easily conduct \\(\\Delta \\chi^2\\) tests comparing models that we’ve estimated with cfa() or sem(). 5.4.9 Use the anova() function to compare the five-factor model from 5.4.2 and one-factor model from 5.4.4. Explain what Df, Chisq, Chisq diff, Df diff, and Pr(>Chisq) mean. Which model is more complex? Which model fits better? What is the conclusion of the test? Click to show code anova(fit_1f, fit_5f) Click for explanation The Df column contains the degrees of freedom of each model. Higher df \\(\\Rightarrow\\) Less complex model The Chisq column shows the \\(\\chi^2\\) statistics (AKA, likelihood ratio statistics) for each model. \\(\\chi^2\\) = The ratio of the likelihoods for the estimated model and the saturated model). Larger \\(\\chi^2\\) \\(\\Rightarrow\\) Worse fit Chisq diff is the difference between the two \\(\\chi^2\\) values (i.e., \\(\\Delta \\chi^2\\)). How much better the more complex model fits the data Larger \\(\\Delta \\chi^2\\) values indicate greater losses of fit induced by the constraints needed to define the nested model. Df diff is the difference in the degrees of freedom between the models. Since both models must be estimated from the same pool of variables, this difference also represents the number of parameters that were constrained to define the nested model. Pr(>Chisq) is a p-value for the \\(\\Delta \\chi^2\\) test. \\(H_0: \\Delta \\chi^2 = 0\\) \\(H_1: \\Delta \\chi^2 > 0\\) The five-factor model is more complex than the one-factor model, but the extra complexity is justified The five-factor model fits significantly better than the one-factor model. 5.4.10 Use the anova() function to compare the first- and second-order five-factor models from 5.4.2 and 5.4.7. Which model is more complex? What is the conclusion of the test? Click to show code anova(fit_5f, fit_2o) Click for explanation The first-order model is more complex than the second-order model (df = 265 vs. df = 270), and the extra complexity is necessary. The first-order model fits significantly better than the second-order model. 5.4.11 Based on the results above, would you say that you have successfully confirmed the five-factor structure implied by the EFA? Click for explanation Nope, not so much. The first-order five-factor model may fit the data best out of the three models considered here, but it still fits terribly. None of these models is an adequate representation of the Attitudes toward Immigration items. This result is particularly embarrassing when you consider that we’ve stacked the deck in our favor by using the same data to conduct the EFA and the CFA. When we fail to support the hypothesized measurement model, the confirmatory phase of our analysis is over. At this point, we’ve essentially rejected our hypothesized measurement structure, and that’s the conclusion of our analysis. We don’t have to throw up our hands in despair, however. We can still contribute something useful by modifying the theoretical measurement model through an exploratory, data-driven, post-hoc analysis. We’ll give that a shot below. 5.4.12 Modify the five-factor CFA from 5.4.2 by freeing the following parameters. The residual covariance between imrcntr and eimrcnt These questions both ask about allowing immigration from wealthy countries. It makes sense that answers on these two items share some additional, unique variance above-and-beyond what they contribute to the common factors. The residual covariance between qfimchr and qfimwht These questions are both about imposing qualifications on immigration (specifically Christian religion and “white” race). Click to show code fit_5f_cov <- paste(mod_5f, 'imrcntr ~~ eimrcnt', 'qfimchr ~~ qfimwht', sep = '\\n') %>% cfa(data = ess) summary(fit_5f_cov, fit.measures = TRUE) ## lavaan 0.6.16 ended normally after 77 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 62 ## ## Used Total ## Number of observations 14243 19690 ## ## Model Test User Model: ## ## Test statistic 9740.512 ## Degrees of freedom 263 ## P-value (Chi-square) 0.000 ## ## Model Test Baseline Model: ## ## Test statistic 159619.058 ## Degrees of freedom 300 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.941 ## Tucker-Lewis Index (TLI) 0.932 ## ## Loglikelihood and Information Criteria: ## ## Loglikelihood user model (H0) -515589.611 ## Loglikelihood unrestricted model (H1) -510719.354 ## ## Akaike (AIC) 1031303.221 ## Bayesian (BIC) 1031772.190 ## Sample-size adjusted Bayesian (SABIC) 1031575.160 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.050 ## 90 Percent confidence interval - lower 0.049 ## 90 Percent confidence interval - upper 0.051 ## P-value H_0: RMSEA <= 0.050 0.280 ## P-value H_0: RMSEA >= 0.080 0.000 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.036 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) ## ip =~ ## imrcntr 1.000 ## eimrcnt 0.928 0.007 126.255 0.000 ## eimpcnt 1.184 0.011 106.508 0.000 ## imsmetn 1.012 0.011 92.436 0.000 ## impcntr 1.213 0.011 107.078 0.000 ## imdfetn 1.181 0.011 104.566 0.000 ## st =~ ## imbgeco 1.000 ## imbleco 0.826 0.012 69.006 0.000 ## imwbcnt 1.050 0.012 88.051 0.000 ## imwbcrm 0.715 0.011 63.128 0.000 ## imtcjob 0.751 0.011 66.542 0.000 ## imueclt 1.015 0.013 78.256 0.000 ## rp =~ ## gvrfgap 1.000 ## imrsprc 0.858 0.017 51.965 0.000 ## rfgbfml 1.046 0.019 56.104 0.000 ## rfggvfn 0.848 0.016 51.644 0.000 ## rfgawrk 0.652 0.016 40.998 0.000 ## rfgfrpc -0.813 0.016 -51.233 0.000 ## shrrfg -1.002 0.017 -58.499 0.000 ## ct =~ ## qfimchr 1.000 ## qfimwht 0.979 0.020 48.332 0.000 ## pplstrd -0.586 0.014 -40.685 0.000 ## vrtrlg 0.397 0.011 36.273 0.000 ## et =~ ## imwgdwn 1.000 ## imhecop 1.157 0.023 49.549 0.000 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ## .imrcntr ~~ ## .eimrcnt 0.230 0.004 59.907 0.000 ## .qfimchr ~~ ## .qfimwht 2.558 0.064 40.233 0.000 ## ip ~~ ## st -0.580 0.012 -48.041 0.000 ## rp 0.255 0.006 45.185 0.000 ## ct 0.467 0.014 34.425 0.000 ## et -0.197 0.006 -35.077 0.000 ## st ~~ ## rp -0.835 0.017 -48.285 0.000 ## ct -1.394 0.040 -35.128 0.000 ## et 0.670 0.017 38.935 0.000 ## rp ~~ ## ct 0.538 0.017 32.407 0.000 ## et -0.232 0.007 -32.949 0.000 ## ct ~~ ## et -0.469 0.017 -27.959 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .imrcntr 0.330 0.004 78.903 0.000 ## .eimrcnt 0.396 0.005 80.392 0.000 ## .eimpcnt 0.109 0.002 60.401 0.000 ## .imsmetn 0.220 0.003 75.979 0.000 ## .impcntr 0.107 0.002 58.874 0.000 ## .imdfetn 0.131 0.002 64.630 0.000 ## .imbgeco 2.301 0.033 70.568 0.000 ## .imbleco 2.845 0.037 76.832 0.000 ## .imwbcnt 1.669 0.026 63.272 0.000 ## .imwbcrm 2.808 0.036 78.659 0.000 ## .imtcjob 2.639 0.034 77.663 0.000 ## .imueclt 2.741 0.038 72.463 0.000 ## .gvrfgap 0.734 0.010 73.743 0.000 ## .imrsprc 0.790 0.010 77.164 0.000 ## .rfgbfml 0.880 0.012 74.676 0.000 ## .rfggvfn 0.790 0.010 77.322 0.000 ## .rfgawrk 0.946 0.012 80.924 0.000 ## .rfgfrpc 0.747 0.010 77.519 0.000 ## .shrrfg 0.674 0.009 72.713 0.000 ## .qfimchr 6.815 0.090 75.362 0.000 ## .qfimwht 5.250 0.072 73.378 0.000 ## .pplstrd 0.674 0.013 52.766 0.000 ## .vrtrlg 0.818 0.011 73.191 0.000 ## .imwgdwn 0.655 0.012 53.496 0.000 ## .imhecop 0.468 0.014 33.845 0.000 ## ip 0.350 0.007 48.646 0.000 ## st 2.571 0.054 47.662 0.000 ## rp 0.433 0.012 36.718 0.000 ## ct 1.698 0.073 23.296 0.000 ## et 0.520 0.015 34.814 0.000 5.4.13 Evaluate the model modifications. Did the model fit significantly improve? Is the fit of the modified model acceptable? Click to show code anova(fit_5f_cov, fit_5f) fitMeasures(fit_5f_cov) ## npar fmin chisq ## 62.000 0.342 9740.512 ## df pvalue baseline.chisq ## 263.000 0.000 159619.058 ## baseline.df baseline.pvalue cfi ## 300.000 0.000 0.941 ## tli nnfi rfi ## 0.932 0.932 0.930 ## nfi pnfi ifi ## 0.939 0.823 0.941 ## rni logl unrestricted.logl ## 0.941 -515589.611 -510719.354 ## aic bic ntotal ## 1031303.221 1031772.190 14243.000 ## bic2 rmsea rmsea.ci.lower ## 1031575.160 0.050 0.049 ## rmsea.ci.upper rmsea.ci.level rmsea.pvalue ## 0.051 0.900 0.280 ## rmsea.close.h0 rmsea.notclose.pvalue rmsea.notclose.h0 ## 0.050 0.000 0.080 ## rmr rmr_nomean srmr ## 0.103 0.103 0.036 ## srmr_bentler srmr_bentler_nomean crmr ## 0.036 0.036 0.037 ## crmr_nomean srmr_mplus srmr_mplus_nomean ## 0.037 0.036 0.036 ## cn_05 cn_01 gfi ## 442.344 467.858 0.944 ## agfi pgfi mfi ## 0.931 0.764 0.717 ## ecvi ## 0.693 Click for explanation Yes, the model fit improved significantly. In this case, the original five-factor model is nested within the modified model. So, our \\(\\Delta \\chi^2\\) test is evaluating the improvement in fit contributed by freeing the two residual covariances. The \\(\\Delta \\chi^2\\) test is significant, so we can conclude that including the two new parameter estimates has significantly improved the model fit. I.e., Estimating these two residual covariances is “worth it” in the sense of balancing model fit and model complexity. Also, the fit of the modified model is now acceptable. Caveat If we had found this result when testing our original model, we would be well-situated to proceed with our analysis. In this case, however, we are no longer justified in generalizing these estimates to the population. We only arrived at this well-fitting model by modifying our original theoretical model to better fit the data using estimates derived from those same data to guide our model modifications. We’ve conducted this post-hoc analysis to help inform future research, and this result is useful as a starting point for future studies. Now, anyone analyzing these scales in the future could incorporate these residual covariances into their initial theoretical model. Basically, we conduct these types of post-hoc analyses to help future researchers learn from our mistakes. End of In-Class Exercises "],["full-sem.html", "6 Full SEM", " 6 Full SEM This week, we will focus on integrating all of the disparate methods we’ve covered so far into full-fledged structural equation models. Homework before the lecture Watch the Lecture Recording for this week. Complete the Reading for this week, and answer the associated reading questions. Homework before the practical Complete the At-Home Exercises. Practical content During the practical you will work on the In-Class Exercises. "],["lecture-5.html", "6.1 Lecture", " 6.1 Lecture This week, we will begin with our final theme and discuss structural equation modeling (SEM). This powerful technique joins the strengths of CFA and path analysis to produce a highly flexible and theoretically appealing modeling tool. Essentially, SEM allows us to build structural path models using the latent variables defined by a CFA. 6.1.1 Recording 6.1.2 Slides You can download the lectures slides here "],["reading-5.html", "6.2 Reading", " 6.2 Reading Reference Weston, R. & Gore, P. A. (2006). A brief guide to structural equation modeling. The Counseling Psychologist 34, 719–752. Notes: This article is quite general and provides an overview of things we have discussed so far in this course. This article also also adds an important new idea: combining factor analysis with path modeling to produce a full Structural Equation Model (SEM). Skip the part on GFI (p. 741). The GFI has been shown to be too dependent on sample size and is not recommended any longer. Skip the part on missing data. There is nothing wrong with this section, but missing data analysis is a broad and difficult topic that we cannot adequately cover in this course. If you would like to learn more about missing data and how to treat them, you can take two courses offered by our department: Conducting a Survey Missing Data Theory and Causal Effects Questions The authors state three similarities and two big differences between SEM and other multivariate statistical techniques (e.g., ANCOVA, regression). What are these similarities and differences? Do you agree with the relative strengths and weaknesses of SEM vs. other methods that the authors present? The authors miss at least one additional advantage of SEM over other multivariate methods. What is this missing advantage? Explain what the terms “measurement model” and “structural model” mean in the SEM context. What are the 6 steps of doing an SEM-based analysis given by the authors? The authors claim that testing an SEM using cross-validation is a good idea. When is cross-validation helpful in SEM? Hint: You may have to do some independent (internet, literature) research to learn how cross-validation can be implemented in SEM. "],["at-home-exercises-5.html", "6.3 At-Home Exercises", " 6.3 At-Home Exercises This week, we’ll take another look at the Kestilä (2006) results. During this practical, you will conduct an SEM to replicate the regression analysis of the Finnish data that you conducted in the Week 4 In-Class Exercises. 6.3.1 Load the Finnish subsample of ESS data. The relevant data are contained in the ess_finland.rds file. These are the processed Finnish subsample data from the Week 4 exercises. Note: Unless otherwise noted, all the following analyses use these data. Click to show code ess_fin <- readRDS("ess_finland.rds") We need to do a little data processing before we can fit the regression model. At the moment, lavaan will not automatically convert a factor variable into dummy codes. So, we need to create explicit dummy codes for the two factors we’ll use as predictors in our regression analysis: sex and political orientation. 6.3.2 Convert the sex and political interest factors into dummy codes. Click to show code library(dplyr) ## Create a dummy codes by broadcasting a logical test on the factor levels: ess_fin <- mutate(ess_fin, female = ifelse(sex == "Female", 1, 0), hi_pol_interest = ifelse(polintr_bin == "High Interest", 1, 0) ) ## Check the results: with(ess_fin, table(dummy = female, factor = sex)) ## factor ## dummy Male Female ## 0 960 0 ## 1 0 1040 with(ess_fin, table(dummy = hi_pol_interest, factor = polintr_bin)) ## factor ## dummy Low Interest High Interest ## 0 1070 0 ## 1 0 929 Click for explanation In R, we have several ways of converting a factor into an appropriate set of dummy codes. We could use the dplyr::recode() function as we did last week. We can use the model.matrix() function to define a design matrix based on the inherent contrast attribute of the factor. Missing data will cause problems here. We can us as.numeric() to revert the factor to its underlying numeric representation {Male = 1, Female = 2} and use arithmetic to convert {1, 2} \\(\\rightarrow\\) {0, 1}. When our factor only has two levels, though, the ifelse() function is the simplest way. We are now ready to estimate our latent regression model. Specifically, we want to combine the three OLS regression models that you ran in 4.4.16 into a single SEM that we will estimate in lavaan. The following path diagram shows the intended theoretical model. Although the variances are not included in this path diagram, all variables in the model (including the observed predictor variables) are random. 6.3.3 Define the lavaan model syntax for the SEM shown above. Use the definition of the institutions, satsifaction, and politicians factors from 5.3.2 to define the DVs. Covary the three latent factors. Covary the five predictors. Click to show code mod_sem <- ' ## Define the latent DVs: institutions =~ trstlgl + trstplc + trstun + trstep + trstprl satisfaction =~ stfhlth + stfedu + stfeco + stfgov + stfdem politicians =~ pltinvt + pltcare + trstplt ## Specify the structural relations: institutions + satisfaction + politicians ~ female + age + eduyrs + hi_pol_interest + lrscale ' Click for explanation We simply need to add a line defining the latent regression paths to our old CFA syntax. We don’t need to specify the covariances in the syntax. We can use options in the sem() function to request those estimates. 6.3.4 Estimate the SEM, and summarize the results. Fit the model to the processed Finnish subsample from above. Estimate the model using lavaan::sem(). Request the standardized parameter estimates with the summary. Request the \\(R^2\\) estimates with the summary. Click to show code library(lavaan) ## Fit the SEM: fit_sem <- sem(mod_sem, data = ess_fin, fixed.x = FALSE) ## Summarize the results: summary(fit_sem, fit.measures = TRUE, standardized = TRUE, rsquare = TRUE) ## lavaan 0.6.16 ended normally after 82 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 59 ## ## Used Total ## Number of observations 1740 2000 ## ## Model Test User Model: ## ## Test statistic 1287.421 ## Degrees of freedom 112 ## P-value (Chi-square) 0.000 ## ## Model Test Baseline Model: ## ## Test statistic 10534.649 ## Degrees of freedom 143 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.887 ## Tucker-Lewis Index (TLI) 0.856 ## ## Loglikelihood and Information Criteria: ## ## Loglikelihood user model (H0) -57914.779 ## Loglikelihood unrestricted model (H1) -57271.068 ## ## Akaike (AIC) 115947.557 ## Bayesian (BIC) 116269.794 ## Sample-size adjusted Bayesian (SABIC) 116082.357 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.078 ## 90 Percent confidence interval - lower 0.074 ## 90 Percent confidence interval - upper 0.082 ## P-value H_0: RMSEA <= 0.050 0.000 ## P-value H_0: RMSEA >= 0.080 0.160 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.045 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## institutions =~ ## trstlgl 1.000 1.418 0.669 ## trstplc 0.609 0.031 19.403 0.000 0.863 0.508 ## trstun 0.887 0.038 23.484 0.000 1.257 0.626 ## trstep 1.134 0.041 27.652 0.000 1.607 0.755 ## trstprl 1.192 0.040 29.444 0.000 1.689 0.815 ## satisfaction =~ ## stfhlth 1.000 0.979 0.497 ## stfedu 0.602 0.043 13.872 0.000 0.589 0.416 ## stfeco 1.266 0.067 18.848 0.000 1.240 0.681 ## stfgov 1.639 0.079 20.638 0.000 1.605 0.846 ## stfdem 1.521 0.075 20.180 0.000 1.489 0.793 ## politicians =~ ## pltinvt 1.000 0.567 0.566 ## pltcare 0.953 0.048 19.653 0.000 0.540 0.590 ## trstplt 3.281 0.133 24.675 0.000 1.860 0.915 ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## institutions ~ ## female 0.019 0.073 0.259 0.796 0.013 0.007 ## age -0.008 0.002 -3.740 0.000 -0.006 -0.105 ## eduyrs 0.034 0.010 3.233 0.001 0.024 0.091 ## hi_pol_interst 0.358 0.076 4.730 0.000 0.253 0.126 ## lrscale 0.104 0.018 5.634 0.000 0.073 0.147 ## satisfaction ~ ## female -0.147 0.050 -2.910 0.004 -0.150 -0.075 ## age -0.007 0.002 -4.598 0.000 -0.007 -0.129 ## eduyrs 0.005 0.007 0.775 0.439 0.006 0.022 ## hi_pol_interst 0.164 0.052 3.162 0.002 0.167 0.084 ## lrscale 0.099 0.013 7.501 0.000 0.101 0.202 ## politicians ~ ## female 0.010 0.029 0.349 0.727 0.018 0.009 ## age -0.004 0.001 -4.490 0.000 -0.007 -0.124 ## eduyrs 0.007 0.004 1.697 0.090 0.012 0.047 ## hi_pol_interst 0.258 0.031 8.364 0.000 0.455 0.227 ## lrscale 0.039 0.007 5.370 0.000 0.068 0.138 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## .institutions ~~ ## .satisfaction 1.030 0.069 14.933 0.000 0.796 0.796 ## .politicians 0.675 0.041 16.628 0.000 0.908 0.908 ## .satisfaction ~~ ## .politicians 0.365 0.027 13.544 0.000 0.713 0.713 ## female ~~ ## age 0.071 0.212 0.335 0.738 0.071 0.008 ## eduyrs 0.179 0.046 3.869 0.000 0.179 0.093 ## hi_pol_interst -0.017 0.006 -2.767 0.006 -0.017 -0.066 ## lrscale -0.032 0.024 -1.316 0.188 -0.032 -0.032 ## age ~~ ## eduyrs -22.750 1.722 -13.212 0.000 -22.750 -0.334 ## hi_pol_interst 1.377 0.215 6.413 0.000 1.377 0.156 ## lrscale 1.774 0.853 2.079 0.038 1.774 0.050 ## eduyrs ~~ ## hi_pol_interst 0.270 0.047 5.787 0.000 0.270 0.140 ## lrscale 0.735 0.186 3.946 0.000 0.735 0.095 ## hi_pol_interest ~~ ## lrscale 0.016 0.024 0.672 0.501 0.016 0.016 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## .trstlgl 2.477 0.093 26.743 0.000 2.477 0.552 ## .trstplc 2.140 0.076 28.334 0.000 2.140 0.742 ## .trstun 2.453 0.090 27.322 0.000 2.453 0.608 ## .trstep 1.950 0.078 24.906 0.000 1.950 0.430 ## .trstprl 1.443 0.064 22.437 0.000 1.443 0.336 ## .stfhlth 2.922 0.104 28.103 0.000 2.922 0.753 ## .stfedu 1.663 0.058 28.613 0.000 1.663 0.827 ## .stfeco 1.775 0.069 25.755 0.000 1.775 0.536 ## .stfgov 1.020 0.056 18.371 0.000 1.020 0.284 ## .stfdem 1.307 0.060 21.953 0.000 1.307 0.371 ## .pltinvt 0.682 0.024 27.818 0.000 0.682 0.680 ## .pltcare 0.547 0.020 27.582 0.000 0.547 0.652 ## .trstplt 0.672 0.069 9.676 0.000 0.672 0.163 ## .institutions 1.881 0.125 15.077 0.000 0.936 0.936 ## .satisfaction 0.892 0.086 10.386 0.000 0.930 0.930 ## .politicians 0.294 0.024 12.224 0.000 0.914 0.914 ## female 0.250 0.008 29.496 0.000 0.250 1.000 ## age 313.238 10.620 29.496 0.000 313.238 1.000 ## eduyrs 14.818 0.502 29.496 0.000 14.818 1.000 ## hi_pol_interst 0.250 0.008 29.496 0.000 0.250 1.000 ## lrscale 4.034 0.137 29.496 0.000 4.034 1.000 ## ## R-Square: ## Estimate ## trstlgl 0.448 ## trstplc 0.258 ## trstun 0.392 ## trstep 0.570 ## trstprl 0.664 ## stfhlth 0.247 ## stfedu 0.173 ## stfeco 0.464 ## stfgov 0.716 ## stfdem 0.629 ## pltinvt 0.320 ## pltcare 0.348 ## trstplt 0.837 ## institutions 0.064 ## satisfaction 0.070 ## politicians 0.086 Click for explanation The fixed.x = FALSE argument tells lavaan to model the predictors as random variables. By default, lavaan will covary any random predictor variables. So, we don’t need to make any other changes to the usual procedure. 6.3.5 Finally, we will rerun the latent regression model from above as a path model with the factor scores from 4.4.10 acting as the DVs. Rerun the above SEM as a path model wherein the EFA-derived Trust in Institutions, Satisfaction with Political Systems, and Trust in Politicians factor scores act as the DVs. Request the standardized parameter estimates with the summary. Request the \\(R^2\\) estimates with the summary. Click to show code ## Define the model syntax for the path analysis: mod_pa <- ' trust_inst + satisfy + trust_pol ~ female + age + eduyrs + hi_pol_interest + lrscale' ## Estimate the path model: fit_pa <- sem(mod_pa, data = ess_fin, fixed.x = FALSE) ## Summarize the results: summary(fit_pa, standardized = TRUE, rsquare = TRUE) ## lavaan 0.6.16 ended normally after 44 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 36 ## ## Used Total ## Number of observations 1740 2000 ## ## Model Test User Model: ## ## Test statistic 0.000 ## Degrees of freedom 0 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## trust_inst ~ ## female 0.004 0.045 0.091 0.928 0.004 0.002 ## age -0.003 0.001 -2.229 0.026 -0.003 -0.057 ## eduyrs 0.023 0.006 3.642 0.000 0.023 0.094 ## hi_pol_interst 0.167 0.046 3.599 0.000 0.167 0.088 ## lrscale 0.059 0.011 5.258 0.000 0.059 0.125 ## satisfy ~ ## female -0.125 0.040 -3.115 0.002 -0.125 -0.073 ## age -0.005 0.001 -4.102 0.000 -0.005 -0.105 ## eduyrs -0.003 0.006 -0.534 0.594 -0.003 -0.014 ## hi_pol_interst 0.073 0.041 1.782 0.075 0.073 0.043 ## lrscale 0.085 0.010 8.510 0.000 0.085 0.200 ## trust_pol ~ ## female 0.016 0.046 0.338 0.735 0.016 0.008 ## age -0.009 0.001 -6.480 0.000 -0.009 -0.161 ## eduyrs 0.018 0.007 2.839 0.005 0.018 0.071 ## hi_pol_interst 0.464 0.047 9.801 0.000 0.464 0.232 ## lrscale 0.055 0.011 4.801 0.000 0.055 0.110 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## .trust_inst ~~ ## .satisfy 0.437 0.021 20.609 0.000 0.437 0.568 ## .trust_pol 0.498 0.024 20.480 0.000 0.498 0.564 ## .satisfy ~~ ## .trust_pol 0.367 0.021 17.664 0.000 0.367 0.467 ## female ~~ ## age 0.071 0.212 0.335 0.738 0.071 0.008 ## eduyrs 0.179 0.046 3.869 0.000 0.179 0.093 ## hi_pol_interst -0.017 0.006 -2.767 0.006 -0.017 -0.066 ## lrscale -0.032 0.024 -1.316 0.188 -0.032 -0.032 ## age ~~ ## eduyrs -22.750 1.722 -13.212 0.000 -22.750 -0.334 ## hi_pol_interst 1.377 0.215 6.413 0.000 1.377 0.156 ## lrscale 1.774 0.853 2.079 0.038 1.774 0.050 ## eduyrs ~~ ## hi_pol_interst 0.270 0.047 5.787 0.000 0.270 0.140 ## lrscale 0.735 0.186 3.946 0.000 0.735 0.095 ## hi_pol_interest ~~ ## lrscale 0.016 0.024 0.672 0.501 0.016 0.016 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## .trust_inst 0.866 0.029 29.496 0.000 0.866 0.958 ## .satisfy 0.684 0.023 29.496 0.000 0.684 0.945 ## .trust_pol 0.902 0.031 29.496 0.000 0.902 0.902 ## female 0.250 0.008 29.496 0.000 0.250 1.000 ## age 313.238 10.620 29.496 0.000 313.238 1.000 ## eduyrs 14.818 0.502 29.496 0.000 14.818 1.000 ## hi_pol_interst 0.250 0.008 29.496 0.000 0.250 1.000 ## lrscale 4.034 0.137 29.496 0.000 4.034 1.000 ## ## R-Square: ## Estimate ## trust_inst 0.042 ## satisfy 0.055 ## trust_pol 0.098 Click to show explanation We don’t so anything particularly special here. We simply rerun our latent regression as a path analysis with the EFA-derived factor scores as the DVs. 6.3.6 Compare the results from the path analysis to the SEM-based results. Does it matter whether we use a latent variable or a factor score to define the DV? Hint: When comparing parameter estimates, use the fully standardized estimates (i.e., the values in the column labeled Std.all). Click to show code Note: The “supportFunction.R” script that we source below isn’t a necessary part of the solution. This script defines a bunch of convenience functions. One of these functions, partSummary(), allows us to print selected pieces of the model summary. ## Source a script of convenience function definitions: source("supportFunctions.R") ## View the regression estimates from the SEM: partSummary(fit_sem, 8, standardized = TRUE) ## Regressions: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## institutions ~ ## female 0.019 0.073 0.259 0.796 0.013 0.007 ## age -0.008 0.002 -3.740 0.000 -0.006 -0.105 ## eduyrs 0.034 0.010 3.233 0.001 0.024 0.091 ## hi_pol_interst 0.358 0.076 4.730 0.000 0.253 0.126 ## lrscale 0.104 0.018 5.634 0.000 0.073 0.147 ## satisfaction ~ ## female -0.147 0.050 -2.910 0.004 -0.150 -0.075 ## age -0.007 0.002 -4.598 0.000 -0.007 -0.129 ## eduyrs 0.005 0.007 0.775 0.439 0.006 0.022 ## hi_pol_interst 0.164 0.052 3.162 0.002 0.167 0.084 ## lrscale 0.099 0.013 7.501 0.000 0.101 0.202 ## politicians ~ ## female 0.010 0.029 0.349 0.727 0.018 0.009 ## age -0.004 0.001 -4.490 0.000 -0.007 -0.124 ## eduyrs 0.007 0.004 1.697 0.090 0.012 0.047 ## hi_pol_interst 0.258 0.031 8.364 0.000 0.455 0.227 ## lrscale 0.039 0.007 5.370 0.000 0.068 0.138 ## View the regression estimates from the path analysis: partSummary(fit_pa, 7, standardized = TRUE) ## Regressions: ## Estimate Std.Err z-value P(>|z|) Std.lv Std.all ## trust_inst ~ ## female 0.004 0.045 0.091 0.928 0.004 0.002 ## age -0.003 0.001 -2.229 0.026 -0.003 -0.057 ## eduyrs 0.023 0.006 3.642 0.000 0.023 0.094 ## hi_pol_interst 0.167 0.046 3.599 0.000 0.167 0.088 ## lrscale 0.059 0.011 5.258 0.000 0.059 0.125 ## satisfy ~ ## female -0.125 0.040 -3.115 0.002 -0.125 -0.073 ## age -0.005 0.001 -4.102 0.000 -0.005 -0.105 ## eduyrs -0.003 0.006 -0.534 0.594 -0.003 -0.014 ## hi_pol_interst 0.073 0.041 1.782 0.075 0.073 0.043 ## lrscale 0.085 0.010 8.510 0.000 0.085 0.200 ## trust_pol ~ ## female 0.016 0.046 0.338 0.735 0.016 0.008 ## age -0.009 0.001 -6.480 0.000 -0.009 -0.161 ## eduyrs 0.018 0.007 2.839 0.005 0.018 0.071 ## hi_pol_interst 0.464 0.047 9.801 0.000 0.464 0.232 ## lrscale 0.055 0.011 4.801 0.000 0.055 0.110 ## View the R-squared estimates from the SEM: partSummary(fit_sem, 11, rsquare = TRUE) ## R-Square: ## Estimate ## trstlgl 0.448 ## trstplc 0.258 ## trstun 0.392 ## trstep 0.570 ## trstprl 0.664 ## stfhlth 0.247 ## stfedu 0.173 ## stfeco 0.464 ## stfgov 0.716 ## stfdem 0.629 ## pltinvt 0.320 ## pltcare 0.348 ## trstplt 0.837 ## institutions 0.064 ## satisfaction 0.070 ## politicians 0.086 ## View the R-squared estimates from the SEM: partSummary(fit_pa, 10, rsquare = TRUE) ## R-Square: ## Estimate ## trust_inst 0.042 ## satisfy 0.055 ## trust_pol 0.098 Click for explanation It certainly looks like the way we define the DV has a meaningful impact. The patterns of significance differ between the two sets of regression slopes, and the \\(R^2\\) values are larger for the Institutions and Satisfaction factors in the SEM, and the \\(R^2\\) for the Politicians factor is higher in the path analysis. End of At-Home Exercises "],["in-class-exercises-5.html", "6.4 In-Class Exercises", " 6.4 In-Class Exercises In these exercises, you will use full structural equation modeling (SEM) to evaluate the Theory of Reasoned Action (TORA), which is a popular psychological theory of social behavior developed by Ajzen and Fishbein. The theory states that actual behavior is predicted by behavioral intention, which is in turn predicted by the attitude toward the behavior and subjective norms about the behavior. Later, a third determinant was added, perceived behavioral control. The extent to which people feel that they have control over their behavior also influences their behavior. The data we will use for this practical are available in the toradata.csv file. These data were synthesized according to the results of Reinecke (1998)’s investigation of condom use by young people between 16 and 24 years old. The data contain the following variables: respnr: Numeric participant ID behavior: The dependent variable condom use Measured on a 5-point frequency scale (How often do you…) intent: A single item assessing behavioral intention Measured on a similar 5-point scale (In general, do you intend to…). attit_1:attit_3: Three indicators of attitudes about condom use Measured on a 5-point rating scale (e.g., using a condom is awkward) norm_1:norm_3: Three indicators of social norms about condom use Measured on a 5-point rating scale (e.g., I think most of my friends would use…) control_1:control_3: Three indicators of perceived behavioral control Measured on a 5-point rating scale (e.g., I know well how to use a condom) sex: Binary factor indicating biological sex 6.4.1 Load the data contained in the toradata.csv file. Click to show code condom <- read.csv("toradata.csv", stringsAsFactors = TRUE) 6.4.2 The data contain multiple indicators of attitudes, norms, and control. Run a CFA for these three latent variables. Correlate the latent factors. Do the data support the measurement model for these latent factors? Are the three latent factors significantly correlated? Is it reasonable to proceed with our evaluation of the TORA theory? Click to show code library(lavaan) mod_cfa <- ' attitudes =~ attit_1 + attit_2 + attit_3 norms =~ norm_1 + norm_2 + norm_3 control =~ control_1 + control_2 + control_3 ' fit <- cfa(mod_cfa, data = condom) summary(fit, fit.measures = TRUE) ## lavaan 0.6.16 ended normally after 29 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 21 ## ## Number of observations 250 ## ## Model Test User Model: ## ## Test statistic 35.611 ## Degrees of freedom 24 ## P-value (Chi-square) 0.060 ## ## Model Test Baseline Model: ## ## Test statistic 910.621 ## Degrees of freedom 36 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.987 ## Tucker-Lewis Index (TLI) 0.980 ## ## Loglikelihood and Information Criteria: ## ## Loglikelihood user model (H0) -2998.290 ## Loglikelihood unrestricted model (H1) -2980.484 ## ## Akaike (AIC) 6038.580 ## Bayesian (BIC) 6112.530 ## Sample-size adjusted Bayesian (SABIC) 6045.959 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.044 ## 90 Percent confidence interval - lower 0.000 ## 90 Percent confidence interval - upper 0.073 ## P-value H_0: RMSEA <= 0.050 0.599 ## P-value H_0: RMSEA >= 0.080 0.017 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.037 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) ## attitudes =~ ## attit_1 1.000 ## attit_2 1.036 0.068 15.308 0.000 ## attit_3 -1.002 0.067 -14.856 0.000 ## norms =~ ## norm_1 1.000 ## norm_2 1.031 0.098 10.574 0.000 ## norm_3 0.932 0.093 10.013 0.000 ## control =~ ## control_1 1.000 ## control_2 0.862 0.129 6.699 0.000 ## control_3 0.968 0.133 7.290 0.000 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ## attitudes ~~ ## norms 0.340 0.069 4.957 0.000 ## control 0.475 0.073 6.468 0.000 ## norms ~~ ## control 0.338 0.064 5.254 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .attit_1 0.418 0.052 8.047 0.000 ## .attit_2 0.310 0.047 6.633 0.000 ## .attit_3 0.369 0.049 7.577 0.000 ## .norm_1 0.504 0.071 7.130 0.000 ## .norm_2 0.469 0.071 6.591 0.000 ## .norm_3 0.635 0.075 8.465 0.000 ## .control_1 0.614 0.078 7.905 0.000 ## .control_2 0.865 0.091 9.520 0.000 ## .control_3 0.762 0.087 8.758 0.000 ## attitudes 0.885 0.116 7.620 0.000 ## norms 0.743 0.116 6.423 0.000 ## control 0.497 0.099 5.002 0.000 Click for explanation Yes, the model fits the data well, and the measurement parameters (e.g., factor loadings, residual variances) look reasonable. So, the data seem to support this measurement structure. Yes, all three latent variables are significantly, positively correlated. Yes. The measurement structure is supported, so we can use the latent variables to represent the respective constructs in our subsequent SEM. The TORA doesn’t actually say anything about the associations between these three factors, but it makes sense that they would be positively associated. So, we should find this result comforting. 6.4.3 Estimate the basic TORA model as an SEM. Predict intention from attitudes and norms. Predict condom use from intention. Use the latent versions of attitudes and norms. Covary the attitudes and norms factors. Does the model fit well? Do the estimates align with the TORA? How much variance in intention and condom use are explained by the model? Click to show code mod <- ' ## Define the latent variables: attitudes =~ attit_1 + attit_2 + attit_3 norms =~ norm_1 + norm_2 + norm_3 ## Define the structural model: intent ~ attitudes + norms behavior ~ intent ' fit <- sem(mod, data = condom) summary(fit, fit.measures = TRUE, rsquare = TRUE) ## lavaan 0.6.16 ended normally after 24 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 18 ## ## Number of observations 250 ## ## Model Test User Model: ## ## Test statistic 27.890 ## Degrees of freedom 18 ## P-value (Chi-square) 0.064 ## ## Model Test Baseline Model: ## ## Test statistic 1089.407 ## Degrees of freedom 28 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.991 ## Tucker-Lewis Index (TLI) 0.986 ## ## Loglikelihood and Information Criteria: ## ## Loglikelihood user model (H0) -2533.616 ## Loglikelihood unrestricted model (H1) -2519.671 ## ## Akaike (AIC) 5103.232 ## Bayesian (BIC) 5166.618 ## Sample-size adjusted Bayesian (SABIC) 5109.557 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.047 ## 90 Percent confidence interval - lower 0.000 ## 90 Percent confidence interval - upper 0.079 ## P-value H_0: RMSEA <= 0.050 0.523 ## P-value H_0: RMSEA >= 0.080 0.046 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.036 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) ## attitudes =~ ## attit_1 1.000 ## attit_2 1.039 0.068 15.365 0.000 ## attit_3 -1.002 0.067 -14.850 0.000 ## norms =~ ## norm_1 1.000 ## norm_2 0.983 0.087 11.333 0.000 ## norm_3 0.935 0.087 10.778 0.000 ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## intent ~ ## attitudes 0.439 0.063 6.990 0.000 ## norms 0.693 0.077 8.977 0.000 ## behavior ~ ## intent 0.746 0.045 16.443 0.000 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ## attitudes ~~ ## norms 0.347 0.069 5.027 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .attit_1 0.420 0.052 8.103 0.000 ## .attit_2 0.306 0.046 6.604 0.000 ## .attit_3 0.372 0.049 7.651 0.000 ## .norm_1 0.483 0.064 7.581 0.000 ## .norm_2 0.521 0.065 7.954 0.000 ## .norm_3 0.610 0.070 8.713 0.000 ## .intent 0.423 0.048 8.769 0.000 ## .behavior 0.603 0.054 11.180 0.000 ## attitudes 0.884 0.116 7.614 0.000 ## norms 0.765 0.113 6.767 0.000 ## ## R-Square: ## Estimate ## attit_1 0.678 ## attit_2 0.757 ## attit_3 0.705 ## norm_1 0.613 ## norm_2 0.587 ## norm_3 0.523 ## intent 0.639 ## behavior 0.520 Click for explanation Yes, the model still fits the data very well. Yes, the estimates all align with the TORA. Specifically, attitudes and norms both significantly predict intention, and intention significantly predicts condom use. The model explains 63.93% of the variance in intention and 51.96% of the variance in condom use. 6.4.4 Update your model to represent the extended TORA model that includes perceived behavioral control. Regress condom use onto perceived behavioral control. Use the latent variable representation of control. Covary all three exogenous latent factors. Does the model fit well? Do the estimates align with the updated TORA? How much variance in intention and condom use are explained by the model? Click to show code mod_tora <- ' attitudes =~ attit_1 + attit_2 + attit_3 norms =~ norm_1 + norm_2 + norm_3 control =~ control_1 + control_2 + control_3 intent ~ attitudes + norms behavior ~ intent + control ' fit_tora <- sem(mod_tora, data = condom) summary(fit_tora, fit.measures = TRUE, rsquare = TRUE) ## lavaan 0.6.16 ended normally after 31 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 27 ## ## Number of observations 250 ## ## Model Test User Model: ## ## Test statistic 48.757 ## Degrees of freedom 39 ## P-value (Chi-square) 0.136 ## ## Model Test Baseline Model: ## ## Test statistic 1333.695 ## Degrees of freedom 55 ## P-value 0.000 ## ## User Model versus Baseline Model: ## ## Comparative Fit Index (CFI) 0.992 ## Tucker-Lewis Index (TLI) 0.989 ## ## Loglikelihood and Information Criteria: ## ## Loglikelihood user model (H0) -3551.160 ## Loglikelihood unrestricted model (H1) -3526.782 ## ## Akaike (AIC) 7156.320 ## Bayesian (BIC) 7251.400 ## Sample-size adjusted Bayesian (SABIC) 7165.807 ## ## Root Mean Square Error of Approximation: ## ## RMSEA 0.032 ## 90 Percent confidence interval - lower 0.000 ## 90 Percent confidence interval - upper 0.057 ## P-value H_0: RMSEA <= 0.050 0.870 ## P-value H_0: RMSEA >= 0.080 0.000 ## ## Standardized Root Mean Square Residual: ## ## SRMR 0.033 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) ## attitudes =~ ## attit_1 1.000 ## attit_2 1.033 0.068 15.221 0.000 ## attit_3 -1.025 0.068 -15.097 0.000 ## norms =~ ## norm_1 1.000 ## norm_2 0.984 0.087 11.256 0.000 ## norm_3 0.955 0.088 10.881 0.000 ## control =~ ## control_1 1.000 ## control_2 0.859 0.127 6.789 0.000 ## control_3 0.997 0.131 7.609 0.000 ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## intent ~ ## attitudes 0.447 0.063 7.100 0.000 ## norms 0.706 0.078 9.078 0.000 ## behavior ~ ## intent 0.563 0.063 8.923 0.000 ## control 0.454 0.119 3.805 0.000 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ## attitudes ~~ ## norms 0.342 0.068 5.011 0.000 ## control 0.474 0.072 6.548 0.000 ## norms ~~ ## control 0.352 0.064 5.521 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .attit_1 0.432 0.052 8.381 0.000 ## .attit_2 0.330 0.046 7.220 0.000 ## .attit_3 0.344 0.046 7.439 0.000 ## .norm_1 0.496 0.063 7.820 0.000 ## .norm_2 0.533 0.065 8.152 0.000 ## .norm_3 0.595 0.069 8.643 0.000 ## .control_1 0.625 0.075 8.372 0.000 ## .control_2 0.876 0.090 9.757 0.000 ## .control_3 0.746 0.084 8.874 0.000 ## .intent 0.409 0.047 8.769 0.000 ## .behavior 0.542 0.052 10.423 0.000 ## attitudes 0.872 0.115 7.566 0.000 ## norms 0.751 0.112 6.709 0.000 ## control 0.485 0.096 5.059 0.000 ## ## R-Square: ## Estimate ## attit_1 0.668 ## attit_2 0.738 ## attit_3 0.727 ## norm_1 0.602 ## norm_2 0.577 ## norm_3 0.535 ## control_1 0.437 ## control_2 0.290 ## control_3 0.392 ## intent 0.651 ## behavior 0.566 Click for explanation Yes, the model still fits the data very well. Yes, the estimates all align with the updated TORA. Specifically, attitudes and norms both significantly predict intention, while intention and control both significantly predict condom use. The model explains 65.11% of the variance in intention and 56.62% of the variance in condom use. The TORA model explicitly forbids direct paths from attitudes and norms to behaviors; these effects should be fully mediated by the behavioral intention. The theory does not specify how perceived behavioral control should affect behaviors. There may be a direct effect of control on behavior, or the effect may be (partially) mediated by intention. 6.4.5 Evaluate the hypothesized indirect effects of attitudes and norms. Include attitudes, norms, and control in your model as in 6.4.4. Does intention significantly mediate the effects of attitudes and norms on behavior? Don’t forget to follow all the steps we covered for testing mediation. Are both of the above effects completely mediated? Do these results comport with the TORA? Why or why not? Click for explanation mod <- ' attitudes =~ attit_1 + attit_2 + attit_3 norms =~ norm_1 + norm_2 + norm_3 control =~ control_1 + control_2 + control_3 intent ~ a1 * attitudes + a2 * norms behavior ~ b * intent + control + attitudes + norms ie_att := a1 * b ie_norm := a2 * b ' set.seed(235711) fit <- sem(mod, data = condom, se = "bootstrap", bootstrap = 1000) summary(fit, ci = TRUE) ## lavaan 0.6.16 ended normally after 36 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 29 ## ## Number of observations 250 ## ## Model Test User Model: ## ## Test statistic 48.629 ## Degrees of freedom 37 ## P-value (Chi-square) 0.096 ## ## Parameter Estimates: ## ## Standard errors Bootstrap ## Number of requested bootstrap draws 1000 ## Number of successful bootstrap draws 1000 ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) ci.lower ci.upper ## attitudes =~ ## attit_1 1.000 1.000 1.000 ## attit_2 1.033 0.060 17.261 0.000 0.925 1.165 ## attit_3 -1.025 0.064 -15.894 0.000 -1.163 -0.902 ## norms =~ ## norm_1 1.000 1.000 1.000 ## norm_2 0.984 0.071 13.794 0.000 0.843 1.127 ## norm_3 0.955 0.093 10.324 0.000 0.792 1.157 ## control =~ ## control_1 1.000 1.000 1.000 ## control_2 0.860 0.113 7.624 0.000 0.653 1.098 ## control_3 0.996 0.147 6.790 0.000 0.748 1.320 ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ci.lower ci.upper ## intent ~ ## attitudes (a1) 0.447 0.067 6.674 0.000 0.324 0.585 ## norms (a2) 0.706 0.078 9.094 0.000 0.569 0.878 ## behavior ~ ## intent (b) 0.545 0.075 7.282 0.000 0.389 0.686 ## control 0.428 0.232 1.847 0.065 0.046 0.934 ## attitudes 0.010 0.122 0.084 0.933 -0.249 0.226 ## norms 0.041 0.118 0.345 0.730 -0.194 0.266 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ci.lower ci.upper ## attitudes ~~ ## norms 0.342 0.070 4.883 0.000 0.208 0.480 ## control 0.475 0.069 6.850 0.000 0.344 0.612 ## norms ~~ ## control 0.350 0.067 5.218 0.000 0.221 0.484 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ci.lower ci.upper ## .attit_1 0.432 0.050 8.720 0.000 0.331 0.526 ## .attit_2 0.330 0.045 7.382 0.000 0.238 0.415 ## .attit_3 0.343 0.049 6.992 0.000 0.244 0.444 ## .norm_1 0.496 0.060 8.305 0.000 0.376 0.614 ## .norm_2 0.533 0.077 6.951 0.000 0.390 0.687 ## .norm_3 0.594 0.069 8.597 0.000 0.443 0.719 ## .control_1 0.624 0.076 8.216 0.000 0.477 0.763 ## .control_2 0.875 0.092 9.495 0.000 0.686 1.052 ## .control_3 0.745 0.079 9.398 0.000 0.574 0.889 ## .intent 0.409 0.050 8.169 0.000 0.309 0.507 ## .behavior 0.544 0.058 9.379 0.000 0.415 0.639 ## attitudes 0.872 0.104 8.387 0.000 0.675 1.077 ## norms 0.751 0.099 7.557 0.000 0.556 0.941 ## control 0.486 0.096 5.042 0.000 0.303 0.684 ## ## Defined Parameters: ## Estimate Std.Err z-value P(>|z|) ci.lower ci.upper ## ie_att 0.244 0.050 4.860 0.000 0.150 0.352 ## ie_norm 0.385 0.066 5.835 0.000 0.268 0.527 Yes, both indirect effects are significant according to the 95% bootstrapped CIs. Yes, both effects are completely moderated by behavioral intention. We can infer as much because the direct effects of attitudes and norms on condom use are both nonsignificant. Yes, these results comport with the TORA. Both effects are fully mediated, as the theory stipulates. In addition to evaluating the significance of the indirect and direct effects, we can also take a model-comparison perspective. We can use model comparisons to test if removing the direct effects of attitudes and norms on condom use significantly decreases model fit. In other words, are those paths needed to accurately represent the data, or are they “dead weight”. 6.4.6 Use a \\(\\Delta \\chi^2\\) test to evaluate the necessity of including the direct effects of attitudes and norms on condom use in the model. What is your conclusion? Click for explanation We only need to compare the fit of the model with the direct effects included to the fit of the model without the direct effects. We’ve already estimated both models, so we can simply submit the fitted lavaan objects to the anova() function. anova(fit, fit_tora) The \\(\\Delta \\chi^2\\) test is not significant. So, we have not lost a significant amount of fit by fixing the direct effects to zero. In other words, the complete mediation model explains the data just as well as the partial mediation model. So, we should probably prefer the more parsimonious model. 6.4.7 Use some statistical means of evaluating the most plausible way to include perceived behavioral control into the model. Choose between the following three options: control predicts behavior via a direct, un-mediated effect. control predicts behavior via an indirect effect that is completely mediated by intention. control predicts behavior via both an indirect effect through intention and a residual direct effect. Hint: There is more than one way to approach this problem. Approach 1: Testing Effects Click to show code One way to tackle this problem is to test the indirect, direct, and total effects. ## Allow for partial mediation: mod1 <- ' attitudes =~ attit_1 + attit_2 + attit_3 norms =~ norm_1 + norm_2 + norm_3 control =~ control_1 + control_2 + control_3 intent ~ attitudes + norms + a * control behavior ~ b * intent + c * control ie := a * b total := ie + c ' set.seed(235711) fit1 <- sem(mod1, data = condom, se = "bootstrap", bootstrap = 1000) summary(fit1, ci = TRUE) ## lavaan 0.6.16 ended normally after 33 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 28 ## ## Number of observations 250 ## ## Model Test User Model: ## ## Test statistic 47.389 ## Degrees of freedom 38 ## P-value (Chi-square) 0.141 ## ## Parameter Estimates: ## ## Standard errors Bootstrap ## Number of requested bootstrap draws 1000 ## Number of successful bootstrap draws 1000 ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) ci.lower ci.upper ## attitudes =~ ## attit_1 1.000 1.000 1.000 ## attit_2 1.034 0.060 17.222 0.000 0.925 1.167 ## attit_3 -1.021 0.064 -15.877 0.000 -1.158 -0.898 ## norms =~ ## norm_1 1.000 1.000 1.000 ## norm_2 0.985 0.071 13.803 0.000 0.848 1.133 ## norm_3 0.948 0.093 10.204 0.000 0.786 1.155 ## control =~ ## control_1 1.000 1.000 1.000 ## control_2 0.861 0.113 7.635 0.000 0.653 1.100 ## control_3 0.996 0.142 7.020 0.000 0.760 1.318 ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ci.lower ci.upper ## intent ~ ## attitudes 0.357 0.115 3.113 0.002 0.146 0.603 ## norms 0.646 0.095 6.794 0.000 0.473 0.859 ## control (a) 0.199 0.199 1.002 0.317 -0.188 0.633 ## behavior ~ ## intent (b) 0.551 0.074 7.487 0.000 0.391 0.683 ## control (c) 0.469 0.142 3.298 0.001 0.231 0.791 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ci.lower ci.upper ## attitudes ~~ ## norms 0.344 0.070 4.905 0.000 0.210 0.481 ## control 0.471 0.069 6.838 0.000 0.342 0.608 ## norms ~~ ## control 0.345 0.066 5.240 0.000 0.215 0.481 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ci.lower ci.upper ## .attit_1 0.429 0.050 8.628 0.000 0.329 0.524 ## .attit_2 0.325 0.045 7.230 0.000 0.233 0.408 ## .attit_3 0.347 0.049 7.011 0.000 0.248 0.455 ## .norm_1 0.490 0.060 8.172 0.000 0.373 0.612 ## .norm_2 0.525 0.076 6.869 0.000 0.385 0.684 ## .norm_3 0.599 0.070 8.529 0.000 0.447 0.729 ## .control_1 0.626 0.074 8.429 0.000 0.479 0.761 ## .control_2 0.875 0.092 9.522 0.000 0.689 1.049 ## .control_3 0.748 0.078 9.532 0.000 0.579 0.893 ## .intent 0.412 0.050 8.283 0.000 0.307 0.504 ## .behavior 0.541 0.055 9.873 0.000 0.423 0.639 ## attitudes 0.875 0.104 8.385 0.000 0.676 1.081 ## norms 0.757 0.099 7.616 0.000 0.560 0.949 ## control 0.484 0.095 5.092 0.000 0.306 0.683 ## ## Defined Parameters: ## Estimate Std.Err z-value P(>|z|) ci.lower ci.upper ## ie 0.110 0.105 1.048 0.295 -0.105 0.309 ## total 0.578 0.186 3.108 0.002 0.235 0.971 Click for explanation From the above results, we can see that the direct and total effects are both significant, but the indirect effect is not. Hence, it probably makes the most sense to include control via a direct (non-mediated) effect on behavior. Approach 2.1: Nested Model Comparison Click to show code We can also approach this problem from a model-comparison perspective. We can fit models that encode each pattern of constraints and check which one best represents the data. ## Force complete mediation: mod2 <- ' attitudes =~ attit_1 + attit_2 + attit_3 norms =~ norm_1 + norm_2 + norm_3 control =~ control_1 + control_2 + control_3 intent ~ attitudes + norms + control behavior ~ intent ' ## Force no mediation: mod3 <- ' attitudes =~ attit_1 + attit_2 + attit_3 norms =~ norm_1 + norm_2 + norm_3 control =~ control_1 + control_2 + control_3 intent ~ attitudes + norms behavior ~ intent + control ' ## Estimate the two restricted models: fit2 <- sem(mod2, data = condom) fit3 <- sem(mod3, data = condom) ## Check the results: summary(fit2) ## lavaan 0.6.16 ended normally after 33 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 27 ## ## Number of observations 250 ## ## Model Test User Model: ## ## Test statistic 62.797 ## Degrees of freedom 39 ## P-value (Chi-square) 0.009 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) ## attitudes =~ ## attit_1 1.000 ## attit_2 1.033 0.068 15.295 0.000 ## attit_3 -1.018 0.068 -15.087 0.000 ## norms =~ ## norm_1 1.000 ## norm_2 0.985 0.087 11.305 0.000 ## norm_3 0.947 0.087 10.845 0.000 ## control =~ ## control_1 1.000 ## control_2 0.864 0.126 6.855 0.000 ## control_3 0.958 0.129 7.417 0.000 ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## intent ~ ## attitudes 0.352 0.096 3.669 0.000 ## norms 0.644 0.088 7.347 0.000 ## control 0.207 0.163 1.268 0.205 ## behavior ~ ## intent 0.746 0.045 16.443 0.000 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ## attitudes ~~ ## norms 0.345 0.069 5.023 0.000 ## control 0.476 0.073 6.513 0.000 ## norms ~~ ## control 0.346 0.065 5.361 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .attit_1 0.427 0.051 8.295 0.000 ## .attit_2 0.325 0.046 7.101 0.000 ## .attit_3 0.349 0.047 7.477 0.000 ## .norm_1 0.490 0.064 7.702 0.000 ## .norm_2 0.524 0.065 8.025 0.000 ## .norm_3 0.600 0.069 8.652 0.000 ## .control_1 0.610 0.076 8.015 0.000 ## .control_2 0.861 0.090 9.580 0.000 ## .control_3 0.769 0.086 8.938 0.000 ## .intent 0.412 0.046 8.890 0.000 ## .behavior 0.603 0.054 11.180 0.000 ## attitudes 0.877 0.115 7.596 0.000 ## norms 0.757 0.112 6.733 0.000 ## control 0.500 0.098 5.076 0.000 summary(fit3) ## lavaan 0.6.16 ended normally after 31 iterations ## ## Estimator ML ## Optimization method NLMINB ## Number of model parameters 27 ## ## Number of observations 250 ## ## Model Test User Model: ## ## Test statistic 48.757 ## Degrees of freedom 39 ## P-value (Chi-square) 0.136 ## ## Parameter Estimates: ## ## Standard errors Standard ## Information Expected ## Information saturated (h1) model Structured ## ## Latent Variables: ## Estimate Std.Err z-value P(>|z|) ## attitudes =~ ## attit_1 1.000 ## attit_2 1.033 0.068 15.221 0.000 ## attit_3 -1.025 0.068 -15.097 0.000 ## norms =~ ## norm_1 1.000 ## norm_2 0.984 0.087 11.256 0.000 ## norm_3 0.955 0.088 10.881 0.000 ## control =~ ## control_1 1.000 ## control_2 0.859 0.127 6.789 0.000 ## control_3 0.997 0.131 7.609 0.000 ## ## Regressions: ## Estimate Std.Err z-value P(>|z|) ## intent ~ ## attitudes 0.447 0.063 7.100 0.000 ## norms 0.706 0.078 9.078 0.000 ## behavior ~ ## intent 0.563 0.063 8.923 0.000 ## control 0.454 0.119 3.805 0.000 ## ## Covariances: ## Estimate Std.Err z-value P(>|z|) ## attitudes ~~ ## norms 0.342 0.068 5.011 0.000 ## control 0.474 0.072 6.548 0.000 ## norms ~~ ## control 0.352 0.064 5.521 0.000 ## ## Variances: ## Estimate Std.Err z-value P(>|z|) ## .attit_1 0.432 0.052 8.381 0.000 ## .attit_2 0.330 0.046 7.220 0.000 ## .attit_3 0.344 0.046 7.439 0.000 ## .norm_1 0.496 0.063 7.820 0.000 ## .norm_2 0.533 0.065 8.152 0.000 ## .norm_3 0.595 0.069 8.643 0.000 ## .control_1 0.625 0.075 8.372 0.000 ## .control_2 0.876 0.090 9.757 0.000 ## .control_3 0.746 0.084 8.874 0.000 ## .intent 0.409 0.047 8.769 0.000 ## .behavior 0.542 0.052 10.423 0.000 ## attitudes 0.872 0.115 7.566 0.000 ## norms 0.751 0.112 6.709 0.000 ## control 0.485 0.096 5.059 0.000 ## Do either of the restricted models fit worse than the partial mediation model? anova(fit1, fit2) anova(fit1, fit3) Click for explanation The above \\(\\Delta \\chi^2\\) tests tell us that the full mediation model fits significantly worse than the partial mediation model. Hence, forcing full mediation by fixing the direct effect to zero is an unreasonable restraint. The total effect model, on the other hand, does not fit significantly worse than the partial mediation model. So, we can conclude that removing the indirect effect and modeling the influence of control on behavior as an un-mediated direct association represents the data just as well as a model that allows for both indirect and direct effects. Hence, we should prefer the more parsimonious total effects model. Approach 2.2: Non-Nested Model Comparison Click to show code We can also use information criteria to compare our models. The two most popular information criteria are the Akaike’s Information Criterion (AIC) and the Bayesian Information Criterion (BIC). ## Which model is the most parsimonious representation of the data? AIC(fit1, fit2, fit3) BIC(fit1, fit2, fit3) Click for explanation While the effect tests and the nested model comparisons both lead us to prefer the non-mediated model, we cannot directly say that the complete mediation model fits significantly worse than the non-mediated model. We have not directly compared those two models, and we cannot do so with the \\(\\Delta \\chi^2\\). We cannot do such a test because these two models are not nested: we must both add and remove a path to get from one model specification to the other. Also, both models have the same degrees of freedom, so we cannot define a sampling distribution against which we would compare the \\(\\Delta \\chi^2\\), anyway. We can use information criteria to get around this problem, though. Information criteria can be used to compare both nested and non-nested models. These criteria are designed to rank models by balancing their fit to the data and their complexity. When comparing models based on information criteria, a lower value indicates a better model in the sense of a better balance of fit and parsimony. The above results show that both the AIC and the BIC agree that the no-mediation model is the best. Conclusion Click for explanation So, in the end, regardless of how we approach the question, all of our results suggest modeling perceived behavioral control as a direct, non-mediated predictor of condom use. End of In-Class Exercises "],["multiple-group-models.html", "7 Multiple Group Models", " 7 Multiple Group Models This week, you will cover multiple group modeling and measurement invariance testing in the SEM/CFA context. Homework before the lecture Watch the Lecture Recording for this week. Complete the Reading for this week, and answer the associated reading questions. Homework before the practical Complete the At-Home Exercises. Practical content During the practical you will work on the In-Class Exercises. "],["lecture-6.html", "7.1 Lecture", " 7.1 Lecture In this lecture, we will explore how you can incorporate grouping factors into your CFA and SEM analyses. We’ll cover three general topics: The multiple group modeling framework Measurement invariance testing Using multiple group models to test for moderation 7.1.1 Recordings Once it’s ready, the lecture recording will be embedded below. 7.1.2 Slides You can download the lecture slides here "],["reading-6.html", "7.2 Reading", " 7.2 Reading Coming soon to a GitBook near you! "],["at-home-exercises-6.html", "7.3 At-Home Exercises", " 7.3 At-Home Exercises Coming soon to a GitBook near you! "],["in-class-exercises-6.html", "7.4 In-Class Exercises", " 7.4 In-Class Exercises Coming soon to a GitBook near you! "],["wrap-up.html", "8 Wrap-Up", " 8 Wrap-Up Information This is an open week that we’ll use to tie up any loose ends and wrap up the course content. "],["404.html", "Page not found", " Page not found The page you requested cannot be found (perhaps it was moved or renamed). You may want to try searching to find the page's new location, or use the table of contents to find the page you are looking for. "]] diff --git a/docs/willy_wallaby_from_wasatch_files/figure-html/unnamed-chunk-132-1.png b/docs/willy_wallaby_from_wasatch_files/figure-html/unnamed-chunk-132-1.png index 05df3ad73b2911481ecaeba57f9b37f850ceb2ad..119bb901005af6767e58870ee55c87b4f319d95e 100644 GIT binary patch literal 69730 zcmeEug;$i(*Df|8f+C=Vh)S1|(uzt-NDK{v(kb1jfFKBn64KH!3`pmI(vs3SgLHSt zz}@(}-?#3%|G-`ATX&YitXa&w=RIeiXFvPd&zX-(3R1*`RD^hVc*HW&PnGfTEYMpIifkI70rW{7RKBijDD7776az{7hR)1F;%?%z|qw7c&v zoj&_{_VdqwFTFe?{rAl2D?jk=t)2eH)4;pt^6v>d9)ZZeClqhb-}?82=+Rk{(&ek(r!sPAaL~og?cm_R&BaCav1$Z|n3&j|I})YluU_?_ zv`e|wGMDgLP??!?5P@!f?Xd#8J3IVxhY_fPPLB(HZELqgTt0vM_ANR(Ix%r|<=O+zzDsmQd;aBpqgrg}d;G*r>3QLS9! z$rF=~xGe4MwYBEvW@Z(?mMh;!y>QzDCW)SiJllhT+b{cbG+kX?`>xH_s1ffo(g-_QQb!xLMyjdb!+UvW<>$Fa+s&=5 z?#KJH5wg>ATxeBH{$FoyZtk7s5fhDs@NkET`j1SB9t8_e-Vb5W?z4PkesLTH5%84B?&AtzH)q;6)U>rn;Id+yZ9Hy$U%q@X z?uaY*Jalp$AJF%fKWigYO>pMt5&+=$g}wcK+*%z;dkmkPj7%3fBLl-TNlAXwKi>y+ z0|NsMU%mSJ^=o`wocTcR^6IKhDo)sW<=(w}a|q1n=x~u)inhF>VvYT*LTe-k{1vNL zPRr+rhzJPzhzN30(!Yy~+6_{a5|)?>HDuoc0*yv5FD*@ZqDrkX<|6L|jw$XjFfcGO z4i63SS&t5ObaZU}Y}@W1t@adVVUf}6E-;c)RO~clJK3L2luc7E)q?k>%g0h^tDq1~ z_4TRg=?90011N3vB&8c4lW2H+BeFavlT%W*t!ruyS1LaYRgxxKV=5q!WfT;|-c0!B zt{v@fvaql;GF?|i_Q}Y|z;jsCVvB(*moM{nTco@nu)M$(y-+dHvMz5L*w)rIH8oXK z#Ov(rY-3}?#g+HPys@znTj{KL8t`mfp+P~yt{bVOwPR!A4xchXa55`W)B({u(K@88GGjz)`a^ zY_ZW2FYFxIq8RtEEB}>L*6DDK1^Zhf*rp>yMMPS(9y2j9&GsmYTMgvKS5<%gsp8>jpD_~`jIRDYI_<^PQw8X8K=$XG0y znVC_D=5E-t#VaW(`O2i&(bmQ*DpFWv6E{v>;e~U%yUoMR9f|7!+`5cROHJKc9O#rT zBJ`VpAXbFw9m&rb$SFA+^-0r7p!U<4FQXm?)@Eh2d}n7zQZGaz=?0JSa|9yNqp5&# zEn*kqso>7LOK$^N7i&;F#%+I)d6gNEpxn3S435O*;t?nRYl70`)WPvAd94 zVh20=km?XL+IPuSS^GZSXRgQHV75(r{vIoEw zV7Gc+h!#rBVeWI)>!ZE(ifO=!7CItUq_+;}SAo@gtw25TSJ9SX+%lL)~&aBS?yptT$?m*<612X0j>Vz%lYRY|ZyFj9OuMhK7cm`qfnq^UCPsPDRn}@TxcHN>{ht^?Tizsz5sAg!A7@r)Ac> zzmt=j03$NHOg-J*6kK{a13XU6pXrUvYo1@elz6h3SCcAF$*J>aq{wx1=F4!GJJCG| zPcHOUceWam-*WK2)i7_}ts0f|=Uv50LpJg+UW8#Y9XBc$ub~r3;^O0{1<1EQ6cAm$ zd^8=U-`&%r#;PVEQQMEwrWUYf&-*(uac2zOE0FtsIcfp+pnyxzaZ`VP2~Y$w#^)LZ zuYW@@Dyygjm+ov2=r0y4)p+6H<`N+Jz(;78TJzEDuhxGgtH8+D7#EXQzh*6S&uZJ2 z&*U3_VmH-%wT_I>ZEFrzUrq&2AyEvo_8KNdCojH<&Ra*%V5TklLlR==(lxvmUr4sB zEo)=_<8}VC$10F)=X)|t+GC6!?(8A+s#%NeaR-jB<&d{kRl~9QGo6WCc|ApDD*OG0 zg;L_;HGCHRN5_Y&U0q!Q)}zh8e?Mr=XZ`kbQ8X||iTQHn1L}0ijkcyLf}_*Vvf*H6 zmaZ;-ks`T)It~0R+Pc#3fW&T!@`>W!*x_Lv z6{_HLc<)BbyLa#0R@Lq73b2{m!2RA^Bj^8Rxj_9lY1{(>N&UW^YhM+`vPutOZMPPd zMoK$}hBQ~}DJUpf7QeP9c3R5_eaMGW&iLaoy-nlUp4vPhs1K z2yxq9FlxMiPd061HqkpW3bwYSjbm5xvx>etf0D5p@39|~Z}Lklhmr+&5D>qTAv7%F z$yIrF$8F|-f%|hdOv_2M#1t4}=MejN=G$w{J&&QztGN^E<>+Dw($N~9rSsHxc`um4L zUAkgouDTUa?n2}CgQ>5I4h-w->r9FXykeu@6Ik~ioqaTBm|GODXBmsu zUagv!7j@iTP(u@*;8wkZtYlX?*4EcqN`~KGAe!oNTrZLqrQ?w?^@M;>UX_ z*cCt%X8zQzEeEtdDF8ZCp3(yCUiQ(Wvu|1c(RE`$XQ(vZEJmTy zfU@BC@83pNV|{&UHrcZ+!ieb1K6A9-2!Mgfo?nCu z!&a3a=`EI6jcmDLDCKMd*lE!jf8>;1ZQ21mrkxp2X?UM`T9s}LrEaMiY@m&yLQ6}_ z1h<7O;0J10Epzh)7^{E{mfhDB*c0T9Ll_KFovUu7IW2o>XPSnVw&~N&yW1tw($Zxf z`x`LuT%B^e`R=r}wKYR$L6^0^&SRbeX5IJ9)K+)gaR=Mmja;A4pFf|YS=bo*EMg#jB{^k=Ir<~QB?DlYc&p?2yUwf%SJ@l-&(c*F;v>;jVm zFvU>Q$08#y9}feBY^9>2!sc=v%~VPpy{=*O1)^V|!wGPUMVNw|e0pl?TdBLbfE)^n*|hr-kWCSIK&c011>i3G<%7=F zro^Zmt>OhBr4Tv|B_7K_rv`FS8u_*T{Z4~F3~Z{_B{dIUsyiJvWCN5zk$I7r>K+PP z<9!keQ2``E`r*h>b9`U{f3L>c2kFAmZ+me-ATCUd znVFfLeQ6SIaNM76OPNJ5O_`;=c^m7#%Ar&CXKRqP(o>~3OT`2#dH}N%NOym>M zD@#21n3yth{E-B7(s+O9L66#8>_bLjzni8@%ggu4jR2cEPCPj>0oZ}cG(xs|e}=m7 zsYXN-_v?qbYOpsxolxec2>56eZ?5gq zV15IZ3Z;hZ4>DChV}Gc+3A(<1of_5xbGz(In zKY0=>r>y)1Yuld=aXJdL`@N4(JGF8Inh%Qecv1Hf*|aZ+Jb>`s72Ug&ozw*-C9qqZ zSFz|QPF*T;@?ywTa96a#PR~~_W|10Bw?s@go3N(fVCOo+dRM3k3s-*nbQwyoyOndq z0k?gXSR=aMW64hQJ_Db!h6b}fA6B?g=>pfy2wAl!xep81wznhl8k?FRN+Y(8$~^)v z(fh8b|FFwkA`!v+R-3)WiuacF&HBW%{RN&*kym^A)z~?CiZ|>}`1$!E48x;t)c7M@ z0S6Fa*ai?ScwXKA!s>2Og+^~Hmx;RXG?VQy%EyHB@_$SE{{1@?^|;b8FF856!*&I+W&Gmc_6qLO(}~dsn;G{aF-WouA+Lec?tTs4jU{KjJ^Aiv&=wvYQUH z?z7z!qJIieLSwXb5J^OT3Z=o+0!A&@U6UdrvQBJNoeqHk_p?QF8wG~t_pf3qOP!Xr zTxDQaJ773L@W>hv#+|Z9%PHVjU-F#h{vQgB@Zs)0yEkzi&v11Glil zR=K+Q0nHcaE*S50jli9j>d5r@MZCXbb!)!{fSE<4#DlWbLY<{s=>#0gdJ*ntdnM^( z_sl!sD)!T~$oHE+{r&vrr|&B}Tf4>sIoA|>7fosi;hTO*i4|R4ED=B=w)b<7bG2$S z1j}h)V6Z{RSbf1wFv;WA9;=Im!^XzOWxCvckYh^5IidIpA6VbmfV#0lFMexl>k6$< zGje-%Y^;8b6#KgTV8K_yA{w@_%uI1zY24ZwTLJ=x7OFYkRt9omaaY@MAODx zCW(u+P!=YTd)&xR>tM{v|B;Gy#(cHN#OhXSPO7o_F}!AWz-M&*rlwxkty5kirb-Z;@gB9W6SpNIV2Ut(|YIn7#`=%Je{GkU1z%Y1~ z8}biYI|l;%YiAPza$NsILBZe~4-5f}Kg2?E$ru>X*#Hn|Nu*blm$!lxoG9iAml6>b zg>l@ubLR@37*rb^V~bE+3p_||vDQyg%F!;(g)O~NyR@|QpM0#Gq~zl6J~A@G#K3Sf z!Gzv7G_=!w70txu3LI`{1;ge|)DO@KAOJrr%F1-ATwde}IKn8JLTD`vi-2doHm@Z& zk>oMy0Dh5j#N-1>`19w_O}E^0U@`!99#kpA)_b#u9xbm%g5* zXM`*R1AqGBa4~Wv#35^3t;$n{U9YO_x&P;~va+~`Z)$%OUm&9RHuyd^HnvLKuY?`s zh}x*<7o54&)UfXc%!@%-?Ox=1IeM_YNKEYt@=y@!XlKMmjy!wj906YXP-UG82T+%{$^mbKj9|W(FFP?m@47Yj0?l5b0cduF+Z{W1_H3KW76_CJ!Pb0+ z^#RJe1qB6E+^++w?%laFajJ4%&Iufg5%WB_5z>lMVs89&Q^nHL!eSBRuo7#`aQ)5` z2??+LY04ftg50GY2K(6#+yXMMg%|Dml~EpqkKt2NCVyI$j~DiT7#$rw<*<#x)KT<* zyt9Fwqt)Yvw&%JgL90rCPGn+YQu=065c4GHj}@44MK z=6zzd$82{YBITs&HE<#~24lLGT-Grv3k!?7@(EyB)3dYc-JIJCeZuxLZ4(Va+d_Be z2=MXuHd?v3bjn^A zON%*hc?iIpAiD`lL?Gc_gVlA6TsD3)<>{1fz$8(XlZ$)ot!>VB)*NlLHdq6fUxpeY zxN||B54QUKE^)>S7x2QLNJ@UdK9T}-J(x=gZNa>BaoGbZRa;vNDljOw!cI#M&3l>W zPt1ZPc|fe#7a5lb001lpdLTcLL8lK+_3Vf!BXq0HmQ zzdqAH0i7c^6{?Slre=8$BB(-(&HGp}+41r52I@O4?iO3HW%B{FH@-mCCTpX*Y-q+C z16oufPy%4^W@cvP<>ioEL9GEg35@OrOjo{H#QbHv@8s|HSzIQ z5PQ(CSzgBNO@=zp{rR;v`{j#i@vvMJC#{If8dL#USy@`B61aJoPg-76rTlY*0d~5D zU+i?ajYrWM#z&0oizW|2rEd_os^0Tm z+u6B7Na!~4kqKV;&EAiP*K2#(@LyNze`>J(WKlr6AVLBRO>|RGmRA zQGiyc*>KVPTvrNG_;dD?vnO1ss;qvbPd;3p2P*2=1U7_heLX!av=Hrp%!5vaaAwDQ zc?;?{Bxy1<0xQmCDc@T=(NYxKpVGELt6^T?;k{jOhHVM2)c^Wr01cErS|Yq}QZ6no z8X6joQZM>$)L%IrCf@%jFTj8L-FSHH8Vf$XS>Dh2(8rkaV@+!NZ*ov&yNKOopA=sAhg>qd~clh|BZb#ffuMvis1 z_qTf(oK1(m1|4~t2<|8o)t0$xQA24t^K`t{VK zh>a-HWcg+OW6t9l^OBt0RQ>V@SGHz`=K6yR+3%Id+S{G^N3@0%%=nEOU-86Q#j#N( zlO{LDarS!r+ca-)4l-MM#Cyu2W~^6p)suUfck7qJqLu=&e0N`e5cJ_u8oGoIbA7ci zpCu6xp^MNmlr*<6(jLgltxA>F4d1yfpQc7YA5Jx0LA`jTi4J=_YqCUc}>CLM>0=t?zbb^7)%=`Tupp}$a?X?#82FPYv`OJ zt!>epoXFej_=cna<#^#^u{Qlwf(0Fo!D$Zn5Uk{RRmzKglp;09!)fMc3vUZ2J4qAS zYj;Kn<&U*jyYt}|YF58TD$rl2FP4}ojjuV$!a33`;M3nJxEraXsf>9ue)Uv8!F#(w zEiR^yzuqOaB5CK6ru@K27WKrR#=NilGL43kUzg_FoaCIl>r%d?)+N^G7D#STJbcbN zcC&m_-|3I{*O^=GHOu=b31(l?*cD4z*U!qBtK%;oQ@Z^d@m$xwd9lkZZ77Mj0~fpM z^VmqF`r|CssvWJw1kd=*AiNz5&@V8 zmExv{?}}?H(rLG>cQELDcN-II2Xjm2q*8{TRM4+upXpF5Z^~ng&4U^v3L{yhMP44p z%ej)&&~tZN?~0y8K$HSHTsIukQrLG?vPnEr6XCju+gv50XB3D%QpIIlX48?cal5nb zd;M)tu$<_u8?g@cWfGbd>kRR-M``Bj8JOmacb%WED4D6AP_BFP(8X$2E6IuK8krNH zxcv*a%+GHoH~dIft0wbZ@?%LQ66)lJ4&R--va%9`((chJMIf}p`P58bXBOQTaw#HB zj&?6(T#>IWtW_aYR$xk8(JVD-7c{MA^*;97A6c>P2|Nil{PAQ-!^3r5PF~NgX8Dkx zkedFfs&%->?cIQ-fPf`@`lm@1O04_B+pAl(4%EmAo-i7tkfOuSZ|U7)uWgVSCq%mn z?H7&f5pi*Npc@u#M0X-(^2VMw^3jyuweu&Pust$BpT}J;xRD!3LzS&7n~mDsTk)KB&bqevAVzlI^3 zKb+hVR}#;pzl7sI(q?f7&=GmK_BSF--D`fgb*%0Li#2iAZK{w6@XUwq?+TfO!_JCG z`TW$o<8x`W?)y72qYfKW>0fm%*?7IiV-_t^@0Aof&-Oe~Or){nF0z?B&e+lSoLQK- z5`?;rkO+_;k7czBC@9D;Fq^Tp?0&r;bxmjYS^#9GvQ?scaviklMnN$vrdrKJ8g%8-SBn6fq}rv6`oXa)F z%WnIPH)W(FGk!+?Z>#ekm_`T)i2(K38rDuMgOx=oaTd9q=%ky5dT4V_>-h+U(D#e;@O>A#b=?z!=00^(9R#H zy`sgiQy3 z+;|bu{m3D=Bp3b}6BMXEi%G5=a)#IoY7Yu-4XXU=ID3TKjKFO^mMI9jg>~U_;uL@h zO40k@6NK0H)3#U0+uJeUzV$Kw8@Nk1g9=g!l$xT*gD&az_!q+M5kI32zr*cG;G;Qr z)mC*D%GC^zfMt=4`2+dI1*l1-*<+tWH9FhCYv!B=j13N9Cr8k+QW1Tw9W3!1WZB4pNO{1$3QqDy04d3i<` zCxhs0(3NWVZ(RNukhu?A6g4O^+!nRTn;n?bPxJFzrlb&NuYtz9a#*_Qe(LM?ulWGac z^w%Ad;=f^*I{fm)zf^nmHHM%SAKB=0Ws8oPBL|(|Z%Y{=SMBeFFUl?LMvI@wXlk%WwyBBtHf0EP7i@qjLXi$)Ej) zXRxskV!0sVA0Df{Ak@s9o;*kgvY+CWFLsLd0Ek4E^-q4lIEA^ zH<5+3w)SalH${m-TH>)xO|p1mHr|;kzRf?)YHxDFcW68#Z&sX2eP|s1;@gAGwl;tm zUs41iGA%w*i{&^08nYE14y>*k%wCNr^W>NuK8Rfe zezBCZl=@(&WNbt%ZRonY#UY^>GEptPfgW+ znrwdhC=ZS8f=|>NqsoNRUUog|y6FWQ@ZK)!ky5?{& z-_xjHJOvCgp=O&mEggwkl(G`18nMk8Pq*_mpK^#;^%EyFZu@qXimC#7fFH?%+duAc zJ^9CEzX$5b)i5%21D<*w9fWI*wfD;paG^)hbM20QC}XS2fL!Bn2P4`ei-Z6ni56R< zbWct`o2*LU>obj!jCjqxTz)xNof16RQ7x=*wUeFFE`ISTtem`+*mivX>UHYArN}~o zt-(Jqak=$c2AGD*)*Z2tnz^F}m1D*2EcF99o*=xaw*xUXPUg*7hK0KuelX^~ILeTi zx58^D5~HUDmfh&Pyi>3z%A)io4dD+79q<};nNEG5+Pox9?^}s>CkmxVFQQANr29(! z%J5Ll{?A#p$Ci4Q39MYr44ILh9Cw$enC*do1@b-<3la0`@Nz>}7E8o_jv|ewz8OZI z)MZ|=?_jaerHa3<-lDIC;veJ~Z0c(I(XL};4L~1J+cMU0IycqxdL8Jk+AX{&Jex~4 zSyQsZ5guyQ^BF7e4t}K_M#&B@Sg)>3|DB%xYb2{;M64F>Q~^3v4K~H~Qi91}sPW%! z6c!dNPV_X?w>Z)+Mi%=QCKe_p7U~a8X4m@&4L{_6V`C^Q49meGsvTLRMt{Bg-PSGQ zxh2eybS!5HCqwbgB`TUFDt0b2-O6H?HwzCngufkIM`k}y3hY);*VL}7xpu06!Q=ml z>;FgA%IUT7h$x~c2h5zAdKNqJqt_J{owo>UYBq7BHM6px-Tk|?Nf;Q{3P`G{@`9b)+hErJl}D-|M21xGWBN*sv<|)D2hN47J4dLDb)k2h?|hCfp@)yuxGr(o*9TDf z3!k2Wc*I8(e7TRqgu>{S+(U%Sr322RN`*>cPp|3CwH#YlP#C3~755hU)z9saXn1-S zOCqyU@yrW=HM-&Ykpnj9EA_OQD;g~{jt_%A)#5Q_A&Z9IUte4k3sMll~B*{^iUw+n2tH*5=uG zqW?~uH7ke8B<%6sb2?|uM+Ou;KT}0fmU20bwYSBx2AltmV3_xtrz^%aFP-VNg+k(A zy%fToQSs_QW9QJSIO_N=7mG4w8hN^|LZt`myz0B`+fvayhf7N|Y8`Cy`dNG9?9)Xe zcclLsC~(!Vm)KCsMFp0Q<*q2f++;__#*Y*dCqPF?_>QwQQ7505^&$o(8@yqAG!8`= z5{0~37t|h%+uF$J-T;hoE1z#g1bUC!MfchD|0?rQQB>qmrL?Xp&M&oD)vHE3Xzy-~ zrSxYn=Pb9~xYp7HwNa{~7RWFpTpne@Zl$D#zx*zsM75o_7Bq^xlI-3jUN^R12;ZA$ zqW9gKA`6St!F0j6H1KuLibJi!=65}%_eGeFi5>2w?DN*w z+;INe`WJcy{v>oEfq^ZZox2+wU-BZxvca-m`&^U3VtOF~A$qRyQ-Z*EM|k4-#!dRC!g)?kh=cemUx>bZ z-T1nMT0Ws0oVZ0rm94(Auux&zl?;7eP0@`` z51UbDPTgzA$A+tVT}eKBE5hr;7Rg|1yFvuC*0_gQf7f!YP$J;i3kPF?3Gz-qiq`q8 zlg`^$c4^^K_j@fQzjAwj%TC`+9^({~G@sS6clG_{t)k3PX7r1>G4Ye*@?LvS+(GVg z#g4eiNg3#i$CGQ|FyrzFl$6ILM;a&D4b&Zjv z9p4XRjC|3l!f4j%=!VBT;FL#IhRV#JY6s^*|HRb zkRE?m=UC$~Cm3Vvpf;Etw96%%Le1=(j8;D^)kb458I>&SQK3G?b*=LH$7{9z(2QEa zz5ezycD?8uma1*yg`%)y_3^=?Aar)3c``CHrEPk^-3e|k>u+cxdOJHiu;*c+otK)L z%1Ti9p{0X@&I<%BAOVpaV9b?($Tr4Dl~mk?f+F*ms#Zk!*0AOfF%b=;uWiAb1?{{% zQi+s4X+`c>xpM;-yK5^J7ae|k!)rRckzG#|z9D6*%e2TH%%~;Cp@yq|x?y{4%zJ&h;q*FhIB>(Ccz7D= z=n_$F{lXOp(^;J8>ONVgewp{<7HnE=jmpU_cZa3M5$g71x9Nl{E?Qa-qudm1zTQ-n zi`Medw7pb)9qh%>8f|H5xh4Tkc{Fqm;XsM?pYM;M{jQ-g2<~pM^>)S!3qWfTyj^vT zGgn0ui6{b6?{%SKU3N49rT(5Udt!Q;##fJfkrSsj}5VNIB(AnkCZy!-IWyJwq6Rw&_!M?fy_qc zO~*n9T60B+BNU>ifJ69YL8zba(2T{x;(m}6?>H~UpU&$BIIMi3?a>5gv$@q(t}!9^ zoh5M09z$2#q%#2n2fWzwz8NH33?jQq-;W|gAPlEvPYXC-BocDQ1J?pjW!JZQYKO{; z0Hg8P_1{%Ia+4^#N9dPa=xpNp8|uTl0jOC~Y#s&JuRyd;k@M3JVNGYMN&$X37h1Fs zsPEH@+WJKFhW4*!<6?PMS6xve+K%fh-3=LXQh6cM3aP~si;Gj0Hb2E|n*0?@N;fRi z$cMRXJXn0Cxt@F1?tYX<^UEm83xHlVRJo~gDFLKC98kpNoRQ_lB7s(--c8*-=kD4T zPBy9Y!G9ulw-*m}CT9;;SAoFV^H1j>Ovxn-s~oy(ji5k~l8jx>5Fb;(d@AWoy^Q`w}9Dp~|V~ zi=EO${DiTcLLQ8Cn{BcNX(LqeWV#9JhVSP_P2DZry(zt24O!k4J1%RMW{}>Iinc%b z_0aduT&=`|&FEN~*z6lTUokH>Cw=wtVk7qlO6o2GKQM-6HX zF*e7m>MtbTZPowS2(cYx7)HRaI6tkaQv~O7GMdnhytXvNpTr zxIA};Ou&R!;eFy2FJwIT;ogfZ)$TGkBL>I4JSwS>0JOUwMDE_Cceu9ZM>XWrQtOG$ z&W0eaz@I-^gP?5~{?1C{TV4LEZaag{glD1RdUdVEB_%_29^X;ABbFK850*Y+@3t@U z-?8<@lA3~^jicEwB%O$w)5n_OLT;|=g!Gm|F=u3BayF zWdYUBd5wFsR}6N2@rEUz=<5pEl!}-)+s`+mp{b|96d&oOR0djOIaz1B4bMr0F{3Em zg7xf*;bO9SraM(tlJ=82U0JF^bb--zo^HY5G4$vcamc7 zuPXG%KZ^P(qL$D*E%uY)QOhjx_U7VlZszEDG4*IRjydaws8pJav=%L&Tbuf))(Pkf ze+&&hb#sE>{yHI%*KpFxNvAA`6`xd*YRJf<@2afrE?eO6OAzzaN@3Z|`vQJ^aMYO> z=MzY@8Mafs!Bv>{N`s4eojj70+Q15^CnHxs4V&$p*K2BQNw-=q0o^*j-%u-;d-u`kJEMWf zhM7k5Q{jeUCv)UU%wNpIN!s(-Z0b=u&nsQnj|WjX+F^oW56T}XJNt0Ucz$N(Xv4W% zXGNri?ex&4${{PlPKaHF-9|KrWNN;tUej%IBoQV-%ae<9L<~@ULjIZFYUQdqJ(D8p zwsqsd#5fNL7)|H|?OubOv1qu;brXCw9B@=I*#JQ%E-nr|_o}KY@Sohwasqp-*{^|t zflpF!EcMh=x^1^K2!=Rt1Xq+tY{+R}NXrQF19cOUih+JDe$4h_Q0W*2+zos6BsboGEkNGzKSEXQDVADXG%B2J!=f-?>~O=^L+EH<%qy zU5tQOhVEow0J5pUA@Ii+7m>o`q3?IGWb^aMHu?`-e&n04Rr$ErB z&YCx!e3-Ev_ok_lfZo~|o74QbUE!GBnT4;2UM!Mjd(QLVgI$^J$@{IIC(HU4Pc(GIaok~G0qc=W?(HFK<0Fr9&aX|=u|k!;6JdNk`D6H`~EyvpORXDkeWK* z@EHpqEp3BI;>S0W_3o=5O%lmhR@AP66XX%Gli;%5Qzu{2+NQ4Af#Vu2mcYcD<|||N zAc~;>sQ`qGYiBsmum;QG(_WDu0#gCM!ealXD+(SF5Y< zq~KkAC|*T&Gm+B9@T^VR4?+jxDu5%s4$)wQ%FpoM{r;LUr z)UT)0(UP2^qMl&OlL!!UTsVUB|0?Msam%()?n>Zuapi&R2?mnIq@*OUm6J<&fG=`> zetssX+zt$V`sH>Eyu9Fq7xx6QR4SM%O*!infA_}2?w;H*dIMHcaHruWg1CH1IXO8u zXKqp#PU+8}N_}M+ur5zTb~NTA;8R821H;4pbyr{;-Fj2;d;G(ND{QbfC9*{R6Iayk z4_QD%kxGq|mz#I`ao~CjFj-EHqBYYOVLp+uybF~{g)AoWmg8Cnd}iZN|53zhfaV-C zsA`hf(OtVThhgY*g;KHF$Z$FFRB_fUSJu!D>O($rnXlOgKhIHY7B8zV>qB51VKIOOeU3S7 zzAoRreu!IpKfAIpId?bMF)^jhbvbkZ8VYjRt~{GFr=8mAu3sr^c7NCvzG%$F_pgTt~`bNpMDIRS!MloJgLO2(!JQos_N$JC<*RwoIF7+w3hC=F; zLw;=O@e||bR@S=l(Dr-o8rV_6TPP6#F4s9i$4w+0LdXN?ohB*VI3+}ItiM7Hg`lRS zq+CD2l-o0b_rAI;e>jHEqQtm82GTpi;78y5uXF1i!V>sbY1kt}}t{o8>)Q^eaKzPOYWT&{c5%?8HB9 zxr#=QS#R%vhhRJI7WRF&DPO*8KGb8Dgk#~THVTPP_*5@kQ^ulFjZl%0m%xcLdXminG#Zt9oh!9lu(qW;bt(){FyA~(7X{%fhZn|;9)Z&$aGGgoP<3|nhT2l?1$x|D`{SyI-qmfDs~EP8 zO~0>ShbAFRe_XB@UzFs(?{uikEn-?0{xeZQEPmtmGcSSS91z^R7dCqer4r=1o_DHD zbCi|6{MB$OZ~u~O_S*f|m|#aJUQLuc97g-{k&5ah?xAyWbK07v+=Qn#sZ0AL z!<3bZ`0I^{Hd}dP-fqnfpqybGEReMbOh*N(mwC1{TE8htfoWBtlh4gsuAU+Su?-b) zl#XV1J^G^-vihr|REz+S-{O+RVM*)YsRwN~!Y78q4WDX14UP_u`}u~{f&D51;+fmq(Wa)R%fI^X6Vd&as;G3h zO3fcov~eYRs7mnE_3GjPT|SaV+VDuI8kviaCyVJwYzYE>2fwiBgC=@S_f|w84H`^1 zWIl@Q=n98LG^+cXQ+bOHI%5&cEYZUr3t;j<8XfOgi(x2A^vl+|YdzGCahG;?n~qn1 zogonpZAbI(V#Wr-&trcL8$j1e<>Z^&-A&07?!!l;XE5m@dbCjvIuoYLF1+JlmAQXM z_|Al6J;+l|IPWElt4{dz@yUAU&mXJI0q@_l2mTGYnV$YpOB1o>KRp0Iu9pr&gr)y{=CuMM67Q>xsy`{ncoUXNg_Eg{S_e{f4{n_%d=@-9B&n zRoFjD?ibn17)RULi|$Wy@x6DBqbcOliduCCi}5ZPYI z>^xMQ7kO459PPw_df^usAkB7&Xg%PJ9k9Tzv{sUs&~~u?ZeeS+FMeaSI*RT54aPg> zZnlEPIc%llYDtKdkB-l$OO?oeNCKcbIR@k z=1^@z-Gr!iLj;-EOR{yd?ECI}_=vmcJ8Z0M9yFex6z9i>>gxJaP_MKnf4v;| ztG|LX+T3q{Blza`v)>3RHgA)Y@#?z6;__;}J#lB8$z*hJ`GDPsaH(^@P*4CWVmUo2 z_So@2zh?GCFP+O`@=qh_L%w8Jp`&?%$7EhIC)2k5<4fI^OemG_axcrI_C%LgUr?P# z3=E++XU1c9b+DU@4fE>5f_(LZWcLjlcRfY;vIZb8;J(+>(>q-vFly_$jQfG* z_32ZuqxGh%cO{v7fbbpet|lfVSeuwQA0N5aq1||k0|ElzXpq?6U!Rt0NHO32T01&+ zfK6|0iNKetlv|v^>-h#rO&!iZ{ANmg9>bfpG88G%PV4f|fqn$1WZ2o*7FQ})`9Hzo zDPlS?9l=}hB?52~70L_7JD_GgoIf?3%kY>uPI1&F&o@j==#w_?{D6 zyX;}H|3%q*$8*{L?clJQg&s8M6xp~$tq-KMUzM_yT~XZSy{=}B0GDJ z60&8l_#LOW&;7mc$M1gp#`~}9(tDie>wLYQ<9Hs=$E4ulJX zA|uBJ2V0+stbHZfR`ZRnva%0tF5lF*&6A}*SOB~dr1{h4k$3%`K5a73g7}lY3dk0C z#YJ-z&kZ#y$)smwpwZQQh+sYvv24TUc{HJO`3z7op1Azq|N8y=cQln;kRe(+G&Qde z1zbMJs~^x?TBQ%JJnA?*f|rWX&v@;}r-zRpKSr?Goo3`|W^vCLsk`%BNT!f&^V&Vo zo>Hb;ga{B?$L8nfp$qV%dv(PTZN8zQTIg01v-^G7%F3|!2$j_G#G4B-3|d@a^i)&{ zZ~lF@->kcAyZ; zvt2?p-fJ7Re~b2iT}c(wJfxATU!2T172nd<_V&Z>Fk$v$774E4;8(TJ@)(HUGVO2J zeEY!KD=>5;MSV&{=1uNUUIm-<&0E8Ww|K~XD+ zy=7d=n@`Pk3oaX4lIjcY>vd0d);D7CEh+t2+TAKu%zjyBiTbIod|>nUJ(VZ;dzXW! zZZfr}|(Ayl{M`eeg%Uvd#-?Zr|Y4K&zQQ0h0kQ>Ru=& z?DJ2mOUm&aK$%%(w9ez}7E|6ENjC^vCST4~Zlh=m+bJs(-&|lj6mfH~y>S2J!|eNx zy?l;bs|^iq7iF*67d394kE?GXi-M|Gf4CLWo>VmO@o;hNp_Z2~MUytp9B}8Zii(PT zvafPp`Qh}w<42BG)VUh&y*E4FijR|fxBsqcrfS2Je5~OCBEtF}u!k(Sd!&8&f`==p zOjryUvZuAHk5TI|7cHkaI6lmMc>Dr8f9QG3i1%H2S>=~w)g#NrvW_^_%e%Y6@vLyp z8&#$nz3<5QdTRJ`2y3nCkXA)dNJxry8Y?HKDY_mZOQh&b=l}EPPn-=N!$NU3wqL+8 z__(`W*}0)Gy^ge^jd4NB1C7fvfsC()6 zMhDwvzciLWwWT}R^wxxeMdh>WkMI+orQ+aB1hC8Yg^yNdJQY}dRx_&9k&%-N!-%{PU zQ_0x-_(j)G zeO%HyKO!sQPagg37HyiG^6pmmi<5EVYDJ*Cj=r`x@BPkEDRJYPr#~0-faaeZ4aL4O zr9Ah7d5gB-OvQw0+sVB4@v#k(n}R$#Wmt9P#XJ7|P%aBEpXrN=_&d^?QllAlG0Mv1 zGs}VFQQa>P&{dDMJ;*X-b34d7PIL{ZXD&9eB{|JCM$HEn0t*LrRYRm2Z)X z9M+xT?J4UB^i1D%vh{>p)Z*XFXJ+M5vOBv}T;1(|4UsRi%>P)X%Gg!SRY&u;ud%zk z`-G{H*CnqjY(+eVW-XD9N3E`md`ccfC1Lr%!gZ(R$%HExO&p{2SLOb4NoEUfF^QCz zNU94czWkSEcqwZzd-iG8On!>!-g_K36ua11+v~D5wolBgI;f?*-al6$b?Nr(jUPJH zIhwbWzvjJkFc0}Y-kKq=bF%1KS-{X>XS7O`B)ZaQ{9n9y;kINx|MvSHPQQ;VC$El) z5-Z3yqd4><=`(9xZLE>Z@HXQKO>J3swxz{YS)Utwb^2nn_jEZQ@7G~}C~C^&oH;~2 zMEkZwRwXB3GP`?a_G=E`+ZU4#pV}*ah8`I#h%m})FR&V~ee0PntAlNFcuw-U`m<}| zMQ62yCPV!dUuD!0^!2%^vF6T}er~Irbu{)i!$Mq_)n7Zl-cwX#cv7mjcxUV;cN(RiA^Yp{ zVAKD#_}6AQFF)N?`BNA@`SpgWLmGc_B94lSnJ;mkadxZ|wV(O@Hff(%nb-3VMn4-T z6KC>scVuT3f&ic?vuAQo`Lz3SK8x?!wtn3bKCHCLJ0f2%pWwRO@XAXT|C`gEn|NEK z;e3_;AGAtxlh1XzSd9wGKiN;v6sQ%>onsVw^Qh?f$Dj0bY4t)scKicpeW^j@NH-^W z`KaU9`zur_;)QB5cM2z_(!Gsmn?zg=_Z#vqPxj^T$-eeI6TV8hXxJMHR-YD=Av-Iw z+$rbk@ZCbEwih>Zr4&R=%;E_3El$$yD)^qKVfDFl#OEuW)|b@X)GA^%_M4hIYx{Q@ zFME%bE2Z)Mo$GK<4KdWWzuK7hw_8}x7z-fo+C%Zw^ET6C&%Y^ z+$r6qtlYj;Lsv&XZ#YGf-1Cz2igKvI3rS~}n<`s%Es~177Ay2MjUT7W2~m-Y-`GF$ zyDlxyWUe=-H1&7Nje>VG@CD~C#iyvv*(*}c&H7aM@X;8GKC|cNI3y?I%II3?wfO04 z$=BP}_gc2l1k)fRQQ7K#Ra@X6R-|?rxiypNb779xBG0E@V$7Zo{20Z_$5btv)X^GQ zd-Ku30w1po-zo1x?mH(|&nP%4xSr4dqp{1FX%BUB@a*TiEY59C3?(;{ZZK0i*$sS+ zfRTtedcsB4@oQ^I&LW?Hr0Mbqa|Wf9Gip_X$4n-z0#jFa6wu+&ctI+7#p?Si*WB$@ z&iNQ;4?6#)w(hwVk2y-EKK+||1JlpytJ6AtJPr5xH$N?y{W~$97}fZ#=B8CiR|K7- zw^vkPw~FIxLwRBFk24>NKL1^*UYWBhDAF@DSh7D**8cI*q>Ri?sx+mTq6n^YOO{7p zgponnm6sK=m#9B&r=m(V?nfKrjB|k(ng`z`Qqe{`S}hD7UToPwX3cH0JbP(wP^7~< z-*k55RZ*dHVpY+CV!6^2mbv8fIR|g-4&CdR^81C?v_*YQOe7|_@iH8j0ZG*MGCL)p5QU!C9U3K1S z#-q~L%VwDS(bS>Qk1%xk#B!;{qJ_p{n(vPN!j+EC;v?d3y%JQ_E!9voR*+TjeK|Jx zEmR$;`2o_md>4V7FqFSo^=2AUHau?PQR1i45gmS-4WAxHruIt zLU==$b*~#V^Rme2!7R}{x#g&;%$FH56&7GQmcq?r`5CuAZHBNMWom1CXoq9>QOVW* zdh^Vi5p7KZABR$v8L~6gxEu{So;x0``3w@)pEl z=80%lI1k=S-*|>z*rrZ&3wjjj)tTSW`dj-=3{|yTjN3|nH;I1fR&b2lx6vqqIo#It zsTX7A+oQ#a^Br7V+(G@@VD3s11)oSX4W`BIn&=B|9-nB}`P=%c=veo8`#863+VjdD zMN`#c5K@J4%G^A4_%D#(zZyt)SjhT7nb&vm#Ky(f;*LG6|UQzVj+C{uw9*~oA{eY zDVLpV;|-Pl%APDtc7!5C2)$~Up7mZhlVvaUnZ%sz^Mx}X$5|TY`{}&-ILYbR>B~QQ z9N+Y4X;*hxp>Pq;YJ!_eRoe5tgG{aQ_EFnXy}X*J6~i+R9H|dwr{Rd;77uk09{tov zVX~u|u}o4;_kwQ45mNa<_KAu8&^z-%2!YjUKzGh#fs4u)!@tUK!y{!%90v9B#kc&lyArj#m3ndk4f@46#C{io1CBRf?z$o_Y7pw%b>D`Sxp2e`wVW|K3NGKNX&?OPlDKLGs zyqq$ZrlcI~L1wE3^uju0e^X8K;MpesDXr?}J-^P4{qXSYJ~?w9i7En`2JVEF>Sm+2 z>35%dRSroeR+9&~++NAq*9`g~%azq_JHSc#U;0gW5!oXmVYbOX+hTtz z%{YnJ!}Z;_xtuTgDYLsec~)P&M(d@v=l}?E~Zt}ynDqWmj%+;bVbOI7;O~-D!r%v)LKJM5Hkfn*i;`1Qu z&uIJozXh$MYl^#)ou%9-oTyvv%|iF|6sZ&xvbWbsDp`yar@GpSR3X?d-(}rs*yiq3 zWb-{nqbj0kAAA6s7JxxugQ3qP==z8TW5vn1)CQ5?jtNBqZmm)(s~1l))19d1C-%GM zQc8HxL;G zX>P8O>PYL~+Uvd4W>|F432}E>exOK9`szg;odibC^j|fMmN$R!yI-^SrHa?ttEYY*5@Si2w7!jm19**6xzexH z1HDJ63)Hx}SO5MU8ZY`&Dc9F?JheUO{y(Ar^c7pXw|{sxKqbEFvNB%or|h_zICF0O^#6)J(<7st@~5Q zXZT;FK257>EAT|D0&hnD=}oSXcVzrq_FM(XSNHkbVjP0~u!QI%$Yi)_b72y25F!|F zucC~i9~bW3mE4ppq`9d0x6Vk3!v5jszmM|TbK_}aBfp9L=^7t54!PE*+Zr-hA@j?} z))2T}bD@=9Z>x6ewb1LUHSbsHS%g-w^yt9)Y4fvCaYZr5WyHK!v|xBzv|HxuK>v+i z@uRM7pJFIxf)^opP^s<6 zBLMErcg!P7WeCyF?^x>NI?LebJJrkxI4I^>RmWfk|TI zluM1_{ybHgT_#K^g`BzYa70?rVwN9nl5LVFxm5C4Bt=8-l9Ov+c`vu`VLu8+x<+0u zXRAyv9zS4qcbrDvX!kM?a?H7A54fLO`Xe*;*w^GZLlXt3-VMFuyShy0jpUCaa$udj z{kvD;p+eM)zcMnjdR-X$yt$=E+xB{Dhj*=DX2)L6nHML#V2pCux*l=5lgG@UfU zv;1y-Ycd+h7pixj_Z%&?h*&oN+`Gkm_eCJ`OfGI+MYO?c3F0oPS#?c^9pI#r#M4Hu zjNgjs8l+1x_{Ll1uojT-_ngu-N97>*_RK2?IJgYVzSp$Bal}p5kv)gARne%iE2kjQ zH}Q|sfOmDDLB(iK$H(>+=g-Rz@*jLkKYhR=Trzn{-HF95)<7q5935#dXlR_8CxaxW z?seWP7%2$)s2}Q}+;Q@POvs$R zrRV=&%0}ed9e%wY9OmICpgZW09Rz`V^{sLz|XyL#V8U`}-ec)d;Jk z+qVzG!cf+(f_PvTNKQ?dK2j30-RJwu`?sAGG-=&N@Y9#?05}lLt>Sm2k>2#iX{&+E z`ABg>ZjftGU*^Wm)$}|gYwLHKi8TIbKRI=OEBoQYBI%0SPr3?ZKWtwcHbfaOP%MQ3 z9$&wHWu5$X$Pc43P-iHgZ-k~HhLXt0%ln))GaIz6k!YSjeT80IcI{I`Z?;cBEE%$z zd;I_OmO{_$7(f5JMMg;TL#R+WQd~!-5Th}`*b{ksPJ743wP`+YVmOKkl=b8@L6h^N zf@md3W`szX;@5|lor1{Re|hB$6RYgeFNd7fZZN>hbS)ljGkNsjf%c_KK4-69edSH@ zadLjiCp7e$iIQ&ET2M;naa~r~159~3e~@kRw)QF&Azf)P8MHAhUkm zlt}Z)w`@5=4td!`x9jK5J+axndw2BD)9=o%rM_fjT(~nxJl(Rh+vg8SCkBUrL|zdQ zN2uzvUbV0g!?>14_D`R7lOFsQzzdL1Z>8sj{A5br7zCXH4N_ADOb}KgUn<`<0k)O> zty^ZU$?tYfr>#|bX)=e-qdTYzHEGDZ5fcaem?d%{3JK=k%-mddMg~Ttz=MjrEoKU2 zK!gq2bprYyO=mZ5d~I)0Rvth>_ob$$qmfuDX~CBQ#bPH;EQ1n=odQl}#QAs#v??Vg zB+f(9S~JyPw7sq z+eIzbi`b4Il_(?5A^r1-~zBiz%MopG}u;ke!Ppd;{j1Y_i(;g^g!~^p@gKgc{GK|x+#5U?Af_{b_Nqh}?v+H=33cdd>dwkdeS3TY&YlCpAgvv6h`M3NKp zcJt7z-`o8+J139@Yt-dMMCb$dTQz7R9$s z@8#fOp!L9=6)T5?hSJf}ULWHkD;Dzh@u||kM~_Ve)@HZsl&_bJjLePh!p_dl0u(gi z&Y(O9RlaeEDSHJ42TT0L*h#eyfZt!eiCi94kZ8S%dAkd8}{cU8=Dk0FnfI@$BhepQ@uJK}S za`c`E?sORM(wEt@XU~eOMY01DMif3@NV%`PRtQW=3UjRx72;dV5;sw;Jwu*|tgI~P z_U0s#kp-q@WeGcOzaGt@Zencw*bDHb&5}O#?%n^e0&@_IgmU?1U%9}L5Zb+a>)!}m zd3<~4+AOpo6=e@cdwUw{0s|}bO6}=Ui4Zjwug5rz=4LG!=$dvtd(W$Aa1qE*B&!0y_$e&;7=>pV0o^z|C5_#KMQSOmmzUf*ijI1XF6a}jN z{KX|Daxw;3D(DywTU|4zVPN25MS(Nt%5^n0HKxr@&jRv*`2`@oOwZ1Cft=xmLJ2DmPo+A|=Kzw~iH-zf zfv!6UE3+bSm-)vvo zi)3UU9wtyt?BM3+=Huh*^(Q7|F`i8KxALJq9pfJk{#yrC0{M?44lRMe?Oo`7I3CLoX=8LIQa`UfT$s5fo%kt3jC z^bq>xhJMdBtMA7&2WuOf-0h7*bcC`grh6=dfc-pgH^s;Gry1{j^N`4$<$1C3d|UJ^ z^&eAAHYRMB6XN--FB@S~7 z;-PSJaXl%;Xc}ch9g5xi*v8Q7!nzhFko7!5o(3UVSl{YjvCp*B)f@0@40h!hkCW9_ z-GN5{uChFl2d^ykm3HtF@=pmXNc_${!l6 z=v;V>!Bu@AYw!anJD?1HmK~y%)ilP9B>lrs>Hw8*iwefo5IFKvCpz58D~ z_3P^UCXN66Ikvt_hYuEO3FAI6mPi~C5UA4n?=F8NCJBH~j?klHSy2*9F<^iOUmv4c zRq@=p8<>k#rerP(V>)sq&!Ma~M&Rm$9HK1+rAw*}ghai@R7|1`&?JV) zwFIJVk(`~Fakb6Z*lAPaUkcg&-#+r>jli6>3gNZm8s_tKMD6dSq|^$=EK zT!vBD=U%xWNe9+FbSCSTv+1Ja-aMl901=GUFJtE0J$GRbj!!CACPJ!~Xy*V3Loom9 z>eOsQjh?fz{>tuVJFYoa&`e@`6(5ZoBXh?*4tF$k_~t)}iZZmNnqYlZYYOt(ao2h4 zzFn>;+*IlN$No6B7MXvbe(!Jb4iF{uziqXyJYSrqK1jsVm~J?w)^P%f)f%;ikn{Toz*R8&+}L9~(~)UY9+4pCke1qFyg@AiiR zeJN6a(o%$8NFj7&$}1}H9Tw#QR|eN@R74mV3k!?kg$oB+F`n?2qaz*erWqCstlRed z%vJL@QEd-g=Y3b=|mPOn_?9=~S)KDA1={6$spo*M0 zf5nIc)*}Z)Rn^P;)i*Aq<>85msF>K&;-dFAWGe?QvgX}G&uP2*(+CGE;7_ku6y1c!W-@sdu)KjGwURTwCQMQ^xY3}Ct^QR zFCbv06vUt!0YO1oHh%u^o4>vyrF&%3_?7(d@OD((WZ5`4IK;)55xlCGkW(@9u!1_; zQs9#9g?Jjt7B}BRPtWkzdYQW;#C&{wUdPoD^(8X0l#6?NWPfEf8%#}2?e)*h%*09N zx=)PW(oC$Xs5k)I9UmW`ojnI`rp&c3Ueplrk9)iS=-uGp>E8$c`>~A{C|;VMXFOqj z8FM!@B2X7Xxj;@%t~y+#2E$SI#lA9LKeGIBlEvI05up9=i97}> zoy6cQxx%B-o97S7SLzeagsg)>o%bbq7z4>u6>A%f7yEb_q^dv%9Rm7gWwnEL0J3*b zQ4uk7dLv`>)4Ed2af8P`fj>T1k=)a>OEDW=otzesfTMcl3ce{9pOdq5b(GX9f~xea ztechW-veZ2zI7F>{-w}$P*-eeZFd-Lwy1pa*mJi-H)j_5{ySH0yB611vK{@#V>+Qi%hEgBP3P|3JR=*lHn9n^CuO9FM4dfLbE1`&54tS4oog`P=Zt4;;wvx zVuiZiwl%nd%;U7O2O=u1UE<>6=22$PFj(!?&(mZ`An@9Uy?w=Ej+mh#Nm#w|Q*hxL zdfT>9(i}dlpy2<4(KZx4D`aHXV>ycb_8&WaQaj6B0}snLPD5QiAA=rQe~yn!A(OM8 zdHFEoeF(F}>+}Fg2jU~TqQVck9+TSb3)aHXv zq^W7nyLo-9O3SE^Z(7BW9CM*C%n`JM_&AzTM={tZ zI5E2nWXhDuT~D7r1(IZSZN_vxGW%y}BbY`Np+_%iqRzAYEGnu9V$Zr9d>DVXNQC;( z#83D+WNJn-l6bqTnOVY1YKiX+xpvf8fb$_8dVi*cNY?%`hxx{3j}4o!Y3jdGa1W2( z6+PlMGCLmn;lyKaf&cR!SFkwuo!TLCLFU0oH=q1dlJ&{d6Y`W7EGD@qX%BfpX;LJF zI2B7l%s^!Q`QyhiOvCB*Upye`I*-|44Z+*KlPm~4B=UNq=T?^OpMRvcPdxHpv~_Uc zC_W=>VKLaHfAuOUQ;6=nx?LiSoMazOe+(6FlES=MV1$poG+o>#F*J=BcC=N$1Y@#f zZ+UE3+Xd1TT8M-rwH(e=M z@X?UqGAt}63y9kNR*mu5ybk(A8(*aFQ6a%%K(sie+S=L}iWnFflkD=9 z6crKns-n<)C~-3>%wBtY6o!ltg?^`B{_;pO z9Rg{kIF4etUQEwbk}2_HDMF}STwFY43;8+>jupXF2kHehCcq$dzY*BrKVf~#F6cE| zS&fL}5n`WP*Mrjuk&O$xa03^YDv|7SlOyU=yLGVmSyqv@{JVNdJBn-wFH9s-~m!9H_<;3N)48-a80-7Z>kl zbyC8$_O4T>PL+Q|$=^JRe(&BGM$8R+^I^B@7SfaT!Lg;H7x)64O9lge%~}*vaxwS< zGwL3D!Rcdaf_mt-{SQ2S?%#hQp+w4^jy?rl>G$t9BRetw%Ar}`k=Q}0&05bh5E^2p zpfUM83?D|I2o-8dCkBso#-^)1WfELp3|q6t2sA_IV_V=bjw4xPk#M#{G=l#h!gyd4 zlTbVp48U8QHZ#5sYeRQb_ge0~+Bz8ckN0hUdF^Dru^yyysL}T@F)|WyA)r+F%=7W} zv1>1lsN;xIP&&j=0^yXiJt`#pFWOiw(qT!9$1TGoD zT56h`J`EhZey*KaWfs|{wW!mB_N<$SJ7Rkb{k@2ukEk*X?a_O8`RvK}u#fKyC8!-K}`}TQ#gNencJS`VH^03n~4;UqR z($YwX`q78*T`(G+HWNVK{kyKvt}k{h8Y9+rGBM(<@ojLz@P?s>Yk_t&@lOCUW!VlN zmXEK0p359;gyap)m|=>Mg+d;1?Q0G!P}pt&DE{s^}!MDC5|Y<1-F zfYU4|(>+8?txRbHgEwgr`!2(jA0us( zko#nQ*DF5|Jrw8zdjJ38G&*p$?MAuM(qMUaGMbK=6Q7Dc(#X^^H{QL9Mh%~{($XG# zdMFRg42+DN^N_#j<@Ili_G4f6wS?@QwRu=e)-|;JU=9~CJj%cz62Tn8gYmp^Nf#Fv z+vY}_Kl_QYfCud>Kr{ajF6Md~!WGPi#(Z}$PZMB4sKh6VF((W7H0ZH>MXuAe@-XB( z0|_z!azmnoh^E@(hGj)X)sR^aS=CQJS|`OiVAZX@sWe62obF zz4OjYqZTIcsqG?q>5%YXUdlW=<~o2T3LHFi$mq|{ul;{%ba~c&{B|59^c_Dw>)^5DC)YIpi z#9YV4zcLv4gR^tSbC)ru^{tIv4}Xio)oZ9>yhjG{@-Y0jS8d=fT0heMW>?+YgSq5} zkdDPloDV;Kl$W;}uJD&}r>mj|W~pV!#UxCs!9@h$Kk#r0*UC+f-gu6k6KGHA074$U z2%OdKqK1vnaF;%o{MdyQJ!%t^HJS(+nev%F$qD4TwdPrfjmNr+RuKw%?xerkmToC& z%;$J!)yW8>e#$@kI*%XFtq*^n*o|$8zCiUCA7B))X)s$a?A^0V&+AHgdC&jsja_G; z5MQnA;pgis0dIwK0b8cxBt&fk_zlbgWb1JnFeu4uE z|6nQ}04zQ@*dFDc0}*=|89R73E~Dhvg-{vnFLft-q&Om?X-P>V7>B;c|2PgPv;!4< zzWB<>e3M!&BSoP+&%=19Y0WSb7l;?DfcwfarWFb-Hs0!oo0)NtIC{&3cR@x@yWl#5QYn!e0nVPnqe7z_ZuMJrbz$wV zfk>B>%uMx@4z{*SHSQ}wyInH4WHla1I+Q*o86vWZF9G94NHs9s`X^E)Vt*sCPk#SY z${KBlPf@ty<+a3oO_BK19qY-4_VM8MZ}j!UGk`Q8@`F%RG$I8+-;WqhMzFh2r`c8S4mTjjIhBFF$K=w>E?elX(V zP&EEKU9afu=jC-9^#e<^d~xl?fHb(x25ep=Xp;DMVw9;7_m-hO)CWtSDy)^9VDPsf z#6m$$Vf!zXKkzjdXQp{sVX#;Jl(L|)^dDAY1{kuKLlj$vzO3)FXJKSTCN<1`m(hWT zf`>?06F_2gIK`G9k*+6!fbTA2cVqhn-v@BoELqS00}m0Au8>z<>DXNNAVK9LE8d8x z7yx_uT~AN>Q9;VRdU|@ub4_=AZv!<3RVZb=a5szxz zIvD+uR7);a!y69CJyG2So}Gq)!JQa0>^ID&edL9vVD#mx>bptbCnZ4IT>n4vSfPDu zPRk=!Er)6unA)3KQ(`UQ0bKzbU>5oy(gTPXcp)@m!BTyulB%qz*z1qKl5gD#&8ggS zu@j=A?_rhw8#ieH{yBc!W+-SMoli^*38N;fM{pI26!u1IdU?5GUnv+|IQvZkPAKj* zmI9W_w;ec7qeX6h#R?7un3>J;2N}8b8W2B|Fn2pbGaykGdMBx#F_9o5Z8l-Y9<-kE z!yw5SXzN%}J=2qUkYuQ`<`6EM!O4JGir^ppxzA*xBw5!~rO7VeGTlX`6f>QsboSq$ zSJ%VQxHBG}u@@FwG}Q6i>qXpio@7)MG9)7svi9GkFLv5C@lKi)XRiIZA;!}*3-=oo zv|dr>r`{;e^Y8upc7v>aeWW7%<&*f(IZ~c<`3mvhq%WX-L`GFXsxMr=f`oZ>vAIZy zK^d9Y&O<1}qjD7^=WAa6$wLko$IeE|tUVt7-)dIHLX1oxmzG!becAq>TZ&G$l|%Ql z4BXi@0zNu?J~J&X2URDpU;X{e;9@%Tl^M4$>oCa`a>g(%eudS7>ZVWahx1$a`=WGzSGB>p0 z;$Z8PGfQ5bT3YscZSVVunH>|XbqHbqx02OXp^%suyD^^a+iiY6=x_*MFO8I3$e?mH zHT`{tO2Q6XIXs`CI1?EDIt1mT+ATIwLX@a1J5CUrgpq^@YVX9xvXY6f4-N?lo1nZa zx7G(tM&?sX>Kul^U+1k-CX~S6zkiSU@F@CbcRIAgs$ux!IW^YkFr5Ei3P1Ri2vZm) zHnX8HGt&+QTBN<`@$-HB_y`x5S_#lrKNK@DT^pPt=#mvAB~{AAZlsKxPOBgM^i+O_M~3-PVE1J=p)KH4}Qded#B@f}L=G_+fuqU9)Y?H_X!bdZyIA-=@JMx*WQA#&O`ZHR6;1AqR6}sD2v_t`2|xeIa%>n zOpI66XZ?Ow!o^_R$Z>R?9YmG|WdQIYU><$WDkzizv%?ywp*A4?60mEqmH|Bwpj)@n zyecCl+7Ucr*|<)KDaD^x(_AkFS3T40G$V$x0$)XJf>TPcD&T6ckdplTsxkqAflzg` zYMGbd;xb1{!I4O(%Fg<874v&;X*4ddly=YEV~u9zy7v#O_kpm0->;2Af?G0w|JLU1 zQ|k}t4{CjVJ?;&=WsGK%%1v9d=Jd=YnVYqlXyC8c+r33b_LR)NeIJUX#+Yb7k6Qmo z&yS>Hk=WHeB{ATdvQco{ws2snv8%w_JDoCqtvYK{`fcxR1w zIIPx!IOVjow5Dl{G@t!M!Iz|;7g9QIeZpX^NUz#OzjGo$PoLnxIAgSTe55nii4!fk zdN0@l%NkI=bUktrwx`6mN}}!TOuOVk zUjAWrlwEGESi_mUZQ>2jZAt=mmd=s$kTanxUZaMD;C)|)VzB`U zcql(|XoDm>dmnbWzdDpyuqiP3_*2^oI9v*lR)M%)s+=D)V$~u1tP=9)Y?Qu6v;;5f zb%{HK@}A6HBV~P{mfPrRXE%-g&qio8BLSBNOOQ>6C&a37K~BxU=QXMiishA+%gFCW zGzC<53Y&9sp4HK*T#LG*UmV-h1CIWU&JT#gCmB|e2I4%O%^gn3n`E%c9po6^da?E8 z$mM-RF~Vc=ThDH1h3gp@=A`T73tTwc5~k%yYAq_1t*tt^LbbH=%_9f`@a|}FfjvDK z{L9wY-;a^G1H#B)vI8*Dcw~4uy!kdF+mVqQ3WyoMChvQMQP!m&%|R|=?20O#li zeo2=dIpL=t%=MI1O;5T1@!)hb`e1c5GvdO8t}w5$X8InxZOs3ARY6(V*T)B-OnSxk zy8$JfYZd4|rCXbp1dht-Og9CiKyM? z0`3Ri271FX7`a;x^v%(dG=r$*b#uBn}S!mDqye$ttCZk>k!5y%b~)Xttgiv&qctqv&1 z)U30!Gm1@)Lk$XbaJeA8Tnp9RAT~MiuF5FeF*Sxz*C6v4zRXq(+9{y~AHX1No{66V zEcrJ}1%(+T=HPL#RYgjO9Xb^A?BPa-()9)L@+T`ox(*wu04|n|qhpbph@;~iSWQZb z3JR#_XaCs4bQRv2@Hg0G;NYdSE})DI5fZ#Q;yDf1&p9>D`SIgTxwSZ=iF5g*@7t3O zpxh&3ebBQd-^rFBhrYkz1XRayxT&hUy9hZN@W{OKg8(*mA$bMSAy^~!j1Qm9S6KeYW`Ai+c zpT$FevTiIfp6^+2FRvCbX@Yaz&<)waDnDmWvz_EooUT|i&BGd>g;m6K_Id;lplRq@TRA%~;;f@o9U2&T z$P6RMw@g{LW+IXxTy_K)Rfcnq9zCjlcI=GW>C>O^F{R|U>SvPNM?WyH`>lr=X`h{^ ze;>p80h2MlejS8Juyy`4zSRf>O2c7j*bQyffvP#MXeJJxdDN2JwPyF_aSzKS^F}aS_FtA@UTi-+B5IB`eM(|c|X43@csqmiV(5t z(q{bgn+?=3vm46%M0eR=upKU{;fScI1qc_AZ{D1bLDPt;vy@)4S1mRyDh}!gWMoVM zrR6yS&hElM6yL*u<-dO=DATH!4PzC9)i~4hr>50=OGMobT;@dA{mt8}61F=jC;+wV z6uK^evX};x7TXaH2eCq?WoIt(9wgr%6q$J~hMH6P%uwJSsEGm}lRI4nvvfi1BfwlF zxTs`s^6(_dD21KCb9=m(l5szVFd@GM8741O{oun#L*Ea|$_~bbGm=;*de8qHp3p=L zjx&P1TgAx82wxr^8OcmXHwXTMI7nQ5oCQhOrb#unc!BYoWTdrz%A3-}2okM|zt9bWjdS7j%I(AheC5}fD5=U;pXor|x?xHq*{ z8H(61C@$Ox3-^M~W^Uk)~HNJT<&Hn8ZsYNT!*Z`DFdzhI)<5S<%%mw3W z1`HdeDKP%KQFacNt5p&RuYy+{QWohH0wjRN48HEqx&7r!Z(j{{62u|h`n9|{bQp~o z@J(>9EA3WA5#mk){QMJ&5`;hqwlDE(U}7P2_V)E98XC<)cDQV(#8&tAuKV}kdZ?IE ze0HtPJ(=5=ZV(X?@r6hsbalfBCq0K+Eod$L1~G@Q*m&t>gqK~*MJO_ce^KMy&hEMH z?A_q(38i?uL7N##@B~1NC*ZB7RKsr2>A{j|RCSUzTeogO5y{-$;cMImd9DPV-%tQ2 z;8XD74C}CtgfG~WyIa*YGSa-^>_E+MjTF=R4m(xGd&nfBR=*DQ0mbRl(#MXirHIn> zTIm}KVQE3?3b3?mJsBNDsyhYUVn5J>MGg-Se+@J-^!G!LtE?nD-K%)K_%ItA+BOu- zk#;^d*v82*EbdTSTV8gHk8xz(_VU5@LP`K6Z$tN0FRyJ{ zT3VavHY#2;eFK%h3l|uKzSYJO=AzgzXhduR8U4ygM$J}#PgUmikUYKe+unH~2T~Ko zo0>=ZdwX%VnKliOGrn?jd43L6^phu1lPBc~s?Yq%#-S?M*JnJZMpl|$ZA3>GFSYth zcnu0QK3}%$C3!jBdWw*-Z9-a-6yymEx1ITo>U%aE@l}I3QZoT(zk)}A94XMbt}j-M z8UaeD-8}uq-~oVB(P~0Du^zMTyt+-|-_-L=TQ>jhI62xqPP+H#cNS8233L1Gn31;= zy@*Km)yNH;OfO%G+ThO3TOkRwW+*o_@0tK4fmf{R4prI*kA~`1XlP>ZZd%_1`-*ov zNAF&L3de(QdyGRt9li#pS-XyIIy$(w z)B|ku!NZ44cw=tLkC6_L2#xS84Nc!PC0$StV!qR|{w!DbGoF?*t7EfpcD!{f2lf(h zeDWy57s~D1;f8#XW@Z_K9EakT5#o(eHD1*n>j41$HMU|HxD=fXIW^!vC`q{^NC$Bh zQJ;b{4j4!k$Yl`MGdumFRzM=;zSQ15q9dUCsVzJ7b%fY+QXoNpm94n+%NH27vtgiW z!5pwPCh1JwC1?IONTl5rMu7lWka5A2CTpjCy0c{ydy1w0!qRZZNv0D$|8>)nch@tD zG*oN@V3Pxze~ye0bxH-ezUsaU9OX3Y8_EJ(oruq_XS6QPcb1DjK$!LZat(m5uyb)R zC15DfC7z46TR*76Fd*ouVMj6&7aQ9rth;HM2r1v0K=J$U@-iEg(X75E?BO&f*?D7M ze=?T+`%OFNLH(Lyz&l_O9FSdOZSF-N%;g+J%v)lySyA6}*B+8_=TyQaU9u|3QViw_ zh&%T@04kbd{;eIaWGNI-v$$eV16vC-s-_0JIjYK|s+%;`5*sf_N`})voh67mx95L& zvfu13EWk=3`+WD_akNdN4wxB=9^?wd(5PcqAJ}VQ8AtFO&DYb`b4cUZWo2Z9OutL3 zxwI0>2+?lV3u# z;K=&WrWVY$-Q7eE^jg$mWFGuWm)o~nV$`G6bvC)$W!l&pAHN+>L7c{slCvHvii#Kt z5(#z&TSn_hU!R7x6+xSw{(U8Ch^#b0q=L8NKU@GU6-oxqgQv);-s%X-6ABk((RG58 zmv5o=0L5nbgce9@cT#LD-?hF;#CAU2iB6L+^tD-ATLU`@;nfcd3Ax^#?vYOP2;mDx z`1kY>eLo{N4!EqMf4~=5BN(wbd>v>S`+7Ca!pY;%`X*wM+qmpSeF!8MT4-@(S!iXn z@(d!tfG0CR5|8{s$hbVHrA6DNdHV?uVHmN54n6rbmK`qi#DhY_M`+Dh#$M~~?FAwO z=A69666~!BMo{kC*MQ&_oHhWCEvf3!nNBVhH7X`G2!sCf9o%C>+J%LMpqDlP??U2^ zqcH-H-u{S#V0xfP!N14P+6soHW*C>nm|+0a*s1 z_&SFRazx1risJ?_G3)dh`YYLFQ3O2)rHxix8NxPz+)2J*a%L_)^iU(Q+*IXGw9nb0 zcnfZI)!``DbkOUYLe6aYURc03+Hu53mLo7`jexKExb<)`ZY{~M=nf7Y)?5m;B!j%x z8)L|0LFs@x{g9F0;;sIO$BcsFC0H0?m#yUF*$CrLG;d_&6;e06lkkR9 zVm_LbcB?Mzr<#EILecgo;Mml7TLh+%rld5a;G`hkzQ-mYn#622bO*aBZ7BbcPiB?& z?b>>5!9|a}S@{1V?yG~UT%&hUjAH;t1(8+>2?1$o1OYeQ9g1{IH;4jC3T!~SyF|KC zq@|lpNjK8XzH96G-MRnXnLBfx(K&P2eEW;{TWh_~dY)%-*u&3<)*+_vLzNCpSGz%> zI}o|jc#|setdq1fDg?Y+uy-qRH7Lf!zGSsLh> zbgS&zCik!mV&KIdc@WU_H~nVm!|wwAAOYN=WBBrQpgU!#FM!n~s)lgkk1$j6FU?gBTfDd?{1}X z@H8k=gXTX73KwQ(6zV6#}#qvaAYl&aknu0bymNV`Q{| zQ4;_#74y?yYA5ydbaq|cHTkbNa|yhPK|<_dOX-O|2T|Y=0K7Lg+Pk`RYMkBx9s;q^ z`TX=qkMJ6j`@h-|35yO&Zulnuz`zm@XwvkH5GO~$m#YYGLS=0dY(>A@_j{CR5RWSU&0G$l3 z|IgaIJzZ{rij0eUYyumGjg1Y}m4%HBjBpD96(-=3oLUd%u5e&9sqsANsp5YNRH#1`XT48{hpD!e=;G8|}C4BQ4l z3f0lcNz~b5AIP;~6TO#PSVv4jVP(M*Ed@*vF1y>sqtjpAL|q^)gU$HM!{10GKJVRk z^{_*UipxtCAN-R1UrOqQ`w|8*rXMs)z}f;i7m%uIAwgX}$jHnDKW-{u)&tS857^aG zG~s0aj0m8lW6}w%nqwfozL^z6l#knVKsF{*>mvb{0l>H}Bk$Ls{~T8%;?sl9S7@rmD;T3pc$)6x(Fv<$ex zYha%SQIRg7NdeXn<|WyjbjKipE!*=l@S1_TmPB0jt&P{mzab$;@3IpR>o714$U(ET zbLc2RaJRg4TTbB8rICfAfRL#A`u`q4@~G|5?)6l-<97p^ir7YyYbuTmKgt6K6ygH; z1#;TE|MkCk0&gbEN=w&vaWsRM1dw;e0o55FzX^-{eD>R>LoKnx^3dq=OFgHg^v>k^(})ZF|$touTMS{_kT zi;9b9W@IGrI$Hi6mIK}eMT?7Aq>DSru9^el0XE)RPM zf}#St`}eV$n%NECb9YU@0C886N0&RR{=VCzyv!0L>QNx&k4Q?5Ssmi1FmQ(;IKh+*u=zS zC;{^cOlRE$_VOS||H_GBCJb@?dh7pd`Wd4JRk3=aTMC`x?x1Q6t0?-wGnfgp)e^d_ zXHDh=PaIgCpxFl-wU{sf6szG8LQ5oHyPA`ShljZqc3#KO?ExU6jkjlX)VViBbSQaG z$cM!t@9Wpyk=LeEIy;NG5(JFe09S}dU4|nkU4-5ab!Al*6w58m%>p1W1)7H@RKz{uMO5Z+WOg7k@t#V}_R; z5*RrC<<&bXuvq!i0<83a^@exj2Aj%5IYWMQje^2Jek^EAwzXwsG`ySJPwE6_XF zrcsn)7l28&a=)H{jDKl4Ga^2p<+WWz8xc5Vl$yF6Xj@EZ_(x!ZOxA~jvrFf|F@}7e zL}g^8xd<--yhCfgOs}<>$g`fP6sqW_nGO%lwW+S&y_fZ7mfbd^Ie3t{zV#;=8|DPX z0E}~mK7xzOUGr>U*Fbw&B9AI5ucN)swi^K)pKO_UGdqtBZbr?FEZW+Wf-87TJ}a2t zRFdNR_mU-axp!S#8Q+B3qJgVAQsdy>W;Mgdj}t%TyhF%3X6#$1_uC0P*Ha- z8mQ={7g-d;F?vL7L&%?_IkxM{AVG9c)ycSWb9Uk58WOJrmKa5m1Ok?N_mJ>J)J-|_*Dfq1a-AU&%8ygA4oeYVmoQa+huI=H5b zS)X2jCQ~`Uo)LNlRcWwtU=DgIa9=#v`+ zfT6o|{{JYLaYDGS(3dv{ap=7lI9JP*HS4%5{W+;e&cM)gR!llBYkz%yHMh>aRL@62 zfmE9TbHZWz`v16_o9Ef50!E^`>5Q@{SLfQ_oPrx7jy-(w<@{Qnrqy76xrBnvR~YvA zCw4`AHYRz!IhXe7U;EVz%LzM_MUKW`F9GHTd|{xJ#WlGz zQ@ea62XQBeG+Q|r8L9F$!tJz3;8YHH_C7Zx_l4oOLh}FxJ1+Hnt11GRF!w;K&aKiu z6}FmJm@Z8pS5{_ga&^X}lkp;vY4aU>)9hM~!~W zx;h@q3VUlNZq7$rvk{@mb?){j>+sNCx6pH`zHa#{hGE*gl)ZvolLn$qOu1t^*nSMx zm4WGL0!R_i^pNiU{b{H<0FcPmzJB?oxN}EGB44(2EW1aIu<&(8=BcbuaNekrDRr+F zI1~c@rQh$c1W%p4RkIU^TC~SUcg!^;ZL^S1Q-heK>Bo;6C{&=?hkNs89)x&6=iquE zWz|LLdSGblfR=qxO~#V&4hYz&Q3MSnj|KvvQK^7^wH>7?9 z!UaJSW>`5kj>9c5pJG$8{3+&qA3_uF$xO5nG9vxkyS?&y?IPL9>MeLgk1_`&FmEWE{Oxj4AfkRZVaB~y$s$lMYUET1~+DeXk zF_p=-emuA0OerQk6WqH%-Vmf-qgL&h`{uP~V10(i88C2poE4bs)V}}8-~<|PhTmS5 z%T7x(UQt32vx&6?Yj4WGZeBXsg!dF;e`idD7G|R&BnrEDYiu14^-NjGzKS=*TpP{7 zwNCQPby)-q5-Ci7P6{BziVGHFQ1w(+Ucm#f=tFb$rixS`xrdsFZ@{ensiG@cy_B)-#tk3cIshF90buqOc?dInnxL)t_)l;WDhu>e@ z-d3J;gZRN3TlyyTct0W}L@an< zC_lIW6Der1VJp|}pQAjO>IWgaCB&@lwvR78hq7c;jwIpbs}N1r5IF}PMyW0FjWGQ2cFc|4xSPL7WvRnXw5S*jaA5t z__hwKzPoBRE2jQ9nCTYgyBXn{4dTPYtL*#z33wc&BqOtIjndOi<|TZ5vJE7C{8oRz z`tOi@a7dd}6~>@@$@!XlEfi38Ubt|)BP$)7k#v8;B{!03u+sLw!?M6(=R@|sxtix$ z7`vC69&d>UkCN z;cZyeZnh$|y&)VIIweWpbJY7eOU_$wK^CBqhNn1MJeU_fRzBRp42Yboj^(z#v#VJr zv1`H|_>ES0{p7ZePRzbvBIC;<<7L--Z#p2SgoL&6?n+tm-sosk?T@$EF}$jkl`h97 z+7&NDQs7#>zQaSMhi)jNdP?}Cy{(oiv;lV0SRl-Zdmnv5c}RT>U+)JG@z+i@3vo zlt$GV`-qFn{rB54bGqIfxK6ocVA3@-4=UsFrxn%@g;Hu%tGFL)oqR&#OD8fRZrp8c z5(P5OULv|xuUAo)j;mb%1~}yirGtpitK;Yhwzu5e+SQA2Ub`j}_E9dotyXbQ1{6sh;G)TVqYc27=j2jVL0oQLwUWvI62yyiW&Gt z!}`G1f;ai)+x<=Ey&2M9L0s&1Sy02JOa8%*le`}`G#1#{!Yjh#*adJ3_3DhM^ym8s zHp48L_>_|pv&;t~c3UaUA@B z>RrUcs5$WTv1vA=JH% zOAcgD)>s&V2s`i9VY6+0M)jGcub78HdXaApLRjr_PEF0+THbqjEk2_c$GMdYJ)F@* zY=ASsd-v>_=@>h@zSN92#bvrm>T^=D(O`9Xm1DibVrJqH%F1k02VM-IH*6HUyY_}G zX=}|bW?sR;|2OFh&h@4fh9H#VhewG6O#x8HBqzHZ8NiMJ;E)fuuH)F-8Ng`}+<^PJ zHxm%UA`+IP-0wJLGV(<8WVs&#;+r=)3Zfn!M`wzP_A~F{9&`&5^2n}_9iL2~2THOns}PP;@tz4VOvOkxnQb^L zIqCl8%gg5{98T5t{LNan`s|6Oh`w4y1?2_%`K#I@pJJfBR(kJ_EQ&YVh z7qH}dafsQM0PE6*LFk}PB-55${&kH{qN$V1>}dPpn6A^ax;p<5Rm%cmt}ZLjRmS-V z_mN`pZ{Nz+J<*ySk!^b5Ne7Ig-$^owDt09)sim((Ze8%0;E8|wO7hUdBf|J8vxL~{ z5G-jrI&RKhD9`R=A)Y@|g6gIByQN8)1_y_w){-BMtc3sV+jhHIGqVw&k`A57_)ST| z6Jv4)l+NG^{60%-wDo^zqS&6S_ZN;AT=bPw)H_2_`_6vdE zzuRrrMr}gOwNAFImcDh1m`@Fd(Ezd^kt2e+-STIJTps4K~ZW2O*xrhh^3*F5>iUS z-Vi&Io((@$mI|D7k^GZ+mT&S%Tg(`sETrROp-+dB*;LO(cy$a3RI`Oiu;N~`n|k9JBuYA0P~DaFUWTQu#g zmmkmFsjOUXBD7$<^T_nbMpWEvlof|;rDq26qmPIe^LOBP?ZVCpV{C%BERMXpFgG^< zJW-%h3IYR_L&ABG<}@PLtoLX-w-BE&^T+eSeDLgab{<`5LlubtVA<%mx18y6W{;}g znw-p0ktQN)Z>QJu-h9=_Zl*G-;Z9=xNjZ0S!ZAntgm|<3KcSM2L84>4FonQCBi?G|;y`8yA0MgO zeOTJY|NQm=wjxkV6oFk-W!N1T7l)BrV|d?NO@u(xI~#c&5+Uqd@ygJ)m(AUM`y(q` z#HagsA*a#nIjY0hcuN7ycE=?#9Q}FfWi5MGaWPwzg;PWy%gc{E0U^tHs4^O#J$A#h!t!Q*KAcJ9sT}ta?!%KW7VJZbM)>frp;L`)q1D05e zTtr6Nc*CW)1qGs1rw>!>_Ufr$Eh{wr$b$ZDG*{l0n>{m4V_!wtj7;~}Zdck7?M}M7 zEK}#J$^YO5C>bNXIfB`s8x^y`6%|icqo7lWF*gBp06=sgAZ@Iy#DL@v)S7@cH?_2Y zWzZ_f48p*>k{>((F-_k%?*7ohSut2yNl6_fQHovJs! zlEt>@^)lBl%``ov(J9o??u^x)q=g-xc8%JP+K`%hTI2<=F=UdlK$(twF>@>nW~AU# zyF-)ZxS2wj0$<+S8)o;*+alp~^~m;@HEojkcV1>RHkO>iKfgKe z1{i=*?9eVbDJd9E&B1Gv=@xdz79J+U+fUDDKh++gQSsaVEYVda z?0gH)eJ>(^K8R~~(^I#oevmR`Dc;c>rRs@YA%^4ZY|`L$s_F+h*ZmlOV$@tk$g8Fn z0DrNNG7gV}_$o(LpzX&4PM?gr%FS(xQmLz38J4YeMG>c_+E2Ly-Cj8tbPK}rto+(! zyzRrP{H_ayJLDv}=Vy=rw6y2ffnK?Nl9`?@NeFa^VBT#j?0CNHqSy8-$nlB!Mb#zF z=VHE)9Dv?D%!H2K1F&|})8fDjaA3RhI8ytiq_I+Hfybc_Gef zfZ_JA_dMS#e7Ha9R_uGTmv__n^zf8}h2aaGqIrEf)Z?&;4;3fX3K4lON=N%4iUtdd zIsnTG;{yYh2rsI@UJk=oj9PW7?VFnN1lB*0+G;^y16_QATeqHh720LU3=U+AmbCmZ zMtdNRQB&I_uDd;Wg2tOVh?PE3MooIQ{{03%Rs}yi$v$nz`2$$!K({-XqRtk%>hpeX z^Z4*}dPcjx!Z)kR7K-7n__7_xrB7%o^o(e}R(1g4{2bDg>scQ>{zg0k3T9x4z)^{a zV1HU!d3) zpAr(;*pwKB^z}z16ny@CdR}VQVP3{~qM9z@fmNo4jLc~Cvp@Wjqtc1@+o)@)j~Y>B zhlpaLqfG4){1(X=hEEtpG5uO^yFYeM5nuo8z&~&xn5-;vZ*|8i_;k=0Cu|`3oySqr z(DwPs@$@@}ech55^lsR&it))kn$A@mr}7Ky%Pg2N8;pWp&z}4FYOhc_Wl|vmE=>uR zmNxdKoog>k4TFw%Lz@(y8x^Yu5obGC;E*0_4JZ;30e*JyYigouo`PmxqMcKTk|C~qbvc))i9ej1`IqWDe@7#O9MbF!Yj6Lu8;T5SKz8*q+%4WBDlAchCtjPAq@ z33S~zZoAFRyDEt8%~OwIN4OtneU--DX(9v=8+NoOn}z&tup4iL@P~0WPXi8A9ng`7 z!zXu2%tviNhy-1?G3c0p=5A#u->fhUWJQ*>nbxV`j6|*Oe=0abOJ;#uSVVbO+0m77 z+LfA?u^frYJ+n>|Oemyp9Uk5QG9Y2!~{>@7(_SHF+`L zzCL8E>k&gMtS4&G#co0kiW)Q3^bBBq#PeomNZ$y0_d)_X_wAjXp1MNNVjHX1^~0rO zQ)JYHj_Rp)J-T4Fg_^*H=pa;FskUl=L)kC`km%09F9rduYzC~$5T!`OLxw)yfnS6Y z!`Gj6C*VK!!V%uaAT1qkFj=N1wKtV0Fj7Q_VELzGf)Mw3>h!94@`#k5*P zsk+R9$OpzD@MgZh6fX$e42yEcJkxv&jPIfYQ#gi=wv#a|IFZ>A_?>o?<$#(zDI; z>j&ZH;jaQ8TPsB|9sjzD+7F{n2obCss5<+p%X}ZPG9(=uTDE>L#eBS2lH`qzU3Vx9 zxnRmVTW4Z_D5NQ6*KU69Mcz~7T3Wg+C7+7#Jd(+Bv_d%wcSQQ=u#HqI3JBQPDkchCzfsJQk!d#a zB>w4xVs%E%F%Nqur&jYJtNLsm{>rMAN!J}unc}bP7BMmj$l~I$lFHL7pUTT~RFtEb zj_SDpAo}Ge@#52H9{AqZ;=kfVT}dvn)Lt6QA>*}M`y3ZPSaXza-tX<1TZF6-UA@bF zK&qDdfrMePYjlJAuKpRoyjolrYdWFXLJdTfb`t)RUXlf|x-6GAB7xOPP*>S`?x8+8qg82Gj6PYe6=V_`{d zr%@dp5yZ^#bE}JoNGKE=hPLdYtcHd>-R)1k#7{TF`iRTw)c8H$`VzP7S0+e&LiXkL zw;XkboS*v1`e|J4j9w%l)iO?U!zf|H+R{*#9af}lVxyyD)qt3qjw)QZa8wOt4To6{ zydzv%2~NUS@y1d-iVb}xZ-GlyU0r*pa(BW}P>4W|bgVg-OALExz2~_u7;Zs@xv}9> z_v-vew)U*d{_`Ob*HkOoYO`)OHD?8s>zL>?jgRZI0s>*#_A$dNESmRAU}M>( zNi}A+J13UoS+r*ROv}giW-I|xYrGN;8zs28p?MQr>X2~X%mQ=?$zQ9eG<}AW>Ii{j zuYC1jX(${U?@pANTNXKD9J_ zE%*=|T~M;OzEa%U*=qU9ptDM6QNn+Z9w&loZ#`F501?7i9jp-KGkhgy}pcq6x|?PlK7D=V~xxg0O=qWf;$fl4BcYccP|tV z$QD2oOkt z!(ePs{gk5~P9!Nxyooy;F+mbfQHel%l9_z!Z zO2{;{dwzj7!RMse>Im(tR>_%b;;Zgb(mTK4{WJ{i}3AbwYCz*=z;}Ge^6H{jE zHQhu%=#G?Nz+#H1_o}6@Zb>$r7qzBp~f-ZHYX)r_D<-}zVkl{If zAPVR;bvH+`CM=FAOJ*aDxtL`OZ<7;Eb~gw)yJ&1Spj7ACT@bhLM1207EG(jz^3%|a z@MjQ7)f)!r=X#=(u-ULidHtbqH{WbUbpGtx(IRw$hT=&$vlJh#wh~%d zMKBnhMZkUd9 zqB1i~$DeA~&?4YMXt%1Z65ZVMDF@b0{#7kdk_xP;u|uEHt1`KJK7S8b_rFPEL9BT- zuBR2yj9ADp>|{Mn{^$=)(RXyB!v)Lj1eKNP2Vz7D3VCFR{hcpf4}C99<^VA!43|QH z^wur;(K<73LJ+kw17T!h`6Kr#Mj&)jNLk4VYC&MPt6Hn2UB|Me?{_--R!44?YWBCln!UG_p(X%~K_%V5|JmC8tGIF_a-E`R zrm2|h*Nj}Y!jMEi3^_|z{?2&a}>%Z`TZr)=UL|J&Q7f=FNM)u51S9spoH^9zi0``wJ|t4 zOfz~HrsINuPxET!C@X%vz3hiep<mMJZdypUTBXFB3VJH%F9Eow>q=}6rril-isSd;RsoQ z=wGUazl7uY8MwIYmp=VcwMTKr3qm`Le^<_>CFR^IC#emYz3<G%%Tg^apP%*>siA0Y7fk|5Tu z3=(rOWu`pyS!#c-AeFLtNnlCd)X3N417wXpCp9VAB1lzxZ85s&B1NzY9CK#vSqb+!VL zhv)6vQ2p@nvEutW%@a7gYMy+iSC*)MM@CWj`5wH5GfixE6*oz<6f^T~j4Bo6DWvy_ z_y?8*Qxt7#yROi=?*x_Bm}OL&%gaMYHeJ%_5U;-^TUf+?-v6iJ(XruD5~O7u;?|Z& zBJy~%37!t_YRb9vPo9X}nlc;iJCVM9ht08o$SiwA_6Mr?kvmdCKW%o|Q0(B-tBI;z zKhM+9y5o@AAd>z&harM8g_W-7HFd}B&rV{w;Qt)$^k)uOjOuAuQ->#dcJ(~aTfJST z9NGE9qrE$ZWBIbndQ=2LvcJ-HC>ZZV`Mtx2((2F0q5@$8rHI-rnyT-71rOUhVQI z2IlYpt`06o3jhrQRq3QTm<$DE!mp!_ijeYv&BO%GDxeGnVJ3K`h!Nvq(dI4uESlsb z5!9J&nHsGuw=@}Jci$h?7AP%JaCbPWe^b2w{jzoqSRpq3prx09xcU9Es5k?S7?-6! zZ@6|?W~M!@hxM2C2*u7AVu92Bj|XOjbWB}5KkkRir(&9wzh5z`hoO?MGnpP*M|To) zKITV3tJQ4Sbk7urJT5s0`+DU}Q@@ee+=4@}d(a?idmk)8uehd5n2dhqHKw@hU__}x z%rz9v=GY_PhkGma$}<{%mn?1GmUJGftNiiUSAd_p`wS;~ppaR%`=%iBfe1cn-iLpp zJBJ9Y%`GPfrp`0;npnp-$k|&p(8>xgU-}ZWymsRT1lyyFQI{EDaaUh9=RH4wIkbl9&EQzED!T} z>Y&dZ7s^}pLeB!hwW^QzDW`=^Po={`AXF!^(}IT$V}@dK zrgUt3qsUcU{R&0nok7)=N$t09dEPv1lp6W2KS&QaQ__7h!GvG?E3sCl+xi@);D@te zXpD_DE~XGxDoe%9EfRDS5AW(hL#N^TxJkUQ9ul_*AqjJ{h}!ifsywtiqE_1pCTg-_ z*1&b7Ve#!d5F_2(l&T$$ABXj4ayaCLh41x>6J39Z{{i&S4&W!1=(df}Mg8;Pp%?G( z<@QL%6A&BJfC~lqv2aMV#Q*g_;ZB6%c0ADZvMft}45K|(wp#WMXS;A`3NBrSt)~4d zk3dXV%IABcVjAIYT;(@Y-xPmU%X3T+544U0(~}`-GTs$WbU%_VhWCK;m#Fya5aZEt z`^3bS1it69wl|B^5M12Suzj;ux zZsbC9eooA@i#w*8p93fQZok5a>NgP#+G7G$2?&3nJ){AT5oW0>6c>!qBR~bCgqZ?h zx7?R1jwgH4CJ6)S@U5(%B7OD4xJxC!udtr}=}62sL!*6j;7$14*!zA1nxQB|!SgI} z@;1JLZ-)w|N?lCy!a6$iDpCRh+uL6YZY^vkzOpHbjdgRMUpBEY#CPFZ*+Ru!zSOG! zwP4v}6M%W6ZpdZ)XtxECCjaFLM;Jq+;|Yz#c_)@mVVVcuc-|lNJXmPBD0Kett>b-A z(tvR|W=@#2G{48GlbxNNFy@apj05<1n;Itbffqg63ZV-`s!wzJt1GiG4<>e?M1~CK z9-K1p%s95r*F(xTb)Z)?%?T(5BTlRp$MfMUzbc z*-+5wKp?rQJVIEk!?n1`{q~f zW3@U`fo}7@YFB*!&uG?YcE^HQ-tK55efq~c(5M1BA7zL>?PMHwcpI29poa{lr3N}< zhqh(8tms)-+`0D$luFH_`YScdS9ZGdXXk8R1{R=~D1*rJt?Jsc4J7=>C$ybvYkO1e zXMLg7;^PB^NgOhbxtxRav@BggkUo~M=c~^K7Wi!a4;Ns~7z^Kl>0}X_$M|_HuYseC zOQw`8To%x0(3c%fuL8>B7%o}RdDyyDs-(P^Kvu-!UMyB`^y|>;9VxnwQ!71NJ~b4j zYr3wHubcR)m(~>?FrSuc0qs*Y0)`wnXXS#t?Rn**6k%bdWN&&EWa1QAi~38CV}!0d zwXQDUC>OB<^!5b6EDZV@h~s0kv)}LD`4C*Vj~5DVFkUELCplfW2ME~0Xq55_xUVST zkj0>4e|xG^c!JL(RMG`Jl-}OE$Ez-T%D~nGX8M7NOiW8RIkZMVv)JeT3woQu6DoM?;~lA>G30($3f=#ceI~wyiw(LI z6i=|QruQYHhJZEf{;D#DLt7yCW5_T-f9$V)VFx-$*XXyJ8Ui~@pkCEH0SV&%T4-9X zDs(czUWUx#Y}n}@QFg^s=&ezTlM0HyOakogXd7_c-&T7oM+w5vHaI&`PWKlJ{IWr3 z3TINgA2TyMm>&!Jl8GuP{{skD=cD4u?s|RiXhKI`#PYf*UsvOSo)~@hbWqcZjz{DC zoX1uSxDBY{?~V4S6*RIci30T6JSfArZ-E^hM8f+N85hT$Th7nts-jxB5+X>d+u!nt zKd~D*=jS`Q%D$Hi!ve#oI^*al5_iu;GhGtXCaQsowY}Zvx(Ph{{!E}(0_#74UpXVN z_#L*~3(TXz8lMEaKfALk9oO~X?phnsDUaO6gzDHuMinV&!n)f5tik?#ixP2QFY-`=_ zdjhIPMaCmfpd#Q_;4pmww)ZY`w_t_P$PS$zbp_)IJrKPJ^jB-wjwrMKjB37e{kESU zEfaokE2w?`{+*Pf){c%pF(h!{XlEIZS$Al(n8$VKo4Rz2@o)i(uaJeVVM5z>;px+u z*cEr5hRj5uW8;&?C5sphJ?1da5)EK~>9sOwt7DCIo%uFc^LU|FN94(Pd3ZnY%9rJ} zM*>?H7k8rK*K~%vei{rYnUv}pZ4H%>9-RIb%uJ4-wl&M)U@kLwqNS}6)*KR*<~S7r zKBMP6_HbNamp5X1u12D&J;9{%9g!{@+@0~OYjR!>y)Ji(!A`z~1%5Iw>U;Sd;i& z8}z4xpD@q&zADs3;AF^dq99{GR#OA(y~?gP)+)zwGfpFzqOh>8YW9hFWo=?2AD>d* zXE(P#+mnU{I03%7#i*lCAtC+*l7i5R3P%`pXizLnk8=afKOFd5JZjC%PvpXBOqS%} zjg^Vl@>u?XI$j_-J4~g*)P7R+XEfn$qbvk@rDmme*%%qYtHwAo9Dgj#L%^DA2VInh zR;ig;>nBvn+PHc=50G}GW76;hTFXY~+F)~USX`hT-jwp=In)|Rd}``8b0Bt5H8r)F zmyuUFD}MY4>7Rjka<9sFbZZCj0D4`4OwLp>;HW`|$=<952fC_$6u^M%j80Y6?VSAw ztpN7|^ozUus2D+A&2m#rHMG9=<@%?5Ny$FAC$fnt6Ub*>*A=m#;9nSh)hEwcgN4iI1o@HflyH- zJ6qyKs~au1I+vvy-ZJ|5qPMqF%5xdnT7!VPQw7M_>*d83uaO64Q##NOu3WV2C!6jw z>@IUy3M9{;ez~__jLc;(wSdQFuF58$6?&z4CxHk{8NZvSiwa9Jcn z#PDk-BtX+%GSYZa2Kc3I;qhczZ$EqltwPxD`3dmG-d#EVErLUwb_FMnO#K45m6yEs zqXN<7vv*(Db417=Vs!2e7HgxgrJr|1^t*1(@8{_?QGrDuQ0WV_cNoep0v-mVXCmlL z2C`6G2^^BZCvCb8M>OmtE$4QEKC*%9<#6Cl!uid&(>+zgoaB5511+#fx2-1~kA`&= zwc=o@Q--{N@w*E@<#SbFF9Ops*H?UfK5pUOD;A}~A%eWmq4#5=kZalI_HD>NVM>Jw zxTLKo1w2om|9$joMZ$)~RE(MqT$!GyMIDcR4;PyCQ6XH=?gY#pC;Pwx9eDz`0OWH9 z0keIUu=M09hRBgUJ-3rObz+71E z{*8OASe!wn6%(Q}H{GT$?5%-HiQO_zrLZe%@oW{?f*nzss~p5M(tQxzG81>F&^fM- znO9WhCOtI4+N)O>f_4l?sfpq(mvx!{PTT3ofPmSu^VjBgMcH{12o26(nY83`RUCh7 z0N49bB3Kf{NmRoY8%(ib;Ce2Y*1NN^lz)rA)$Ve%Bk4uFBA#m6@x}8lW)avnFNpRDPjU`D$`@qR* zmfJkFjQkl^J$JY>rQcpyw9!oHj2d=6@vB4!Tb<+@K<{LIf3+dHQ$jMF3br{x{eAkJ zi?fpF46VDk_v+6<)f{PeUSm0_1TbF&?u{ZC-0+!JP%whgY(nvOAwPJIJb3U6+;o;f zD0qQ?URUVXSrh$t_(P=)mXM(C?Fvr+z*^UmYFrgDtICJ(vQS<3?AkC49BW?m?2TMs ztZzyg81N39>l#@^JMI!iHtN>YA4+yHlsPt-bHenCu`+Z!fp+Cg+^6emZZ$KrY-@DW z!ka);vw?{d(xW5$Poq!la;>WW{#z4H@9U?2*HhB;6jAH|d1 zh?@rSYhXx-ZnZ=Hl=jURJ^>iDHi*MI7d=Cp7Cmnnxt)W7NNjK6`{0JpuvKcb<%JH+?yjbvDN9}9DW?*khGMu?nVrHuh&2r6hc(tq(HEc>o21TI;J zr_A5s?@VAf8iuS0pyjbXXJbS%SY9{o|M9iJYPkF-CqXTC@x?Ug03hdI_!F`qX#GtK z(8?VKM#5pzt4re}eG;lf%qKgNMY}x#+FA%}wTWZ2VVXy$a*Z_qVp*INZ0s zD1Z)l`uDH=Cn+#w7-hloX#p=WaI_{QC0$z+n!W@=bQp65ChYNG!0_k(!l+$v1OcVG z0~`Vuex3Dy4)MRQ@n$lFI|z9Hzb_wi693O0#&2r}ETLD{U{VKzt_3l(IjnyPXDg&* z+#UgQ0$VLG-s9zEfBLP)c?bM9b3s?x33FISi<0hxce5KXM8UbL-f2q@y0R&+cjB0@a0&;gAUZ%MKaLK6!`~dtPqCxj*L6XRA zYY4Jo*o4Eh1Q?F&475$yjw|G;6K0ylPW$o4>U%?%#0X!4p$nI}Xf*lr< z36O-jxvD|F52jp!^_TaBMMStP#+05UKkSCH>p0}rsQ`~;L&IW>g%36CQI0#mfZY_R zy+cRMnZA6Mr^R>-xdII6wpP>ver)DNYE&36c$lc2r;;BN+Pw*;T~^>%2rSQdGMei$ zFg^&P3y7)b*M#K$?d$8Sa#%lxIa=u%84@`ldjwZLj1!=Cg>^DGoalOJRFl))ci!IMz@ky0#SMxBX8;%BRKWwrx47}S&KSUvaK~)>zk-_y;E@GeDw!xJxr8}VQ)L68_tr)ctVz)!Q>|} zW~+O%G#{nwp!dF26TnmGH-Fsxdig#Zv_2+nlnI6pAn;qFFRT9r273iMc3=fSH3E}Y$V1N+KW zzyg)IhIa&Am(k(j;iX!Bm}Zud3eDubgM+BK*;yFuy@LEF^br$Gf`xnp&b;iF<0I#@ z{C<}N7#SJa*_EIkLV@!UE15ftj0|F^fywtrq`cb=%=M?N;7A6RLPR3DHsK9mE#79# z2BYQj#4){UQXD+I?(PQpR@gW?whc8+!Z-mxoQDq|244C0?b~HilIuF)P7aHL8di1C zr8?Ss;JL8%&SAvU@%+@8i;D{ZtA*ik zevO7&FzSo)h*aP@V@?mYqX~$1n|<@kIl{q7$nh{ zG2akcfKrPdRMq9?eK^NlJX&MeNsNHY0VPV;Sdy^Mbuk<^KLVy-D>JjRK|vtc_rn3W zPv1O_#~zB=E0^)_ZpVP328g@O@xaytkBkSoL)ZN1T^e{X81icr={1325E@1zV$6U5 zb%Q_TVgY=r;OKE8aaTN;^kYCkM>58f4GfRc;HePvb3}iZ$^g4>_zH~SQZe1LKp1(In8eiiiuK?fr zRkAmTj*}KiuaakM4mKDKudq?)YpiY+?=g-9sIty|YcMg|KdKRdo6f-ruk8GnX8>`6 z)f6EDA}ZxubifHbsx=C5)=A}7W%+r=C4vCA%mdQ<{;7Ucgb z@66wF&f7koC5$D8EU7FbWJ;x|s7Pp(>{|(yN~)_gSwbme`I|*BeW?^65=AE z;U2V?p#^PxIiBM=j(dK=G1v0FKFfK&&-eNHyx*s`TTPv1V}oOL zfUT`9*y;T2$jTbOd6sjcOkE_02&im+q%|>ze$IQtPoi-7mdJ|gWKORmRzg?NlSGiH zjo?L5H1{63#)JYAYVI3vZ&wCHHI+LJM##Ags>H;TF&}$<&73##z{?;#$3{ciiT-2VepAc_twAe4-JXJHji_bTlP*`Ji`{%^+X)B*G{*#?$Wt)7-9kJ zv~R6>A){uWfSFVla{cd#24zosBnn&Dtl*MYr@7o;B(dig5lXGgRCSg-%1=0KHT^W! z0GIKQ6G8yhUo zTR}o}i1+$J&V#~t07T1W+h+d%F>do{8tc-i0&n9?R zlyEbfqo| zZ4CgM3A*N(Yh}r^X;B5ENC?td@4HC5bIY^uiZ9WsLbG2eCq5|ES^V3_t2o5admIs{ zfA@&xFusoWTc$Ph9{SIN859r6?=zeEofJ*3dHj!3(Mh32OL5#CP3h)UFCq9}J zi{$8PUP7n7hw02WLtT$##1VPiUkZVH(T?YOi=OH{du?s`q*CTq*tcpLS92r<5V^|L zHJ~Q{fJ|l;n)|~YH$yJ*w3C4j95}#|1W`H^u^*n~UjSF#ww88Wm1qw(YG%lCH|1Y- z$`bYnY4OAS2F@)1B?H}`k3u%Dz|6IB@2QlO6plep%!ma(%9s6-q;9nP&S{=VZmrVE z-Dkzb)S^$?8W?VbHM+9qpA0t=7C(fQ0?ADP_?k6qI&|sg!O?OB@m2^#HIz@4-kTl1 z_^uW!=e+Z@X&uxi2VFmqX|#Gs>kcAfp#F`9JLva~B!qY~LaFa2Xn*7aMj^V*QZZVn z1P(rFq|^3`T+S^|hSG3NEl<^kL>Z1~*L!Y(7K{~*nHk1U*t zF&#h0ZBK9K6sE0R8+M`;iNEa5rPcYLK763>>rUqHm;eCcUV{zykV>%{inwUiJ!15~ zDlk2vn-91*ihPV%Pf4mzWcC4n?d?FQaQ9&37~2uUon)r*<|r}f3ytt~0Rd$IeM(?M z%@QCAy97a0q-)t(+Ux8etgp$(&oSgs3dx`9!p$dvolkO}>BW)yr?FBiXJV7}%Ua~^ zSZ{923!9T36RLYj_o@-RGJ!#UNaJc7VsakmY1pGn`QS+Cr}h5@DN3q@p=^Y>`N zbF2^9?$2#)#F%{Jg6wTwWPC44 z)6m#?%lHs>+Q?o*%#YP*wn~S8ytSebp{~S4Mw*TN=8tFh?^l6*#zve{QjK7aZnpfj zSKN4`u{qdSVv1rfs(_Jp=h#+b*>Egi8P(QD9(0n~f;X9rSs0Sf$VXB)d83p@b@5e^ zmKC~lr}gyd5+E3>>EgwUvd_)+-|(w2GqYDsVF`+xOS}WdC+_LVfqnWUP;JncR5S7x zZgsyc*71<;zTwffdJ~__52z~MM6I+a?OusXHw(X`9XoX*w;qW+f{-Jb{BixJhJm;H zd_MDs#(mzzW#Kx%*Y^30_-bu;;(p+Mr@zz*!ae+I(S>0H|w8tCHKHstNy8zuY6sjJtuQ8z1MGq&CK7)y5Ui1nZKC!=8JqwtAEco=8S(-d)g!8X z`^s_`7x(Me5C4!rmJ`N;yw&|T&#o{1F;_z$ai&*6(%I}#DVSy>_0jT`mzklDA70h{ zo4^@lB@|2k_vnQGFBH*#?vHY7VBtpAO@~fRp0;s2;-^KWzX-pr)Y?GO;`qs2M((g9 z;HM$zlp`Z`*0#S#*N(9GVo8)6p5qJ~u5#b?8)a%1(R0+?=r059ayuw#2bo6+{lSp= zAf=qgD;#*I((OTl55l298X7tqF~(cbA5*EOMVug*4|wC)LCNxaL1b@p1+|C}7o{2M zosTsTo7#z6t>TL57g@nCQN_`4JS0O|C4ygidU% z)<`X_o*JeGW!oRxV4;N$Bo)`fXp6bUPXkqyw37r|_rM5w?;@1WxlPw^(4cDkKe%j& z6t5ln&oIBzWP~YA*H{0+zH0mY8EE?_mnHGV9z8cYWx}q#dyQA&OpR}U60Sb6rtdB#rQRt*7rEvrvl%#GR|292M1FM_!cBpz<*JUt?M>s^5kHS z9wKwty<4}2PoFL`dOBBcaQ^my#vB_vyD0qeruDX1bTeo(O8dlr^RF$h1!&aNn;WsT zexlN;T|)80jId(#b^4W+nUk757HZkn4L`wEmM4w%jJW^v1WEo zv^QQ`Dp{1Ev*P!U2(Blm))yq7uT7Vj4b-)FuKB_cYfPR+YScPBW}!T|3Qkut<&>fO zL$|Go6#Jl*GB^nA)n3GMxAF!Ny>NzKfv_pIg6^C5nPt! zX5;VQ$mhjyeiajRQAzrieCEeejQq=uXrw0-Ce*E$t<^_cmKKCLhD@9?;sDira0!@a zEs5~qVIA-Ryd&=%iqmlW75ve1wN0;8Zp48a3bUp_DvmET$v!k@;iJ{bwblG7 z@hEWmW7E=J6oyz=#MjP4z3^`e6INl|>Oys1cVI(CdO9-P7852s-&hUkLI(NF&r8U{ zz*vd2u7Y2@Z6b|c(-$Czh5OKhSe;Uw6_Ryt)q(MBQ`Tgm)7{69YcbzUo{&6cyvv== zuPbYXUO6|U-(C!l%mfw{LdyG0!;Php_%bq+S{02oScfHc^D-6OY?m>@`xI^OV;>y z_D}ejE_9qWY`V>kL-7WpvQsDfB!En8Fq+rt7P! zDvpE5z#+DUsB%(oMt(|fd?RLkf@mETUjn2HtO-*ES^scL%JkiRYAHGtwkF zs4)k|$Oe1s!9em%Cj$@IK|O~gw{s_HhRNkU^lwDG;h*gq8^nxgiWxB# zdB!7pbLGA{tw<>o%r+7|nBEwA07zWJ_W0#DQq#$;$%Gbi&x{!{(*l02odlX=9ABkn3_vZr|a;w!s}udI7cyFu|_#C~Ca#tq@mBlepkN%)$Akg(RZYuEet?zzt{ z_v!aq&=ULed?Qm&mXrkBAjB=_B(Q9)aayJ{=saDWAXrb-H~{JyY#|EZ%#8PO7&!Sj zPuvwAo=K+KI(AYZ^p4!x(sEl(X%i*?>=Wb1ROy)ASRI|knmQx;sM`W31hx2$2#ld$ z#H|^(ESbYbqOXVK3*oZh+PFPMZLO_~5tWh7D)7BgM*@Lk?*Tr#)RBo!^)SFKQ-n)*X1%hSX4$ z;yJ>T3wkO?)f^!c6J2yiM|bMoyLarGVgeX@De~HFz*E?SJljLsAMgpa;0gQF=|NFH z)~by(rm2_d8~;%6E&11B|4lSTNS6k3 z@*B{CRoeDIePG45U;duH1>h~$Ugj8aBnp6{#DF^IvZEGYfRyM|JQXe+RA<6L|(9fJxl%Vs|+HFo%I1$aV9cgoJ!90 zk*=h5z%rl3aqBeP5?xf!n#7$uw@+i@GyCy3=?%#XL>j2!>m*!bC7ZVKN#7beK z>4Ts$KYGlVw~%RQvxo26b%BEz5{U%N@zNva$eLS%ZLQ|6*mYP-me?h(4Ee{bBA8sI@+^*WUoO6rwEu7APk?N zxZ?Vbv~@GFw8%?OPY37oxS0rZkbVTmMP@;~-AlJ)6?)A`H@CHZk6{CApMoX%-YTkx zCuF)d+zz@ZE6=dD2JG;E*2&niX#k?_yN4>x)3RrAoCGkdh5P^!ax!-zB~+4nbt!xI z*s)`HU_DB$3LUsI;pDLsCnouoTS}F4=gF-m8Ibj z<4dHMKZWyJZ2{FZ%|>8cGXbP863gpvH0>ivzL-fx(kfb0Mzw3L<*}|Dm2;! zy2FN@0u-nAvoJR&bl!lVV?7D4I8L?z^ZByQ+e_zyS3 zPqjq>WN!JcU$?WN#`zXsC-L>9ZZtQCn*IwuhUj>VmIV(2c~sdDPhmr|&x^6-fQ0^v z-&0UQ7oGL>ZyRBZg*TXW2*+hM-*X$2-)KN}RglgH5ew8XHTrGQ7?1nZ;Z7?5J~%V{ rZ~u`0qcb)Cxr7P7{C|3E>D$o7S&CSTX$PysFxt$RGyRH{^Y(uNom9Tv literal 69915 zcmeEug;$i}w>35rf&oa2h#)B~t%!6;OAAV;bYlP_DIg#qAky8LS>ixD_< zet$mgpwn>Q-w6o7&CJA&2|E9tF#itIvx^dOy3c0sGJik!pid}J+BVtmu{$Fo-0Hag zHg>r(-ioiyCL(HmezJKbmfO@@s3!8`cp~>aQ>k_y+yQ(H#w(~_`OiyPsq;ixj#b* ziP+iM>HPhh=|8`=#(C9!t!jP;FGDe*+J08Hq}0|tx^9a^Ew7>G&2)=DA!R6)fc@$S zZn5~Z8rNNwR+`Itt(;FH&99MDFSTLmoI8m}WgZlN}UXu-e zgp_<^rRJ*YA3qpjHX4ci{lrdoA|+F@FfU)ywY9Zz)NY7NOaCr5?qJuePK`m})7d*X z_#CfR440T*xp=YjDZF&Z*5Q(5COZG=&C zVB<$3S`qhB=hczu&{bvCGM}rq<8@w#TgE84eko#{$Lh&fXmGOoX4)cCe*DN5#KSP~ z@RVj|X8zUFDYwd+dHM1x5smQeM|R&o4cBy+St9+fQMWS@;7@YU{paQ$)=P#^5R!9A z=tsoHl9Q9W?X3*Ie}CT1OJGMD4MRS2EslP}xzNj~1&UHk?0 zSa*iYmZ$?cTpW|LJT-?*SBNKJ(YPJm+S;o0?UAyr?_Mai(1EQb?{!K_kFA-uh`scT z3|e8=jjW8s?9R_vH}|t0)hu8pcsbn?vK0e26whz#y*F&So6W$$ppYc$1-G%)P*8Ui zcoFT0VYrA>LjUf{+xF@C`Tdjqaem7oRtAP=nqR+sxkXMs{Qb!XBccb)%z+^x3W|!( zZ{B37=S0`n*N5+|k5xr}{`}ozacvD~?rUyNo0>27&CShiYHEV7 zZ{NPHc3Rfh8ulZGdl|AQtBw&B6}@)t+E&pJyup|FlkyZdA|m3(&70IzR22L+D-#nh z-Cxcfnc0ljC4BjEkHdVpI4vedvncBHV45gRCCft#A{~~7R7W*WqqV6iH9h_4_!ya` zDwX{F_NQbj7T>5WujwBtDZAC<;~wiGjVpS%%J3r!@q7%7jI?6jx|*(7%ppAEG(KMT z;8JR8>X$EHe*E~MU1o9V@?{G0GUZ<$vm^)o4e?+q zfj;RIG6PTL0xYc$CRuZSXR*K0pD+h93fhMdH8z&$zLb@xQ(?hF(HMQYTj0~i;2@p#57g0%VQ@9LK9S@AFvw{nNcxfW?uLYK6tG+ianP zl-68sb#z6uZZCveH`DVJdm_0+spx4p*BKkZ3~^~Ukg4q65s%DNezh_2%d{tvZSV6I zs${mtxYNm1rt=`CP~OGG(Q#+gwjTB489rj@{Q2|xJ}0|DWE>PUTWyRcA79dOy*)&Q5toMn(i~kQY5ApVeng2Rl1yB_-r)S65fUqc=Y`7jafr zR~hBMevXYbcXu^0AAi(H3%*3=|{ zxX|I=W5HtA@paaY9@gSdQ;klVN>?0jM{jTM?@?u@TRmc@o*Ej_n4iV>H=>Wl&p*}* z|9Ro@v#8zry#92B{+xp4epTiu5sy6{te9MaMxHha2}zp{pZsM=_(rd`D=tYnp7It< zRCPxD^f^7*->S=vsI01@hNMZg8q;>XUe96sr>940Y5W=(=iuwtkVEyuWO!n~fpd(+1+IYin1C zXb32HG6(kWnxkvobalV2{pDiVAU!#I;G$qE9JO*5&HRO^1;Hkhp2QP~mDs=Ckmc%6 z_H^h;)R+fL&HH7w587+znLid579#aM6w*}M@U5+_31ddO+uF!^%>KX~c2U0lrjdiH zbt|cT4PmLmk*UvI@_cXtBRC zoT4d}rvU-|T%XIN^EPTPj@NP~%#YFqN~D0p+*g{!v{4rq7i9dCJ0c>G?$_Wnz9_R8 z^us5U)s`r{XHn)3wa~wRglxGziiO*x^P$SG0}-8HI_cV|Hw_E03i1hp>gz3>Q*&MM zDUu<5i2cdNk01TdK9){|_Xkv!WMGp0d=e^o}aQC@W~VZ?4wr(2M~XFc+pMLtOs^-xXDhwj{WdBFbP ze-NCG3VI%FjMWE)g^4*W4Fr>O6U@^j@t8*pbnh-BXJ^aH%axUt-RI+kT=Hub;Rw!p z1a;AzU@#g5`e$+ZOiF6J@(n#BV=L=?_RI4H`%Hl19#XYfeWhsQrro~35qV!BqJs3+ zt$X96nacEsXn+n2w@{nTtAyMfd0YdnMV6XU${cXDaTLRNv2GJdf?LvFOFsOaQ) zAGwMLg`I$4Ag4ehXO;snb8y9Y^3#vgtzpY297E#s%#l!p+YGMJ2)`)sc#g|y*x;Kc z7lW>HkkZ`h;HckOLyZO-N`H}^h1AE&;f6h&-_>W=t$e<}^(Xo1{`xpSm8)(pTw*aV zco2^1b$ZF)JxsJYx>sqtD&I`c;M+iop%V3UUHFq+eFG7G22kJI2+`9&EhvLRNk}cI zhJEey^5w)g0f(8`u=x1+PeoeQUUa@8AzeyhC#3+N;K;VeaI&;P(2#QIV-_=#6ytcz z54UEwCW#>4>)p1~YquaWc0@U7X_L(R(@*BbPJ5$n`-uKX{kNP^V{c}Ul5bX*WaXVLs=CZ`y)R8JDCBj6@6|1fJnb?r5uEP0pYr`;uW1m5Y%x=)a`>^l9%?k-#T24o6 z^{!sMy7k}>AO?n{dq8<8_7dDZJnDfSjQgB=6>U#0a?`b?OzgiOg1QZ6km zUFmta1sj*ITV=o4m$tF7VMHWgH}UfnyI;RK-73ccpcG(L0l&|?dw1@<1iYT^tJe3g z7F46&xCdhxSfkOe?!zjWA@lW$crAx=5C6UjYon!osLR$@*^UuE|MtOIy25jHclX}- zbp@))XBQ_~6;)NhR|tps?ty^;yu@egZPrgsgf(Mdzka>YRYFZdV%+`?2&#a9z#MgA zTx4X4$G(kQTcKei!Id;3sPl<=3Bqn%bacu32=^9BevD^G%<_w)6mlWwT;X?0ii)BR zbMjD4e19Fd9b&d4;0Y7Od_>{<|3(++0R*6b8}ygu%=j{+JqBW8+A2S71JZlb@xf3^mL!7$&Jl-aksWEaQyJpm+)qJ#&X6)_!X6tG|_1oz;fn{xtScBwBdNH$H-ap7}*q4&v=KTF@0 zOPBoo{JeI0rQpyY^qCnl)R@yqs<7Khr>rjufA0K+FnM5s%ol+I1Eu{GIKGnbri8b5 zosFihuC9p*5V;KEyZro`-fCx=+Z7$Lc6NT;BkzCN-LM~jGmDio`-Uc5o}4sHwTJUQ z!{K0W-|k+$wBdIVhZ>ZshxKTL$Ff48N#MYM8lp7MZF_FLcDoxm+|oc6eUv9KOp{NA zTs_9W?eud%0-!iGs0>BpS&(S1JT_7V&0_e*RnIgp9$2$}xd}2hZO5aXMZms#lMB5m zL8}ybS|xLVBrK2toB=kUft=wIDcgrtV-0@zLQYFNJN-)WH1V?|VF3Z3i3;o-9j%6P zW5bUC2*>kUMx)!&6*i~{NtOjKbCxaQ$~Ppn#sp$b?N!t5`!R)T?gDPFU#Et*@2{aE z>v=X8NkpQt4Hw=no&|dK!AzvmjJXa74Un97+-Ho8jB?j>{VfKw)1EWU&_~9wYN@1W zC3i7AcpzgdXxX0th#1!5*RNl5gjs!}ae(N1z6q$NsmO3Z0{7lu)1%++p)4*dTLm!% zC~-8q9t8ZlV3&{HlhF4(PLsvrDx z?}^2;)MP2$m1?D=q+$l74GohqX`8*A{>-i&z_BdsTVtQW!_RU|PftT3ly%4I)-`<~ z7T;%iA@Je|p9m+t@Bso3=evq-t*bJ2YuGf27h_O|UQy6^!~PU>B{rS%FEO{*pO$=s zS8A^vH(}*FZwIILiBbRuMclH1wc5H^Clh&}p^MOjiHT`p9C8eZ2yxNTdmxo88J9~| zv#?a10LlApKuSWgI$l41KcMSebTz=ZwDxL_wLDtB)q>aaJMl3x+KwOmkNXFnUbmNvQL zBpv_@EYz_O!bT?kNJyxDqX8cf*ICmQr09+4XB*U>810WT^ps z)G4t0`SzToHXlZyr_VqdhpB{FyY3neR)B!klUl(3wZO>YKvraJI4A;sA3rYdfgGk+ zK&0@c{F)~uK~|T1y!;Eaq@?LFjna)(vrgXFL({|aCgK4JLN0MMkF$1KhVagE+bhgx zmvJx6_0NTc58W=^6g!Xj_Pp+zCnOX>5Wf70G?&Nfyhb-ZpW&EgR}T+}PQ3y}cX#*J zQ0m#i%LKPy8o83#jz0nj2Pz!GWod0q)az&mSWZKOxLz5ES(orh4~~wO)Ds09(rceI zhWFXlZu)EI4V9HRE~+5#pnB%5+5KOoIKSRk`(|Aut>?x&bQJ&_*e?Y=znPiWjfZ5A zIdhSrXu*v$29on~k>eiD7m^+*%1*u6>NzE*JxluOX=x&Re_uiB@~x=Y2aI&Ks%mPf zF49#WeKhB)0!eP%*jg#+IpZy|(HsWe0K@>N@!6t<+1OTrurn%t`%K%H{PfDUsD{NO zFBBxh5Lzy)zYh`YLlEb`k*{)~#+kt(OLUuyg|th)W-(reK$|(VIrNAy`uGd;@|HnO zKRMdPzj}3ZYpcDZBPbx?k>kR1XXkQEW{kdjThnJqw&KYzYMia12F&^m4-bQI0_WK$ z|2dPVrzaIPHOK@61tr7{z<$I|55sC_;A;=CiW%Bpz%R;eP$PLd{2-cu5U~`J4$xs1 zlAuDcMn0%QPLX z)oSD@$jNQ*@AK1qP=vIcl9JLf;GMNIGMj-AO3;Hz;4(=XoPW$I|C5eqUk&dBI zv(pbsjjK${Na~E4U0t)7ORnhW3?>S@so2)OeFI#J+iC{nA=dIOpgC;nncpBbzJ^LU z;(V`mg__H#lxUSI4oL+x2nLQ`i!yvN_T0I7=1(y(NWfD&#S2DdR>NBry^2&*nB^YQUL^(P3Ak&ux1Cgy_+3RpN##kF&3#m{llxh)h*h$y!q&@|~)&-*JSrF+Wq2!;N#KarT zWCpsGcE4Q9+HDQE-b^(C@RX0`N*f?3gl|gn+8aI-uhJR+fI0@v(0pe4rJBS#z=vsUJ+cY$FfOsM2{Qdg^C=c-H5GnzcqtjB!6aedA zJQM8bnt-n9!f~T(0)zlqFpb3)dKUo8*&Oq&N6Xt<-_=xCub~k-`XvK+_I>n~AP&jO z$}%%Cnf>|x1k#OmsabEm&uPR_Yx_W5GMrRVkfHbc+S(S820pxKGR6l?Ot-mS2Y|{E zHBew+uz$Pi6YVcRksA5BSssg^7(e=gvZoD6gacb{JFc&PRtc86vh`R)Bpm?%K=J}< ziJOg$jh`Rc-QA5siGWTEhwtgfYixXZTq3nwWd<66-~bTRxo%DY*yrNrhPKb@>Z*va zFeHsL09o`(sxe7U_Q#+FhhDYg#LssC8#B_=Ey_mVQ7bHwHUM**6`Ag|JN3x6zv=k8 zxnUH$$ziBf-?+HA2;GWkw4$V>Kd{cQFeS+FR|sxDulGJtQ6eKJKSa5VpLLQrIeVL% z8DL$`^rCvCS~uj2V36o>Cc@J-auU9MQ)7wH4YHB>h#l=G?c(Wh!;ov;KQImZQ`kiX!5GL58Pt8{OxMr)q{M)`pwW@3nxUwE1zN$f|G7H>sME~M%s0dj4h}9} zx+H=T)p7e|Z67 z44?~>eeL4e&a+Y!1=OOKg}u42Ug!07@q)}I{;;iSd^Y%_0(80Y6VMG)}d zR6_Yvce)ex@gkO1I!g_Y&Yyt150sbePzXsifS$EsV%P&vD*@~mEWN<_DGUjU!Iz1d zIUBA|0@v37)oa{+IqyH$M*vOrftseMsHmDQpY`u28lYv{(2y+{*xqu_`Rp$Lmze{c z)v3=O>SePEZK@*Ht9Wr&itn~S^`sf1p~F0eOjrAk^_A*N?CZ5-R+~||vWnVmCKv(w z`~2zoa;QO+M!bp`otWVgt*YF}^5Vkc^h|%9^D*O+*rd(6P2IzDs%Lo!2WM6@lU+E$ zeq`q;p;m~T)TXXR3;BETOTX&V^vl(k;~48Aw02S-TzcS>j(Ol~rjwIbpHp8I$Cksg zsA*%2#LtW_1(v5B2yCMt90jeFBsXs@+BUciOMhtxP+8-OGscarC^BQ;$FXt5KvM@emDQ2^XFXhbHWi=;0My=Fn_7`Bc6)je8 ziC@iA=kr(QR3^@Y@3b6QftO4y zZdF-AH~U0ZUCLML(phbQJ5G_D%^93Rxrm5i{-hK>=QNhh*S`dJUf?q*?E0sz@#%IW zk-0)#X7;^q3rm+yn>5J>zOI5B=2AMXhEEzb^b8n$|B{zDDAov+^H4d zCAbNtFr}2Mu97CQC2sl1V*l=)Hk|HTb{51wi|y;+MIx;bHR`xU`K^MsjJ9-5IgAEf z?CmzQgBGw3{;T8dwY$F#iF}D~#woZKV4&%zj!_U0QLRmW z7pr>vtgHa#E%D(xY+M#kHZ8sPH|*b?EkEt0S!PwsgVwXcjPuwJTo84{XAeroW=|&m z+rjmL2XJk8z!EnH?d73%=AhzgybyvYKFLdAH^h0i&4!v%WLz~|WuLTX&8Q-lcCtko zyEJEe5D#5CP;AID;qy8 z!XazE55IraDzHbw{)<50+^EFEXIQp&6^)C`R+o;KZE$mnv}uq1Ic~JFn7nm5!p4*UX>>q4tQ? z?r&d?oFDNheTB~5cO2s~_=A2NR9=&qlkL8a5yi3(sQ1%bQ#ZD54ruQljV@1bPuDLU zMu{^e=AOPjq1au2jXD&a@}O(te$rE#H7JQYxj3N}Ug+>e;3*M-0<-iD@xM$~?@{j& z3ANj5Zps46!|f}c>e%9(j#?^&z759yS+~l4YD*W#!Z%EPxAb}{^-~|%!)vL`H2r*w zgBE2!*UM+H!31Q3!zAxG^H=nwZ1x56*}Rm|yMV(N!Nf#{O|^xO4iX?;T=! zFA-k*rr*?EW#RFeTw2aayH_5h6r=i7DdD#L-BO>0I>voHuf3%op+Q!+_{0MUCn=fj z1B;7GiY%vkC2*4mnAVNQSq%Q?t{2Xt$2FCKVVIOOTEFWSs+y#3Uw~(qKcv(m)EiuU#M2(e6$VQ zMGO&%JXFIin%wixz$9_mmI|7x_TbU2Sgj20>#ol(q_I>~6!}zbcC_9Ir4AIk6;2YK z)0sEHamjdUdWD|L7c?#6n&GE;iMCYgerFE3iC(9Px6qz;kJ2w#xl6z5ODrJ3H)lQG z{LEv|%DyXZknO7H!KMnf=ZAMed= z_bY#__dM3kim>Km5Qj8U$_V{DXP#0@o$lX~J6^h+{Z#x*Rh5ZHu1XHc{`%<5yqNr? z#n3=yjb#X}%KZ?E2bKd5O+T2t&M3^$Ny!h#%MpMdsw*9bMGo+42zpvh8 z$tlfF=oRuKhD>v@F45+{dnd>uF%;r!Ra^~1@%#g4XI1?_^;5aITfp|Um+OLmam$sI!tPKnQVJl4bXyntk zjs_bScIZeJSBZV~iMF`+dr)(RHRA%uZzO9p@`o}@oUQL$JkDYxDgnPaY6lYCZh$1?{(Ql5h%)Rc78r?s$=of>mWkh5nXeGVP0is6q}_4q!aY*ZP9DqcU6dBQp3|Ay z{{=LD-1Q3!Gg%%g{rq{V#HFqB&7oZMYr#im_;}20r54Ph9N3;2F+bwa@=&EApFZg> zs4O+d&N1Z`QZ@w*k)c)(i({p@>6*G15_HyVdWBxM4KnnlOoSo5#tNTANv3J`{VaP& ze3y?OjwPpCm39txp-<|K+Z+E79!E$C8gUcl+6ZO>X)7l*60o{g`^0+#1(;6ZMU6mC%9=DN*yoeY*qK{LlGYA zquh4sZUM4>APwn%>6%ciqY7>jy?g5wYq8Iafl+QMlzv^cd)tz<)Oo!YCW*0bJey|g zf(XiLe)&!%f)Ii;$bmaKy}}KkO^?1>sw|9wW?8SILM0^^sxP`lAGf!;qT}#*-jDcH zaJK-E>(RjS^y}qY(bvoAJR4)ba5wu8lqf1u`h`g5XxX_vDd+Ndd~z=FT4Jw2FOn-i zucnGU?N0NrmM3wqY+=6uE<2Vwwu)e<8jH_?Mb@7*H>jMtC+nFemMRP)K>)m0ak zZH>>Cf3h-^$bFCzfFV6jMmoO4+3eYDk91L@>`Eu|?R_m~?4#C0#PH5sg%lLk(AIx5 z9h~>+fT;0ITEY1qp(|H^pP>(a=mc{Vr9F;9>+!QA#!?o#occ85YKzy~ye?oaKN=BS z!b1RNvD%qhf9exs3Svw(yOWsJesW}9q~US5bNA@6ZPI{ML_Pk$5~hto)&B&fw*mE60Mz6y>owogtl{N|pZ})Ew1})^xa_ zI;vvyztT)lOia(vN9oGcL=?31pY#Fxs&qV6k?|0r5~)q71;D3Zo}HT<*vGD^rE}cs z!Q(OE_7BQ2iCv-mm|v%-kgMBTrKlzK=y-5rJ7-dJuaebO#Z+h>*ps|@YZp2%O|v6n zkJ(pTFdUt7PS3OYwt4>`x-69sASry#GUQp7@i8b;GwLw*i*>0)Jy25~nPs-LJdiT_ zcdo(Y{r|$#|37?tYHLbxeZ|WC1dcb_`Xpa}(#54N-pMAqbaipU&8^AUgVLkK@4%(1 zeayC@e)xJBy&!+}bG=NKf_W8Nv!8#I38i?q%}iX?K2lTP*uJTQv_QB_=;L3~usPY= ztBG=B`OUj9(4Ubl7|r#$!DeoF<@W@A`apfWR|T>F5sbX*NJzCJ=BlzSgH}Y(1zaDz zCUQ3W@R$TYgkaFxl+>6;;gLo1%@YocPlAgxcBegKQfqPx?6l#ZlQw4aIIgTwRxTYe zhiJS?#>p-z$ujjg9jVzbKz#UxuP5GbHjbYQ1S3^ zs^L;EqNZ>%gAPzoz~-B+Sd_DK)*vuSB+IAnkUG0lXt2`V(oQB3w+b5(EyO-9kvbU6 zE0{C*#$nBGXZi51nZ#Sm$84?ISwt$pKK5GH5V3BmkQb_o?;##~>DRcK5Pgs9(43We zErm-5^Vr+G&0*82yozkWTL389XV)gMx*z7--%OPMw(e^Q3NaY-bSY6u5x(Wa zBo#YtvTo9{VdU_N%s`*jy2p}#y)3<8mQIPZcd^G7t7cxt4Z@J zd4SJD_lb@P@EbvgConLuy{Bh?Zx7trCvVSP0DUc6FyG=y%Ju&At9cr(t^v8Mo4>A1 zDyIO}^>-~KQYbds&)wR{t6xGVPMt>8Xn_o}T%#KP7P4p+C_!k9+i-zI>FZW*> zm>Tqt*U07swuQ7Yb=s9VAnqr#=Im)7yK z?&oh`l4i`7Z!b_RSj}6>;y@(@fC25+0{^6#l;-}AhpY2R@`B`%ln(sIYnx5eUd!OT zigV+T%gX!3Dm_6v0!}=6`CwwBmSFO|y(0(j`}gldLxq=zN3CG2tIHbvq#!yLxy>XU zq6?qsR^C}cl6mX}C?+)Vj|Y6Z`RwC0b!t*Lpu6eub!$}<^S;DqWoJ}?EE5kaeMnz* zr%K*D`wh_mQ7U1XhIMKjBIG1>E)tk|YMk?F5{JA1F9S2btS#vgQMZpB->#KUJ=U^1 z&uzX_sOFXPva`U5peW_e3TAIxZ$%ex(66I_?UZinW9JWfb6d6y?e%hZ7x_j`_Y3Fr z1vX{f4c&VLMJjgKOsgwM*3v+KOc@ zd{jbpZLX-JdXGwrynJiwB&cuYGG6k+1Zr|s!*+jT)le2Q3&3bcyB{f9yOB&hmEj>q zet*NHmNmWNa*69ezhc6fVHM^Z&Rozzfq^^D34%`9V36@2t|7k5$Y=^3B2bkTWM#9z zJj}`ZM#X6g5{u_imf4`JeC%Bw3Qn);KNdNHTxantpQbJ&Q}7|A5ooG7E}2jacJ~OM zjz>FW=Uzz_kDcWm>s&;gMocH!^2yM8!BVw_I#L}VZF*_$0F!~hW}vXgzXR%1dpun^ zJDNC~?$d!L-Ckg1D@3w)y1@gW>9>=C|jbPGTGJ%C#HR(DD{TU%bUlrTps#|y^#&nd}K0{359<(f8rplbOL-+e+Z zJShpL5Ds#4S?PR(f;ub<5JND}ahZ_f-o1OietzIi6or{3aHqmN4lM8Z>IzNo^5JId zT~1}kRc6UBQt<#Hv0GwTCdGssVCs|%$zGa(gq58`zbOUi+m~4BKxA@qf{x@G5d}L# zBPeYOR(!p>^X|a1K|%2)ChKPxt()I`z?M#sEtUzoUtonN*Ji4QnWeyB+~(h4_BuUK zZR=YwN8|CQ>Xw71D&0F{C~m~VlcKmV1o@VUbl2x5OcO8v&^KveI^P%G)`yU^OaHuB z9*^+Wxn3@gYnV|;+51u=soD!b2}n>-5YRSk_Vnl^zFSUw^?wzIw9?OHCy*l=bv1ouqC5wr7_>4& z!T6Co=+}MD{NpxI0nBH57bIIkyo!z+xEi3g*-ov~&rVOTr~>R!d7@(f z#2&s<&kLH=vl?3ynrsH#|CW9XlUVe3@A?vh&j~a6krs1uGV1|Op2I#l4lwzgcE)nU zpn~g(6L)U^OQ$t7Nsh)|$uZhyk$rshDIdWc|L1m47D%CJst{y-M*%gFsH!_X_NCk%QXh`JRg+DJ<*j z$kXN)j%$DU@7(8~lW!-_`@`(p;Ho-~#%i>qr7ndXrA{+eH+dAEPH4v02~JjA1tz7Z zScB$Vx!70;yhSIic)dc8!hz$2FQ=O4 zZCg5w_J&^K`TwCz8MLTBUK65|0B>G})$otAQMXzIg5P>%1}qr9FDh*GNPJ%qc7CP9 z!*4#={PC)_9*QG6@6oRb_e z5~B05Aq|oh{DO?t!s|^JLF|R>gw4l7hM6*1AIeeWppyvYn#X=PZW)Ejc(5{JDjtxY zs1W0*{R3&OKbp0dou8BXWNv&6E2)WX4&g{GB019?i(NiJwVBrwC9)@E-q+y{uR+w5 zhO2)`W8py>7fDz?yd<*GBJu25T52jToxa!ZAXr05bTXhBcb1$PlRVGtve0`&=3Dz4 zO5X^Pyh%c$9(p0|+-cPG@9RaDF?J3&@kudxV=a=N*s6tLRaSjgeQ&Lc7uB>SfT0nr zG&Xz-rHLxX_t<5&=ueM}v0Pm(neXf!$~G>Sd+y{cwH`g>=KSYtXXpTFzDj;=u({I3ybiQ#t%f4!#dyj%Jz??8^WoK zLy3*78jJi@3-HgHLDh%OLR3DF*g4n%zx?5Sj!Tz!-Ylmuq1WDjl8z5AzK!jEz001+ z%~wpcS4Kbfl3L(7`OQ_cMw=0)-+9M>pN&mn#0)&(;TCK(40 zM;~v1Wwsv1zUJ`KUob~~?uP#S4;#G`1ga(|oW8k7ENHl-d4F7ohZvePikF`HXYQ{k z&&-%5dOy;c^E9bQuugriPm@n6#?d9lfdsXIL#kk>VdwAI-|eh< zA$ONsCWGxc>MR8s(8SB;$<86yRANjLO|U;w>wB@HCFCmU~AI1olJ9$ zCVwyA-LE?17w8~7!l|hOqik#`52qFP5*jx@x=5@(-25E0PkT+Ku%d`#@%mNNdpDQj zwTso4fup9*aJ8NheKT5N!Xc`v&p)4~?%r%AYt#_M&ezj!@QFO~I&uWV z24GdNa-~0ivDtskX-U{qqY1PM=`wgvmQpsYFWA%;1~}~O>;$d_hlO#lvhsXW9A6pA zyNS&IT*JH?B z_-1p78#D}$dz-@uPJ43WD?B|0oJ$xt1fBVuL9Vvc%t3u2Rke?*I-cI+s3~XVbCbB^1th4WyUyyE;urb6w>6VT7h$^N5F){SZ{@I)n<&Yx z5)%+O`CjPm2 z+vw?^V9YeAb?t=)c<`^=?Z1h)=ke*9Op~s*-Z1&Ivroot8W<)<+Mf~RFz-vK$_|Ac zm(KSuO_`o-@1XYXdRskZ7BpY?Td76MRbL1bI+wXn55JsI#VUc0ED2k7VWsP-QyH{bF{eVjx3K>&GWY*6gqaSOK>}WiFu+r(#uvg4?JtCgrJ;{ zNQ(CM;*kdKGrgGV-m2)h^N;h?KbrdC)2&#qd_bMbwstycvG!8PD|d@s1kWW*qQpc? z?r(b>c!GnQ0^IHFfRn~xDhvj7!5T$Wm}vUFthlTLtY_em0}KAy3^h1q$b{k^=M2hT zK!F)9r{-4JdjSUt?YMcrYx3Z*ylg~p zSdc_#9vU*j>iUv~Eugjm=1~ssKOx?1T-h%1Bc`o5JxFWeAkSK~n#XuA?;AWidQUuM zGS?OOZD3`l$38|yRqxT(#_!BACOU~%VJ)vHL#%ajtQ~shs*BdP*}J)m9-y|oKVG5D zhX%@nj{|S=>38#6XIgG&MQeODF(R0?o5WsP%=q{@j;>m2zspst6HOtn)_u_%?0_nJ z>|_)a?z8PAbP`}!Yz-!N*jeZW*H&73y7$py23X#*%u=J8k;sb8Q_H?o8E{P3!Egvz zs=UD_0oJ0yoZMbBwA*|FEa6o;$>7_DgFX0P+CnI&Iyn6D@(wWDDi(r5LC=DkD&H6l zv2L!EG$6iQv4dE_9PrmH zx-+9FqoQZ!$Ryaye2HPe04B8|p`KDxPzbv3n9%_h^NdVPw5j{b#v&$m3eFN3%Lm_? z_iU7=WGGdZN|rA%t@pv?WTQ{K@f%1o-~vAePpmI-etv$5DP?@g4DDQAg|93F_QZDc zbbFti+a!3Q_vt;xpynVpT5<@TvJ4$6n->#>5ydX%1*LBuuD!ezi2%(jRclt9{;4WW z>g@uIPM>X`Bz_MmiGX>POi3N3V5EIZ6KQ^$K_nk^%G9ikV!tMt8L8@d)<{zdrMN_^ z%}3oV*clyNucGf+G+{SeLBGHa*ZQK>C;nDv-I|y4t$k&LGXoEo8m^(~EB^G}+j>>x z96`f6A3n85{xqm|d;qfvxJCA}?O^DvuBy69SE{`VPBUNP>Kk-TFwX2p4CC}Lgylah zULLF@dN>orQTG+*onWl*Y_^v%!{{>CB3sWM4Dx|lSrUGWX@y~o=r5A}n zRzXPt57G-ji$VP*mgboLga-5&7a}`lD z)_gpT0DReGmZ*T{PcaO^|Qx5d?wPNtYon-qDHxSO)mnx5Z09cw!O>x3oJ?SzMi_qD(f zhNB;hF#YcAyd}P~j8xLh)G=6vaRZnWW%BtNjfBCH?gXJ_eXtT%R#pa_y1+l-e?e)T znzMlOHw+$8Kt=|iercpNh;McaiR1QufA^xgl;Fxx7X6JIO*QL*2Hr)_&x9b235mX| z=I$FWD<-1#TO_%eeT_gT+^s<3)0Iov@OuRw8zE3zt4}vS_R*v!$O!bCDkW9uAVluH zlopBksP~*A?>--gv(8L8a4%5o#RE*=Cl-yZ>@%2al^Dz0$iK^eIhVw(>8O|N5qBgp z^~pRVWAUg$Zudo3YidZEW{k}O;#U{e@brgD3Wg6eN`ll_Fp*ViSK4y!Gq96Qv}f@( z0t7{WEeu-u2-lTnKWehjYcy;Mc?APZ^6I_NJx-Pla((mp1g_Y2UGm; zQ4gab41;bk{Q>UWvx!WhnrvOll!yc&#xw3UHTCSxrco@>c$MnRJU$s&Nh{dRUQ#>u zBSXwy<{s;D37b`Z`I#gu8nt!j89H8D!%1%mJFuQHe@zUO@x+Ywc%L^|OwhL8svc?; z6-8mJZ5SsUyD~ZY$~>cZXYgxVhk`zutU@VeaOD%q;r_0?)OsjzenWD@>sr^flL;LO zgS&!w85XH;JbVpZQh4|bRYjMP-YjUKC`t7!d|S}bssMg%@ptexNW8Wy`gBc{KDC_R zv2U~JR8cD^$Bw)_0v$J@<#LsjZ==pO))^xSUV`<17n&8W$3s~L9YV+6SmtKfeT%#! z#UW~Q)FbF<#o1U=RPjf&KCD2E1(}m^<8g!5`H_}l4XDE(#3Zzt`zGE?u7v=J1kcB@ zy>8k@?#5b8b%)#+ZR3#-JSg8o@c_tV<$I_lkkG0+RO*J>H@)dJ6Fa#dK75Xw+-nAt zY+jx;*y{q$D_@E4K6>;*G(!0PeM$GXV@M>D;FktBJ}GPF5^?rpWd2nBDM@VXtZ?m{ zSH&)8-Sl?#Ihp{WJJw=1cjU8nM+Yo!PEZg>Ql<=LgOuutSX?F%@Zb(JX-yKW+ga-C z+UJvVfPR3(Zeo7Hs+x!WcH7N3#mtD^QRq7w{4xCXUJidt{cxXlY@F;E`QZZe)+jw} zSLzGuA6%`C%|@hyiP53$J;OA_Q({(Q4Z^V0ak zCf?WHVTT{QayWhyI$zZ%L#+8-;Aom}36d@mg`{OnUnm*;f44%y0uKu2!04(;`<30Z`t_EVZk3fKrHzAh$JkCXA%zxFg$ zG4jjk*G0{KO7+Zu0M`*W_C>^N{ia%!rZa83=oh`X1cStNqt(9dtc0W#gU@o$zi6uE z^&(7j60NC5E>>M6zvjHTV3bwlSd?X)L?;)LwpFB7!;!a9xcgi`ud4W{?K9)YL&9Ux z*<*)i^D2R=-6RS9>WgceRi;mRF2N-v+@7gxs2@E#xSiOCN~x&{@u2+3=l|gs`>*oz3Q`VX>9;aon6Y9KFdvVut&SeGxKEyr?I)GVSZ?RfmpNF>zn;B!myeWh`EtQA zGS(-7{gGZii>>vzHti$cA8ve@;WtmFy}J%}pYJ7qE2ffSX?;BZ({6v`%81B!9^Bll zNSzFYr~Io`#q$+%^?J3=$~pbzPG&*_whKS`F0k=NuTqch@2-uavyzXkM|Fl-Xtncn zt#uJAweD`a6jMSH^}BisEhjJ3YAGpXURAETT~Jq3^Y!oBx`CxM^`|uVA&;@KR7-X`#+TZ zcRZK<|2~czbtzhsT$!aPD|qL5{-#?wFS%0gfW8` z0*=|tgX`cyq&Kn0t%+uLGP3EqP4g8M@;zm>nX`_5?5ABxL}5O4;*?S@lg2zeJeY!5 zjK>P#Wpt6Q6Ih(gp~Hv!PzC>(w*F?ay3fG87utQcbh}tE5g(1e0y71jN{a)$7CI~GO`uD zRl6DFyw`8);KlqqtO=UNaX#mCpC%5iUv5@R?~dg*4Ay>o>Y@j#>LwSfE;+quH~qYD&-|t5beYM~yC!Je$dmrbxLnZP(>lFfh-xilnQbNVgK@bc zO?2$nb27=eyJIZ|3B3O7cr)>_1+{)%iw`r&PT3)|3K| z{LqSZr(Ya@3SY|M`^wMSO$uT|jbhFB(0<{r@~!fv586`gt&~nx`z5M-z>_0A(+dXE1x8DsF2rf^m#r9J%_4T$Gcw?f(#9N8zgr~-aFV6+LLdR zPp?kT{-WMgN_?j4%|pwM9g7XE?S&KJV(K<#gC&XVllmQVTgH{*Gsq?R`NuGuvI0Co z`I)%3t}ZV-W~K}c4M9?aX^}H|cBXA7LMf zYB9)j{YR$ZWr}$L*_d(9o+TTLRDSh-?j1eMI_~xp8Er=FnS!1J9+qP?4%;nX$Y0@R z(G#;`LE3yJR4uAPD(9%l&$VYO2C2$!Z^nm>N+sU(+~S+nnC0Z!oh!d$_9QbCY$0Kr zH*eSjIO-5(H*Pl<9vV9J z`AkizVr;qS^Gcsmg^M9+e~fl$_6LmGa!?lO+sRdTtr$H0SIN zf~N@k$S>recjHJX@yr4kg$7@S>tmSdqO7=tgd0ZsbFB@m$Czl2QPa>|uZoz#{Jy$| z24lRIpzuh``Jsm3NFGzuS4R;rmf2HR(&S#ZV4!6%4_{gt_4=?zYqEU*5u2iVaI*i< zKfj))V|Hqm~Swq^SZyR9o?V?|jem3wJz?d_jp z))jeIfA`dz{oL34EN3_)!C%k%Qpn!G|`oQX*xmk;;sk3AIIuvIfhGuNP{ zMC#^RXfThXhU2R8YH_ivT&{mw?0_EPhELgSA%4L>(q3tsrYuLToY6bOw$=HA{LLS2 zw&G*+X_bZF_B{QXExOI=Jw6|L8}M#ER*JFcLizw(ie&HDhSt17x~%u^sCf@ z>96?1Qx`-pi@3GA-Osxt&N`h}lA)_%+Hfy>JN)n7P=*DO#k=|U{kJolo12=?TP=E! zYsjRIJ53sdXO7}ZZl?8}(h+&{#+s#f@5vV{Tj&FCCeODe#hK0Z{4CX5}7r?$knawXj~3+C|UqZ=+OO z+1w+SQ+?dq){OYY(l2`m*C{D^H=5-rgz?hSVrY>?%yD{DulUm|3_^l44%``Nnp%Tv zEd_lR!KxHW&oXxT#rkQlkyFwGnWpyx_vp!DyK(2vZuZ@Sj4d@e-OcZ^v|nm33C#T* zx|x1HiJO~WWk=D7*?{4!UWJoiz|ji6N+IUubCT3H)Xr0B!c^J&E>1lP-O3{+I&&%L zO+w53Bwc04@Jw&ao7+kL2Hg46?q8?yZT0TS<#G&b2Z!1C?Grs2#*N#*+-{~~No5g! zW%+dP38jTBpZ3EKqKg=2+TXA4T(j?r*QUBgZ|=7?JveZoVTTseFUA*aTu1mXGQR0E zs5baJ*T(!~^s~PW_=mzj@<%bAba2dV-fmr&{_Ca!x1y`$tjkM|TkR29=8JDem)`zl z6C5uXd`QEXd(zRQ}Wa~xtU|5%@TD&YWaT8T!m)ROV?&vr#dvmdRFetg@53` zRO@#6{u_^?CHuMIG>c1No!2Z`d0+?{Fu$pI2}%0zr@BCC91$R~E|p zDwif_Cvr>AExu@|2%!#n^+~aDFuUEc{eA0pwqY?Xks;TfnLeXvud&~5YfW>3S1oPj z8Utb}WY4 zJ%Y8l)GCLR6uyLE?r4Slt;b&9EWrp3sMG)=CgL$O^-fdPmfn zvW!k^?rL<*+HrTTjQzId6ZGuA%(_L>KNTU&eMIJ7Jh`~EbnYI%4RrmuZXac6V#vXc z6V&Y(pul^sNZ-X{VWLO5L78VU#W`W+OvQ3jN>hf|M>B_(gZXkQeyU=poDyd5Rs7Xk zV;LJhqupyXc98vS%Zor|k8Asnb;c>ZSVkdoNhv7<89_!R42%vvYsLsF4Gm3{GTd(b zD^WqOkygL(aO3NhcWG)%iW8Gz$?MhBnns$QCt9|~aMSn=&HdCgvG$UyE*%k&7m&4P z)BChZ!M`WB#;c$*iT6^2Pj8L46f4(oXWR2gx!i7-Pw8gqR>gTUhHLlzFUP-;UpiyP zH#e)mEq^pzSMooP{hTLcW!#57O8i$Fnp>J}zWtlGIaEC~jIolR$}`bGU_bktus2x+ zRg;-h5l+s}G~zv0`iBg5EUUN-T%0}7^KJHF+KK4_J(;H|QDHOsvp?=e%k#*0-|XI= zZg*@c#z#wAE4Ac$&i&N;o^37fi0sTsza^i~s5SrHO?aHBlqj=RcB9)Lzl};>UW?5w zo_jialXF#a;$!;vo3|v3{mwmds$1K%5@;7VXVKLhL9Q>uYBAS&ctxs)Kh!D1@yQPl zWt?h$Hdk$#cG1~}kI$LRpVfb^wzTeTj|zA6^@)z{=v@+aYjGKrx^urp>i5|1weJo0 zx$Ws4403~tMv6HMWMtNM6$&o4CVtEtYj~Hqb==Czid{urJrrz`tgQNV_uotDZ zb#l_wWcw;z{R~7Gq!79@p!>Dq?bhi?tF!vzZjKYt2E{*mJ3L>~zq}W?)DYLUt3fQn zZQ6apg?6B|AU!%Zx=SH&Uh0L#!MQVoJ7sr{7Fui)YEsd1E2Wl6;%h2jN^({1o7bx@{nd>Qjma}c zO!UXhe8zU|yu9k|o65j*yr1 z9$aKcBK?%ebD(!UL#RE=)m z?iCBKKC^vKgwN#T+sg|froVGl{cSr^@2b*u&Il8PFgi}&9R&{NIjY|&&dRoL0rSKY zA~hMm-Z)mAYut89$<~%Hy}3D5a=w#Aa&0!NRx8KWLiN)oKPllSDpsYH(T&wStg~Lu z6E)@PDmhcN-r98_NCF#?|M(DH!p zu)gPl39YSz1Nu3S{ya&Mi2e+Oj3ZTSvz^7Ix?bA*sL3F; zp|!|?#(ry2v%=eNZ!7-E;dHN0m~R6J#kGBqB8t-bP3V=wjXJBdhPsPYTg+CN?)zHD z@A&AE8+KmF8=&nI^tu$eOvx)KV7B~}(Z`2q(atYeo$z3wr%wgdR=F=d2yWR0Fkr+K zS+E*^*F}|Lx$FI_G)$g$&bGLiUT>cfr%g;rdivV@G|LIEb1QH2$lFznnsaBABGV*T z=jSCP=Z=^UOP!jzW$4`~wMEZTGfO8_BWy5G^Jmsz)`x2NVBXrj2!C=-ELy`dlw_Dw znW^6Fn^)^sEiO*7HgGt$)G2jg^`=2@iCN)Wz+<_Y=Ao)#s%vAWhenvK3z`c{vr6q^ zvLg!j(MPwckEr*^d8p<_45toDjEE;&%3s&zD(9-W6?i+aE6=txO02KYYE*5s3cOoi z!777p%i@@r&U#G)kCmCm)*N$&zyChM&c($AKDC5|yo}ukMMYh~I3t_PQ1xHwn!lvu zz(DQjtF@DWDIPPeEH1dRe3NuXf-L&)(;5lJ@&5bql8U_N6iXwXISN&E49F^~pFBnv zr2X?+{gSkuQ=nAXuNJMQWyFRdrw_AymK?wQ(V8lA=b_(+)%v;9)OD9cR{E+2BIEBV zMKu~d^IXN(xlc7kR%Ar_UltCuiux&v)SW7S)697P#O#Ay&BJLYd--(_dxWJW#=Vdc zNoe>P!$=<;nOmGYX_T`vm~2|hLTy)R#~-S@JE&ncHFS1~y4zFCsm|$VTjI8HP=tU4 zLl@j!lRcH78`HRSDF~Fo!NFHgeR})03xpX@Ph8&ts=+a%b+Cd@CgY7(vre0e(ZYgv z^$4H&Saj)p(=VV32>Q>o+%xU^VRikkp_So-%@5dv-6S;GIwdTvvsK(;eo{V9d!D=a z4B7I^WpIQPl=?i2fcDtm`N5TENDSBbE`iobV zsI1@ITb{i*`!YGS_V&eXl23FJ=i9JP0LNukW~%zBTA%Dg5Ln3(&NOK?xNP5bxTnm< z;|}q-M~KJ$mosffZmz-WD{sMC4Wgj}lwrXX`?9M@Hg@aw?fwA)&rVzqNKa3n0G#t` z%t+WSE>QtKKacOEUXy+CWP#7L`A_SMt^Wu|&LV%PcrnDL;+0oLuJ1*a*SW~Hbl(Ay zD)ZB?VlG+xU5_kX)G!SLiARIg<4NAd2Ss!M1;Q8=RtDIwTAO$UB0ZUJ`|%^I&yPxf zPmo@?6BoIsQHwuc^WHffU2`hWp4dW53=yd~dM7W`uJSB}-X49HWgWv~DXLzCBn5Y^L!jJQ=_gMqqQj&RxUDF%%pP34DOelVv_v@{r zu+=$Edd{e?QZ8cLOH%6U>J_(Y<&|hhm{L>zTAQ!?w7B~FUZ#B3#}Q^>7r1tbsnSJd zyPG?!Jo+2&-?QEaz(Q+kkH>VqT+VQ&MbG}W*f5FI>5*pZ$n0cOCKcg2=28_`lac%! z_N$hL9tM#ut(81=9#)jC+x7%iBvsHK)~H*2V;z0t-LrsnHWxh?GRY#wk|6d!4aO`y^L{=!(d(x>VnJuSd+s%6#$qiZ~c8Y?<;ssUc-7P=hw zXx{Fk>y(_8R6QNZxYx=Atc@Pm&Z_IbStr5SgtZ zl2Twr*r;b~E@R9#=>IX#0=cblAm!imEp&*GMA46lBc zG5Azqbs=e~?b+b;quLW*dV>Y(50bZCs8JsP-tjY9du8glFZ^ruC-KtWVMf}F?^?WL zS#77M(ZN4jHddgmnssM2dTqJSZLC-rWHjo@8ueoNn(6xa`T5{QbnkzD zUrm>$_m(7>6k^UU9GJP!owz&bttFM%WwAzy1==+^M}uZQcBa~sUjrXo|NS!Jk!Rlm zjVAx!0aA3cUxoWh0iLyZn9PkE++P#=osLrshr4xQD4g@27J9G%(iD;5ppmP+ zD8+nATg*dfJb7y&r*Y^t4h9Z>pUW3VJ^-}5u;gXd6k2@$df}ShtKJKdJaNclNOlEpa)EGZfU<&HHjF8wb~#E>U3-eZ=V()$A@!hWC~136jv9 z9nQ~`TML#IoMP{#f5ebLGrq4;;87HMHk(^q46E-VNzh=@oEzCxxQXYVl|C9UX?I6cmGn< z0{|{fCl=)N%c4u0*7l8Kz^s3oQ;A=W-z0B#Tf6%7;>#tt5+s zrkIRUsp&$`9&~ zwOoD&$(^)Fcu9XhG3D~%Sh>pAhM6u|gcpjr3q=SpvQ3VUp8~gSqPb~Wg+q^Rvn}CH zI^Ji={R~Neq<@}Eo|abga^tbw*Y}FNKIUvQVR1XtUb;hF<)yfrwTWTqxyBn5IUbMa{j?s0mmPcBc?Q7SP?nv*`(yu%c&)of4;g9y-Lg{fXR%GQ}IS8*EQ*tlXly|c?#K$9E-ob z_udsF@U4Z+jMvBxttk&LH7&IbGF~x|>nL!G6Z%h9aC^Gywt|ZKx79V()iq;bX5(+B ztem&q%yb+atjTEzM~qDwMY&z>-B0rYe}^D%_)C;ZUP6y%uvtaf zHW_CvueG32H@Txd(F04k-HT34q44seb&M6z3LhWrxT@Y{XW_cK7_zbc{!`b#F|P!I z4iveIH8?sARwu(&UHemY&TE9}79H`-oxOk-C)|r6c{}*M zQghpD*XCQy)1ihTEDkcsT?v{pcPv^tZEXqPh}g!&Y&KzhVqyR^p6Ka(sK5OyQ!+s? z0(v%~>aq(@1WYO(amn&;z7|NfuI-Z~l=Oh$_isqirUTOus*aZ6CG?H`mWmzPrm9KaMuAK8 z`;7j<``ejb#W~StWGws`4FE+JDn0DMPoIikDf5Q-gsjHashFB`GIzEJcM-SN*?_1Q z0hsRj^ACK)CqB^OXn`h8z}*4Nez>HN!x#1jq4L3?@BDn??mO|Kd3#*YS~%5Y@S_DJdzatP)?qeRR{*8KP!{9%TIf zPHJkE9cVSxocsFt@!viB_wSc3zk4OjdE5GJsiDY9wVm;IXXk!)P(#&?9nw_Q*KbeJ z&XW1?@#8h*%}EM(o*YV73F-s7{5j@7=RR^Ovp;d#b2|{obkUeVZZY7vcVPnuCz zg2+IX^XAxQnr(Y_?##D4d(>J2 z#I-|`1k{or>9=ORs5uwNnrU?ZL;068JD(IV63cOevyS((odnqD*H1{U*A5hwKZmsi16-ssX}a z+C6*XEQF})e2IvQjLasPk*Sj&m;?GW7Dsx#3%JNvzaRMXknb;YOm7#w_VqDHEkU^` zJahN^!@p-q{A_|UKmMA>z#rO+dr?W<9}ZL^{e~bPbVWI zBj_DjpFGURmztfeCIgWSb%QO}xCDY_5#Zx8Xw?3ihxE#UGxy?Nz7*e<;%;+)GEV~IdO=#J<5 zP)qPD8LE8Vt-Mf0;N|^Vr%vW=i?w)Mu!#=4=!Ad($nN;r;mg3?2O@DEZtg+!;hC8Y zqA?sSw=!`>MeeXCAPjW@T{b9z{dI*RN@Moh-V9L)|2=79`eVJWeWjtTt?l32x2vPX zG_|yD4C;`P4FU~>E>vh}D0}9O8c;b4)7skF?n8?oI5UeuUd~b0+SHVyU%ZGKJ3)et zeu1^c#!|*yD?ukle;Jw2-=ge(`-soS5IAmGing{k2PY?pd!H2SB0ClVm4?OAwUzY5 z(g%}R%M%S68tA`eIezs6h~6Pba>cOb_vMD?E)O0)iW1{E z8UW)c9g>xuZSUkH>_AU;MG#5#tD|o>)!o5iu%OL&{raL+DGvl#WPald?wuew{AzAM zvH=8vjOv_UX^E5zM)y5#6M8f3FMbScV~bUxSVmH|?YgLl$im{H?`M!`GO2QW-Az9)a_yBwX%Zw)J%I!YD>uL z#!p~|A@|Ljy+Hsxa`N)#KQd`+p>4(YGn#K>qo07rJ~Whr{no8pvPbszo{bG+m5`AM zU?!Ez$!tCWVARvogMJi0I|N9&y1K?nmeK?yZ91iCwSgkQSDig1EbIhFP@#ao%5pJW zyQjXsEx${1i@hhNC(s!B@6pSG@1vzYp?TY)BbnRGCo< z|Csc5?;Vo!*&ZiQwlqHvvm^=U0pqv$wzIPn7p*Gtt-88#z@B|K7(QVFgi0!LcbNYzKu;-c{Mq>=$ zZ&;|dw`xb9Q$4)9amTx6sOG&^h71_}U%_cC1`!G>u{96D0?9>lMq%AWl6l`tkSgo@ z{oChrxgP_1ej-a)vrm!y#|@vw=v$E9D)xC476wi}kLCG^hK2?PexL?07L-&w1jhxi znz%uKu06d;Msl_eq0{zXzCLP5RU#Td0Bj}>?i`0cf}Nuyw$>{6zR^ESw=$3C1nn+H z=O<`EI;!81F6jm>wrA2rV zw_IkC(%hluv9HqA>xe27rXaKS)QG}xWt%G>an z^rFcMc#~v_e$Q6IalkNZdbs}c&n;R%Q!^GZ->fXmK{q%5>M@9R#2{h ziSUA^5}Lvaue|-2pmE|ifL?5Po&6(8N4MF} zK?m*DR{m@q@lEqOn>KA4#^rtq=DZ3Dk{-W;4s|(i$bzFM zKudl~OhG5Y`1(c@SKidFPA22upMzlSN};cCmT0{9={{*o_xkmJqd(8e1h5K`;xU`P z#L@>p37r%4g2?LkVsw=C_*I#M9Urp6ZaW426@(mNZ+0oFvQ~N~g z>Ru8nU&iukmCOr2RV}%c5)?i$HDx}E8k-J60f>C^~Z-v2Cq$Vh1?T_$@m(JgZ$l{o!>t@Nh|DPZ0E>r)deF7gyjb&v~caBcyWo;a6v-;ob2L@^o*g)2UG;hT>j-rMV zQzr=c5wnD&I0tt3xbys*vNp_`PU6_<=qFDWMhnNF=yvVewRQ_-9UUE0(?p-~JrayZ z1l5>XIE2*gZkGS}vGKDkXH2{=$jdVd7@xyCZ>JfW2!Vavf#I}*ktx^D!phYZyXk~& ztQH%)HQ+f~bg;ier>)G*0o-dsD9WUWq!HB*?~U$r0fBXOlUbayFGqS_a9p7$IETpU z;yBA7A(moLsxO1y%XE}EU|NCcczcu4g7>-#84*cbfRpI7h4>e8l9SiI9CO;(O(>l} zhcZ1SWls=V3IV@8@WBK(n9R@0bXd0IcG0Q?kFgPp_-h{f%0{wrW-2aJ`XZo?PWOBb zE`Ro5X#KIl0qPXkC;EJimNJO{A(W(;eJD!IFE%zV?wtO!br_n=TkKv0S#(qsA3Gr$ zG~#qO^A3de{y&2Wo!u`3_0&U|CCmmP8;c_y@SL-(5Q7gs%JDqBymdRd$ye(VRqvgM zdivDe!{h94jAHpG?&kB)^!;D?h|@M9@)CrM6>p(1G<_TH1@0C%30Z@CA#hP=;~+-m z;eW$rBa28va)c=gYfDQ$c8r{w+1gI{3unHDY94(MVceunKpcgKM?r>ylG6Voug*5o zW>t&Dr4Yy+fJOo@ zH@7R?3n6p*QNuUvFt6h92#1I^HT8$8Dw2utK97c_=%-IpHrYu@m15$uQE`mD76Wte zQTe!B_DsZCP;?Z|yqGPmp;4O}siyv=CU(6ekc>d4nG(iaK2o<4u$qI<_AxOfGuuOGviz?1xJ^bb4CL>%RkL*>X1+%eWK!<)Ha%BRTg6K_ zQbSvN;hWf)k56gon#5dNY?@@Y4X!>X8CRcS^ylj3mAo-3U$`E7qj&i+oSmnz^BE*l zaL5n~jg*aDz((fdt<&AIOQVK}UA=u0+NnHYj2QiU;ldY}r>cb^31M0r2Q`*WfBofq zPF|i+A%V^qO)CXGq%`Z?CMVpuWL%ua!ewcX;>^*LlqF*Vk`6$_KzO z;M${4TvtBGpOTVGHE3eVr&n0<{EXf3V+eGRtbCaaI;s1K*Nxb4iZ(^#IA+m9f1(6F zhN`ZbZR3Bh2m8A^Yqm`fa-3fO@$Z@SrFT3LOv1?0(>giLVGfKjK(GxB-1oQbdyyt1 zJz(TC%j-cf0F!0wB2LGpK=-hf$Rc7kIj@|151U2X`J9pT7H^K}8>}RACrzWd#xjp8 zxzgif|H!1ISSdb3FsXV$!a*S%Y3Zd)%rT_C?(gsF**br&zqZRgMF@x>Fy&CpUXn@R zV^-*|5w{;;cXSMXbm;8aKTrAjCGu57OwbW|3)>9L#OjnJqzaI8Kv47r& z-!lP(L0*Cxy=f3hnvC`L_wPyuekT8^5DIDJlz#z25X_X9-U4`(j7;ZbDEVrB-!mRA zF4~~O&z#vsL|h>DcrP8gf8ro`Lc+(cqM zb;!!13xah*hFfs`nu!S-hkYS?aLM*S4CA^jTz}X9SG>&IY-164kvjksL!UOg01nBD zvYik*pg~=sAY)-+0qT~x0vYLabIpvL&=c255Liq(ydw)U^P{LPr2J!rW5|0O2J8#W zZUQ#j%fga6Oa@6rwd>dWJ~mP}HJx+L>$l&{X1(FQXERltKy5Sy_$&vRAaWrD&;@`W z5#dhE&E@9gDB-23h$H(@fS9ySFP@EHiCAda9K zwL{aK7KjVu74ZY4NmEFj;j~Vso$K5g6qM;IL_#HQ*#2LGW}8-}Nsg5-q(W@m74ae= z00Iwo{yb1&=o14hB_^2QyJ{SUI8UAoetbkT-8Up;y}J&`(#wqXJexq(I`ntkARJt* z9EYU-;>-z>gR?12FgU&QF*OMj%a-;ppfu3Xa765;)7q#`w|8#?j`>Gjn(uxOA5}{p zT!$@hXajPFR*=)Try_V(+n*HKbiISPPeRGwD?SADkG;J=RG~kd@A66HgnzwqD$ASp z7*5kWu@T!BCgf8_M@O$=YM=WPa&GvmIN&YN>h6AobWfc8iruZ)@pUD=)wufCk z8;2GJID8Dg+}K6hpJB47L-O#@1tIQ8)i57+_5N$1K`AA5U>$3X*^G^bq+aGX5*%FS zF-HmM+khyzc{n87H_{Bt=bU9kq|*Ok4e-EzvhyLL%-9!rrFB1oTfuu;jYOOh zjo1KfubPE@r*xta6WdA?QXwzm*+#k*X)rf;8eq)I_wqCbpBanoHkt^2IBo46qSlb82`oEMmiVx?3VLBl#>$m$09;YWw|zR5%}gxFuhGo)yfjC-6;4BbW=KQ&s?I z5$G=l5%>;LZS0jZ5T}IK1irhi59l}k@C)U%j@O1~P?)lIEB=pF;zNth?%i>5GR(|> z7GpsoHd4p1B4vyKKTWzdZnP?fQqObs|&DA7O^+3GiG!R9l;svwH|4O z7|t>|85y5*%F2X#{X@OLFUS-nq3j{S*+qmljmEnnkh7Q8&20$@ z<-P#m*+tW}3Kf>KzvDXbZG1B+w zHvGrF<7I>*2xzV@dMpBw^jH7FlIG@=kr{kv6Bi%7o7)DArgWx9avaliD3!S)*I1ev z%EgjEc6VQHN^3ecgN)foT1S#PPfr;-of>Yku&{*Q&dd}z7$yi|V{>zJOfq$pNJvR_ zAffarZ~MZ`d`M7D`Pn17!*Nc3usWqP$g`mU?>_$;DNvqmHvuC;6c7WJ4$wbCz>mZ< z=hLg85K34%3hbWNKaFY@_2H=~1cr}%go z<<5jt$Duk1qKRuAceMxW74YMKvZAb)oMqNnHLwQWJBVo?ypW$1HFF_8a&q?GNo2Or zu7~1$#fgBa$KLtQ6^p ze2c)P0hglr9l3lnwZxpRX~#QhjRiy2(P&6_u&B^w`q9fC#>B0Ev6 zzy?W3%hr>$U%nD$%cAD&(&Zk0OqTZlh_8lhDlZ?Owb88vJ>xj{iOuWQ+q>CJ%{x6I zK?}zmZ6ux(zsk+{U7VePm#m?FkUtgeS^6Z>k}-ZBp1|rd$Biwc>#qNp^#wr3xL^;ynUDOZR?$}YA7;@;)=Ld9ljd~l# z>YR_6)7*f@)6+)PC6GAoO-g`E6;a`Hq>>uy!Qo|@w)N;FKVUc3@{iOl}Jp+Xm zJt<0i%!5mUE2(%31vBVWS)EMH%$$sJYNw_3@>+8n&6m^jDB~lQn2%{9_JKK-@LYDL z#YDy@L{HO&Z<+A12*48vn1P6*Nx;JJdGG16RCdx+*5sbMWl-#zgY*~AUvh1E3c}Ee z5rT=Zd8lKtqnp(ZrW%j2udmPVGYAzaDJcov(h-E^%*eGGKM&qAbDhpZ47cczkt4K$ zrikNWK+Q^spsK6CT-LgH@ii2PMLB^kXZOydTmVhN#9xEbEKi@t8Oe!w=qNucvK3x0 z&wafe5juiSCk#*Kq&=ll3?LN@z6Rr9{~tN;!=nN!C5yi6076+CGGc%#c%kXy81=>` zRuH^@ANb-f8X6JX(`g36q_Fgg_$KGaOaS{_@(&%_DDNF(r4awRYh&5ji2o)XpM?Wu z@3(CPA;dpltbTUY@W7|QA4J<4_WXYQxn|?P@m6t?AM{4u|GoIe$M)f4bfj;4k|opMM=k=o{ePo;qz6p=#Y?u>?m3B9^U$V!NxB5 zzg?stVYT}zD`w6?o;c1g+FIE5^LdZ=Ej|*%s)!g~F0yrp$)1Y-ZawA%O`qyS> zTGQ4yBe2k+-9ShX4?vVnDO$4l3pFKUW%=2EZ}$vqpFfAN=})i}fFk9Y|E-`!`8UJH zGu)1&gHjb73{aAQnq=~aex#jZft?}@>O@2k59jYmAKZ9K{vX%72sHIwdH}mSq5y?g zZmyHPebG=_A)44jqM|Gx?Iy2xon4L%Tqgt&G#{jN*VcwKb0w}8!tBg+bgH_#DG$RR zJtFkpf$boXgMD&zamf#o-)!~7&u`P_^V(0IikgrtK#VQAt5Y?^%<2e@&_p!-?VFgM z9>#Hl`WI|aXb?oF0jd-7rYO#RIg|2seRobv6JBhJlC<1d6j~VB z#qAm#Egvth3>?sf^1EcQ8SEdF0}E~SIJ zZ^`;T+~Y4Bz(RWL-q#fW41c^Kipxar78uI_>)xXS4|Th9w?6b7^T|ft8E- zwkd(eDICq)|L?#1fBoy**1^!q&(dbOd#vzKx^9^3~~8RPol8^ zDUVfFk@ta`nwH|rOu~Z&L&Jo6A{p7X<93_%gCiqbp?S%|#x~Z|!;$#LOxdt@|0Th>f&X z8mq|rQq%OPv`4!VSNWQeRJ$bT^^)$an(R#?J{-i)>Wg#q+Y*SL1V3fp=C;}M|iRy5M1czv^`v>WIqa_M2P>~3kHxR=! zAj7zp)!h4uYTMMD=KFj20k37lu_MHl+W1$fccLHeltt5PG#Tt2YsNU5npz#&5?c^M zq5^>342{n)u|ro*d|ihmFP`1UOF)vt2n1n##URN1%U54ZYySJ0og>irb#5ZWtR07) zJIlzK4sEu$MF`Q5#rO(nbsD&hc-_01_2}VI<8G;%sNEpH&g@y{DxMBBd zlf~UXA5rPV>;v}s_VcI>K9TbBMCf9&hMpJ>5MB0zrO{^QntkT1OJG_j{D6o3?Bwq84d&kc(}U%Nmqhn^!vPL9J_0fhJPPWo`}R-mtMru{sCX?h+FV`F1+WRHLl zf=7#{zWp*pS&&WG>ggfGka*m!M~+?8Cnq=iE`3<}MxU~q9AQkn1x~1R;s!9n&&S2X zGxVm_ie~$EY_7C8Y!cKdv_2wNhm0+|goIu#-39SW-~?_p`Ku!N4Ap~;rP*_Evwkf& z^PzC*(8YS;LN#(+ov@UUY&;7zL!2`MeGSJ!^8Y#00e%^|PZVF_al>p_1NOH@K;52j zDb{Pbn++{&3;_>(eDdHkIk6kj_T!h}U3GE2-9k;|FPfrqzT7suO1|7jiz<*5mzI`x z*|9^7S>aX(u`?>0o7+usUtrGNru@8=yG4vy`$`NuyB6sLsb7txus#`yv}kUmH9RQD zFocqV-hbmh!Y235qv5UygrO*Gl9Q7|G1e6u@enW$0@NW0pZIoD(Em%?#%9O$J0Yd( z23kp6KY#-k5Q@J8+dEAMKnLul*}?iDpm#kxwl=Yno^k!v*MtjugW~P$Fw@>cJIa`Jj?|@iUW4=pb_e{<>`w{ zOF)zJu@QrTNWp_EMjCInNKj@Gx>2^V%eNieT!INZluX0wkl!KRhLG@TwNc9(%R;ao zVuM0oT1ga`j>v6_lb?9kjiI_v5CtTB7r~Bx9*EBLTTNuZC=RF!mXn*x~jGo-O-D%f#n}R>R5G$13-C% z_T;x|gDaiZC9GLFzOa6_wC7vPnk8Sf>9n$QbYve z20p4oK>3MgGT;kr9@8}_UqeKG85N_Y#YK5}MMYjuJJuws4bQ;+aL+ta?h+8e1S6jh z4hi8Gwm?jX`wB76_iwZLP%(kgNo|c@CK0>ezzO-PGjjWzv-&`+__M7IW)SXKJ26f4 z`Fg;Xz`ae9$`iGsV3du)s6yNdkaFF5GZ-6eR5s_%TPWWuL>x1Gch{ro_()U68NzSl zWUelo+1pP85V0?RVvbvh7DdT zpf$f{+EJkqL~mDQPwFu!(cXuFIaet>dHwyp?MonY!L}s&v5?c_1HJ(SuB?ppFJqdT zn&Rl`?QIdgBxJ1p$!h8vJ^hz2ou9OcCBL}up=KDQmtp;~p`t!%g8r(w<>h6FeJi*H z`uNPF=k8kdGs}<|`YHkKPP-AkV6=)rGWt*yQd)>E(UKFPq~ zQ=s{r)yCoD5pdqz&Wy-W>(-!b z6qs-xU0hwmMDeNKXKtg~c{W{CC@JD5VNBI-!~x@ZpXpPUvxr140@I!@T4ZV5qB>MN zxw+0)MgDPa`uWsFhftY9NgEO3cTUq`LVOxUVsu=x!PJ19j9-}d_;Fhlvf;Pg!CJcp zvaG+Ts5EbIdwF?Lj-l`HjZ3f@uc(gSUM{ZlD~rt=;RmbJiVmUFZr9}jTocc#9AjrM zfFHsX1S2ACfY2~XQ}Rj=sQ=iJP1Hr9(a{E{PycJ;hUY&s)aW&w@rhIMu?^q_1zAZr zoqrALIN6DWXHjFFlfWJC=G~A;bVJP-_2^?clk@$IJ zUwsPv#d~OJ4aH+6Hc?T9o7{B?`C|Wx#i`Bq6HDl>t@QUvapWmUm~p^Hu(tV=j{)2P z1TR8Y9C9~vBxbl7(<7`G_les77pD|ye6rFw(}|H-?ITKOQ%|$)*)t7m_7hEs!PlUt z0PC&s{ekynR*DhmHhvb4kmB^Cu~*p+Ue~VCt3_`U&$NuR$ZRiO^_G9 z#$r7>l%>rXQv8Eh*|KtMI1Jgv#O#lhBOm5DrFGyy#pjD=NfCk>q_xk}Dp!6^m9&>+ z%x$#xILB;S4+Fg47cA}};C*7=rbk?fNP=dEwZGjaqxWAosbz+G#RLLAAhot|LF9sJ zsiz0EICJz%{M&}}V&0+Je{1*i=hRy-S*6X6<~Z#kt|_mnKE3QydAS7oje5lW0sD#s zlBosEj^x-U1_#&WWfaG*U;5@W{Z3%WcyS+~?%eVsHg$8!1-6aVp*eP!HE+w%`mBPD zBE5bm;TEn{MZi%fy|}QcNyhe;aO9sYb1`GHBq##0|C-UH2XGskz&rKMU9MAF`S~rm z)$NyvAMB*iETvkt+rcI&j;w;{XdUl*AI_BJwYG|yy27ucu4)HHTFsjUI5-k79o)E% zwc}cy=mw^zqa)rY5cR(NC63}pYEOkz)ujzF|n7G{+T|^%U8aN&YMp3U<(G`YY{0Wlgz zf1n{4oQb*}5Pd{6DD#PN# z|B*ajo?HcRVV%G5;lrm=UaR*(16^1qeK*;_;O*OrFPEA}Y1c#WiYw#PMH}e^td#3{ zlmhwMyMO%>Fs(gw9tP+rqCQQ<7`q&6{*5Df@VH86R%YfQBqZ8LOD}7^`i^8y|5PSE z$r-}M$7czr+T*)b?WomlA`0ug@oPckMu;QNc;cEOdgEuLm6frikEWZiN7Dzk!Fj>%9}==^Yx>LZR>iyj z^|>*$^&!YQ--6O6kU^B=teQJgFAJYX>0lM3X&fvPt{ClTm{h@EEq5~{2!e91ZM_?} z%n%_a5g!dUrt=f`4u2RHMf3yv`6*y31F#__1;t=PYT>miM#AQz`7|zdH)ZJU4cofs z_GZabr{?>jy-HwG(UqoNl!d+>qM5@)Ojx*avOn+Mb|yIUFsUOKzmPI@W?#Q9r+atr zdVp6Qi;A7e*S_$}kDuUjAd+k}#A<12A?U+UkqHdvWtq*~^rX{&2G0%!U0l%6@bENK z`n^BB0KAxnBl6#Onr>*f$+~`_HI8!lu(+?Z#yN%_wCz}a9jYL5S!DH36ETwMpS()_^e*LWo%N{;_h{cP?7E?w;GlJL^ zJQrj##W*ZTzRxQuR9X)D=UNOKHe}iZP`}a^3hIj#Q457Aw zSSA(bgmc3{?>fmdUJe6s91WN_@u-9s6c%1~L}c60dzgZoob>c=+hmM%A$|_yH&_Kh zhdWHT_~jk--v8UlF%xgyGmo`D3fvli2z<1d?Jt=L&%=k~tuV|n(mGeNG@AA0fw%l# z+UGI*O-#7>l)AcRm5H9gn7?>5Z&g=TUM_N;z_(ApSMS}kM|WNYTZtg-N9B}=6Zj8Y zxjxcjT29TuLGrjwA;^nDLZ%2Nm5Fi&+-bq@-dW_(prG|csn)VCA;^C>h2vfpOTO`9 zH&Tr0&G_})Ew8C6VYFNGxBWo3dkNk;)W|O&PqX*5>7qGZxqnTorg?@ocNjk?l-L$}A=(1D*fWr(2H+5=O<0 z`{3qPV(Q3cM_Evgd3-2O=p)iDyCSSSyMs)PJj_idZ~J=K!E&kdR-qqDl)O zImgi><7M?i771n2ERrd=sYtF!TIuX9s){%dw+DXf0O0U_ik*+PW~~p@HZg2wJwmsR zs*x=ryfY!CrdJH$krjO#GZZf5lKiL8ZsLt0=V4lO?+|wL%Hg9uE;s-A=Pg=mYin!& zAKu*#&n@%{Pz@s0P6=Q_qS?(xL_?Y;I|bImzdo&dD0iy-BjCi2Af`BLb$^DExH;M983l1^OXI$An%bs8Pj0MxAS1O`wLF(AP@fJBrBfRjHLTK zc4g#gZZoX|bn_?bRKp>I`xHVmSCovJi!=nR5C?A>v%Y8N&Ix#^`w}ue?E9Pd_T7h? zlLLK>ii~7119BWRX{+oUFJ%$52{>h|u`qr7TcZ`-*6{IynlmEs-Mb6Jg6713?}Zl} z5aYoOIRy^c*RNt)^XKl6{oUrZ5L2O&Lj?f@v^vWF=fC{T2?|Id6@>;33aTNjxMEO% zX!0i+uc{b8N#P;?;g9CG;6;G$0;?tBq09OtoQ7e6>ys5S^b!(?|M12IxRZb%4e%t7 zS&t!HM}Ru%0@T&<@*JW6f9vHF($s7PM<2Ew{LdW@8(g+PSBGv1oKywKoZ8r=;GW*w zgVwHx!Z}~DKOyS{DXa=4`_KXB7!8HKM|ilSPdV)G#;+f7F8#ORNF~#L;&N+*(r@k! zd=bz2OXE_6rvSl>sYVk1EOd4F$_^W3q@?A)M@%b;U(;e}S0@=+%}bf#_xTSmJ#q5^ zq)f(t0gCXh=6g;^W1g5`3c-lT$obV(%Int+LSLRI^#A)PHpBPukKwKv2VFVPsJ*?t zp{5W9f)NV1-|;|^fg7XY=H+v~y(O8)$6@h#?J|E89r>5HIM#uI0Ln{{Fsy-PB*5NY zQ_|2kSgYB3js+NY79r$j+WB6%v87V7=K}VS$3h$!W zN9TP1!-rbn9)e=h#K;JN6>@{2=BcEWz6w8&K(nKz&J#efjOb z!9kgx6nE`1K0eBjCkC5t-oEYX6Jlm&2CIs~Dkf&LA41L}0;)KEB2Ku4AQK*Oi(}B70n%*pHE1sY@GJy?MfV0btL9!lJZp=`uKLR z7XQIN@V72ex=emo1B3H7<3oF z?CcI)#Xu#3Z!rDmA9G9=of*{m0E|Pwv7yu+Vi-_t3N*&?ouF-qEdqJR970oIGfhBW z%;{<`l#>+|e7H8Tyf(USZdHK5cF;^+@9sXFy7a%xP??V=`={2Arlx!#Z3Dy*!2whX z+S&Pe$o$lY-Jd-90dv`~o@y}{av#D@!n%;vDkYjv{%tyiH%^kc_<#QV8Cq~a0)j>+ zHqVC6ktNTiixqc7EwUHRvq->?$U30tPPpn~(?4kn~gn*>A~S}Ls$4y2?|&~QMX zQ%(-M%m3S_v-cGzvFis=nT4CQyu3UlB~an6LP^308n&;9b4`DvqC7vd*&loc^5=)3 zlex*J39SUdaC^nX>kn=6<~nBAZ2r~znLJb?~46BpMO zq-DndgsrZuSQW3~7Er0Fxii4K;>hFPihXRCg9@0(hT(7BMmvDpaidk z!wr1l4%iX^^`M|WrUZ-@R-$ic2qaV%(5L75@JERx)&bwpX0U*oj!vQne%Q(JF*|Sj zzyOx}9~t)fGhBI(+BzZIgLsVSfvoxkEeG~1;Bcb=S$EwsZj*#~oiDTr2LBJEWFXi; zKY(?72q=39Ealbqk0cbwoE~y3VCfa<^Tf36M_}1>Hdvq-`fIVTm2e3G3_V%f!^U_V z%a&)a!D{if(Mz}})hH|V>n}>Tpx1}iK#~{US7N4wn||@m5bp$|fM%`RkRUd@@m_@T zcda|w7v$F|ki-LJM#owmf-X>;I9eJSX0e@_WjNI^CMh!ocEYWeD|}h@^V`Q;oH08m zd(pgI|7~MI2(>R>OmcMz(U<|61}IgOpz|u-))bh!ibCfEaI59fN~IovsNcSMv$?f} zRoli`R{>w)>%_#_w@XwdExb5t_kT~I&*0N~A(X7I+lICPW*HG39UU2Ia@%#ps=Riyl?4SxDO)hqLEQ=4S1uX7 zOr%WoG|2S(1;pX8eM6JQjHGN-(u0@F6j$Iw#JCOC`PZQX0q24oayHNxrU2T=z`y_u zPAmowE*kc)+>zL}B%9{1)m7bNNm9~CY@;i!9~aa3yIa6X#(6BYdR5UNNrqlFHsj=I z<$v#7@KqYjgFr_nCoAg-a!pX>&4CTYbR-n4qF|H)4U_V5x9o~qX#W|)euq#Cd(OdC z2{(wAUt_J7K}}6Ih8__om@P*YCz7j>yc~D+?{~Jv_bi0Gt?jBS0(zj$9UVJ+doaWL zbgLN-C8RrPvFW&Xe*XL-fYcTD&2lkn%v35h{3vpwe75gfR`dM0PKyzHaN;GhVquH|m=ND$?BE7$V;RD+hA^F}|k=$!EvxN58n0BoDxU7 zOw3Zamvj$NzP=#OClbwx)a|L=^GnB$+{0l^&nCqt2Xb)&DvE$;u;JCY$PKvy4R%tY zBPHEp(-0vg6IN``giGlmUu?XPo)Gpa@#75^{y*)}#rI6x-j80r5^z=Ao5;`8k*0dV z$@D4aZVPk> zG9H=vP8Em03mIXBUo&J0QF9F1;c&yt68^)NbL>%jTtq7y&&rX{S!*Mbc;>0rAc>wf1wm6rnDaU)9%W`oxu}<(Hd}i z;VI81=rOahdO+I_J2DT7W-nk~f`Vf9**|;{M519S#~w>R_ANj|9Kot^b{e@V>Z*AA zb~LZt)*M37Fr~D`d4<`_9lkn_`YqOVua^N@wlk>8Rh>UM$fgTXdZm_I%pfm#i z=@oFeUKwdDU&+8LcQ@IC$1G~hTuNJWOgyRpkFZE@&0DO$z~txm z=W^R&e_|BZZV)zrB^E)!2Kic7Yb)v@pbhzL9Q7-+vyd6OARu`Gab3*R2eCksTNxsr z4=o~g*PwXumfrZrDPI9r)Z~Hd#$z@F z=m^C|SVx76&o3l?HVQhaaTTz8!lGvW;~$BqOWS6-nyDc)m@jW1i(}B}xh^&9yxL;y zZJvQ6x%#@RvoiuNS-_T@XG4G^27VZD>0z%r6yeu|zO2;MF?cex6#Pk$H^Or(dg4oH zwyNC`$6{h4A)(O(Gq=?STHAPDC2?|bdkQSAV6TQU$Z0R(;UQo{KuY8n5K!~I8QMtS zzduGmZ?19kT*aFzM@)L~;dV$$t`W!tRS@BgEGf3ibPQ zdovU`W8eo+dQ}*7VW~@fVo)MZ*wzz71Y+qeuD@dW?ODm8oZ+^;hhsEQRqD7<9UF_{ z-vvTN%XI?Eoc9Enl#dsQ`i~C;yhe9TIGB63y4oxwQbK4Pr{Ar}VZU>qEtIB^B}^0< zNh6?P2$)|ac^R9pr5B&HzT%nMe*`K-wbzpSXP_g6L3I5eS zSze9Ex?<(k%^0qy=+msj9TJM%Z3o$C!H(fT06YH))S{Dry z&mkSbru$zZ`yjiI>3Sf5C~V6mj^tN^xCxtumKT3?jj^0*xA{^ z;!$i}UQ{rb2uLE9pKp1eBNfpVp-J9uYU4Faq81Mj(mf4V3k&)3Jg%=iU%;vFEm2mY z*sE70TXFCSINagi&T}w{hTe{TT zY}0T_x+zFSMhZSR@jWmC?*CpX-x;wnA5|BBAGT?pAhI~DqM;Fe|4UCO_Q_(d;Zb^D ztE?pBeU5_FTwP5KM!;oTJno@{JXVP6hL?CR0=)F~1M*bqY4Bo%)^<5s$4N#xj1s9m z-)S~jASHvXJe8FI7#Xk};pvEHGclo~*ERaZlHc4V*A}q=nm#(LX`Z|%fSB6F`t=U0 zvGE>?13J#(A&(xB32UO_-JhFusXoTugNefc-Bs}Dd8W>`&93DazHEC4lGwf^AEz~7 z5vka6Vqc?~46nIp(#6dF9Mu@C7AIij=AWINCL5Zv??NG>{dx#q@K^JQ@OczxUV33ON-Aa48`6JFWr3H_dXW?yl}Jc?z~Hh*A+=ULSF zLf#fLOBOS@r~AnwM>6(vrYCN-XLaL9y+E=|BywjhzcJ9t1Dp6HE*e}D#BoxsF1tw3 zGh8}cE_Z_=>($GZvFukfeJLLW9Daftx2xRw3B2~}IiN=I`CTM;>fqrRJG(RB(tqkHS7@Eaj_7z8`lHUm4DHw2+NX$zgSW zid|D?96U-f6_T52)Z$^Nq(JhWQTr^Vr*Pr_YDQ)5%y$k6-h)^DA?2y)I+Pf<|NQp$ z@X%ssAvx*ZQs3d^3kcb=G0Sn-lCA&`o{#zWF&$xd^Kl^xzQi9e)SZdpX|V~5z_36m z_;dXI%vRt%CRSa(-2I7qxJCngmUoSYnAk8*lj>&L=UcFl#U?8h*b#sgPO$AUs!RRR ze1(SQwV@zHWscHBJLLw-@jB;`arfk-hxZHMh2D7ZBe>`1jN>#DqB1p#i79eNwL9W- z*>#9-XTq-Kd-1mTj&Rl&@luiaJZsdptMd`*lHpV z_D4pDfx-5auYhCg{V$@hT1>mF2X|Z3J{k}YS zP{5-au^|)DrluY)6sM8=-Nw{dn+(e} zQnea3Z6!`kHCs?@Xjt5tPF0p73A)rrs72!l)sHl;88% z0%tfu-iuhgSJHAgrP6+O4iYRWS4@X&D1&s*=LfBwdB03|3q2}_OM53yjsM^&7-&-P zl)9JedTE>;MylqkYDNRmfTY2){}VM-UbpyMn1Mm!UuM!L^*52`+WAYN=g~ z+&WzH^XoyMrVgs>IO`bs6#JG>^zW{0ZUda*wjnX*2>?uyGd=tgbLgeS`1vcJr26cO=xz(h<)fZ*`2HbbY5t1XJAqUjhy<^!2QT3yoAu@(v9Yb-OV-xe zzY27HnaQNOj0cnk2GG`pkf(i#uWPZlqA~}ui!!)wjoH4nu9zyqwXRhm z=QTg`S{ctBt1fjgYk@Uq;@60e5C8WDV|B?xsM|vkA3qu~_ne-R395Dpr^6+8v|{JX zVbm^qN-ZY`G5zf&$MwYBH8tb3p5E|y-PRXsYCb+XeT!_QD@y}VF}R`yam%bR)oW0~ zeEf)V=#{P7)Aii&kqqwlr5dIVrU)cYhiz-0R`2;862L41e4%A#9@WA^0e97E=cQn% zsh6Br0ayWVoG^gE4|fJo+1qTI_fIIpE%$BX1X4fW^6_EfE|syVT;cKYW#LxgGG=7Y z3p*zw;x-p0?L0Tf%&n4{Dd6zz_H8=yqg{=&VyBfYpD!CML^j4CENgxZ5R~ z|NSumMOJ@qt~(?QKwb_OQMAkflqack$<%BA(Z&YKp`O5P`GKG>EF~tq#gb=45Z96w zMm#`|7I`!>Q3s-3cay@>=4VzDUfq;xYTV|4JN2apQ@Z|mUMR10yc?5;9^W7F^lIE{Ue!wrg%^20RQ*(D31_BVKoy-&}b8?ID({hIFRX>fmb zKCtZ0PHX}z?&v6|ppPXG2$XVuJlx0H$Sk;eSSi5V87qq!XK}31zvq0-a2EwJDKDP1J(UL$frM7OBmKQO$z@X z7eG<|MV$QlVm4{1Ga^|MDj_<$h2-GY5=AkUy)`5Y6M}%8nNMBxEa%+?*;vyx8Tj@B zNnh*JI^!|JrJK8|wQhbN3QjgWJs~!K{)`ef=`nd@B{u!zzg@mQAw%YNmC-rb%6oTZsZO3N?JJHqdWX%fsQVSN^CpuNP zT%~HNtA|SzVt7((l+%N*-`JT6?w8(<;Qm&3)Ool#ah8X7h22wHUth(|g;d_y%WHFs z<%E=uCHvRo+EfgcNRF0GZR@$-UT!lK#C6?6j*faF7z76L2aq=|X3w5pyfawwso47#JuAX%%RL zGgPu8YHbU*rWforw&DM z8UN-$cK2mMCBki?dYTes+C*5p6Y-}EI=R3}{YcN_i-doPN2PUj$M^v@yM6qcqXEe9 z$IxiPVwQm1N0QHgy)bxufNXRgs2?z}#L1}9-V_{cw;3x8z-AA9rc+tT8P!+#vgH&; z!>`6V>f~Qp`z`FL>FUbNtf(5n$ti^oEpl*&%^3kJX5!P08E&T>TD*(qE1IR2W@BjU zI4Cdb>OiBwQIiPObh0mP9Dniza-csKpCBBKcC^jagmx{YaXSOj^N6QUe&?1u4CuT+ zd*QwKYd&FTqsh)$?eX{U_kc)D`R+{l!aC@Ayc*J<@5+d*N8B|90F<>+!7CKnjWam2(0AsV)_@t<^xw?dL zQewJ&sTL#kazuyKRPy5`1Aew~EFH{;$7&yrZN|rgOb#wpCr}13@v2o;(v38nSq<^I zDl*713)jjh#KlVad1k7dp*oSWa{GeL&pB%DA~rTiPs$K>_PxJas^%#JZztDzs-$;v z-zoEh%O3Y4#CYhO+S;SLn2)Q@Lyu-orKp|Mfyf&`RB@10#40u9JjGg`LtO{qs z*++&FyVRTPVYHzDq_uWnIMCW|?SiX633#9NMCyU$2X#Wu6Bd7fJ}V7M5Uw_S?cMeu zDlFDA?$M{urnSM7_~w|N2)6#N^J%ixinv@+zowJk6F^g`pBpBDX~~J|9KIkWGslQ6V4;^SC?xD9lExJ-6QzB29E&q6374a%Ah|5~Mk@&W2)a7K25h%eR2OS?dlWE{NoffxvN( z;1}*Se1h^r#_H@R9N;3WVLi05e z5~jOfha2f?*WWDmLngw^3|xBwXLd7#r$LOOdd9^ywIn1)V+Y&4tu5VMqS=XJrREEB zG1BaMnm6$y!FsAOFjtVuL!O++w3i-q9vBtmfiz-I&_zBfiZhD%UwO{#{eCSlV|7}K zfZ61wrV4iDt06Q}a^>uDw!1lwCcSC+{}e>PjVB_~Z;*2)+}>pu+b(1An#0RGK|=pa zsioF!?^{o&WxSkL+8s_T!zjsFWb ztifOUqV_6oSRi?B(^(ht;Y-Pdm+Y5SkI%L4Z>Zf~Ue1|7mXAN@bj*?ru4{!Tf$ar& z7$n6b>$deHVnnocRLb+P$;!|-hzH$IIZm_+);CdTG+YQNAE7uocMerz_!0bY804Vr zWoO67XFfRMj#4hqtEnNT6v^$3@g;P1b%)vYCTPB97SZD7mYo`Zck35Q#VrEkh9bx! zr)CY_wNMdi@>7AunqsZ3IlGD4(uSW?^XY5$P7y-Fm^#mv4PfjhBzSm0Eko(vUM0#V zd-?KkosMy_hX?U(COE~qd3n0?ojwPLtz$a+`fk6FEK0xT7Ekteny9i^4))bWQaNMb z)_zV>Ma7HhWZC3jBrfSH&rKakXl6IvJW3MqEsqyiLTTQrtG;{u<8quJTU>vpc>K}L zip+EbH8mZJb##Z+b0}lYd1l8J&o4ZP%>d^P1kDanA2N+0~O6 zS=%yqJi#mfesmb&v4;6RRo{P!VAL@(f*k{3TuDnA)g6^{EiH?q8lFdsW6oF>pnR^v zWG=L4X#0T5kmQ=^!(1Uf%OK z)a9ax*&2?fdhxSQXRa)@w;lN#54r{|G$Z=W6~o3wzMj0 zMz1dPIfq@7a?J`D&;0gQa5uMsxf;$kwpb@I9&PYWy_R3dV z+u^h(h-zS}9@3G}(Bl(mP&giceLY+xLC)K18Bq;B*K7yLBWS&+LCr1DHlec+81*c# zJ605QYPO8X&IUc=?Szmo0av3sU|Uuuk&>jY+foQI7Hf2N;$LAo5S!nBMIj7H2V_|L z0}~@!l#~LFSz+j*H|OD1h)aZwENzEsQ6UO;W++JNy>x6>4@Tqto@e})9xzl-DSBu;w8wlYQqMpg(M<9_OP&& z-4!~sLx&|UR*n~_8d{E)LklxbCc1`0H6eXJ<6x$f&!{pf3J_0Ojb+_rqct0TSi5hq zv2@xa+1CH{%J}R`E}K>k`h2lT`5hsbawBB9Khf%xY7ZTq7gnz|| zv!w+N2Ayca^XFMA(yOcZg#Aep{Ycvgjsfp;{o#+|;^1h<#?8(%Akv?=3~8*cnh%0O zfJv>z9Au-vLW_=Vq~IdzPD;^F5Qep!UC8_?Pe-ec`~?;LR+rY6$tOy{ZL(g2Aq3?o`ER4O{Me7THr5MHKarjZ%F8E0MsJ@!Iz|FXiqdl-;n_Lklg~^-qgB0@ z{c|1t3HG9B|JAm=UDZdgceLu(k*mQ}(6rl`p!4UPxMRp;(flu91A7w#Utgzsj9^q9g)D$sw8ZSR0m>&5;G@VLsP-jFSRZy9M|uN)fq z-qPYma(I1pwd!CF#o>)4?w)ThLNt6G;n zg#=1NV+r8S`R5v}n>(eB7236cz-zhl^7_0pr1z2&$*lJU#Bjy6)G{h6FE1p2A5sH(ke@yc8kGFemmN7oZMOiLcvqLvpHn?aKIRxs0eMzy%A-bK$adyl` z?}HedTdxf`ha?rBJY8xf(K?S@{%mzCXbOqZvMz8hrgH$aNjkg^!xbgr!J_+ z2M_yBkMweNVk$+KDm6+i%13g+oIZN`RCm7D-%+z{;%(`KMQ$=UhQllGI9Bkzpqk?J ztfHSApE9tu2jHY^?-RM8w)G37u+6e2PK<471iHFhW=yQHx}68Pkc6&(am%+T&uAwYCPY2g8(` z>?N|?02jTzxX{OhU{;dsh%q_!qNi)A7dL8!#oIhuBjc8O8azCFHp{WOTh?I*GjbLd z%9s?AC5Q6tS7Xzjml7C2ix%AePbmZQC6D|_pjxq#Q$*cl>xmbJ+SCUhx(i!7h3b~* zJe6$YI%)voJIF@d*h501C`wUK=}~d`OO%?A{k&#B{f>|_EI&W@?VyNy`^ZQ?LwIgI z9*z{tbr*6lTSpild5n!zJkvJ=p&0Om3?2E-^T+|i6*|g}_R{Zg!h|Z>jLnG6+Zqk4 zrtFJ#YzqXO96aLAF!CN1A1S>C2OUB+rSZ_1-z>s(!YV)KX4@M>rS`?h8WKjnRVvkj8dS*#W{A8b6kZC~{G zv2tNjj_SxQd>fk)w>uiy3PE==jXNxrERRoY82zH+^TLQVSn)5n$dpJ1DI2DgS;21P z$R{_KOi62PlV)PVztXFBy6K^%!77E^b;XlHpE74VZgCi?CfM|6mBQ>}1Tb6difU@D zeAt!;kkgk>POCvN7pUIi=6m)XZ=@hBr$La)OMPizL~`226_H>&mm{T)QD>55#!gg3DED^ zG=rM;wfuy+Qz6wajOQ{ z(Q2d@-ZIk2R_VP&H7J>Ld|V7YAen&Nd|Kwk_t#inzb;EEzCk~Dnzgsm+Z+O9233aC zS2OKP>FF$Y#Y9B16eYyp5AHsFx>P#gu$~AB_eke0zHdn94es3{{mEV8b2)gcIxCA2q_DO_rNirBnjSYqemy{?H^yGT3Q!E>sTgP6GpnQxAC=8F}qkE7hihtt#6`qy{_g zIrVQ|ho`oBc>!{({+U@-*`?yKxJo9<5>c1`nCsMerhiNayizC+ra$H9AMUD5)VY>x z)@h5Lr2Lc@J~RJt-paY}EupF8OS5Vxw0r%}eRAdeyxF!ZWC1CXfhGF&vL!YnJnlLH zHI6IMwX2fjr)y{1UC71ncp9uWwJTf0S|SAGi@sD)fXCnZeOu*p`u&danbf(c8$dwm zr3p4&)1{*|N# z8p`=9Hs+e%(`!{6(bKC^R|84Ioml?wM!)5CYO`wHt!T8D2MRSEHf~$p^Dw#b5O7~kMBg@OJzT6a# z6Y%@+v%$a2gp|yDN#?^e_a9bIEpxBYX7sKDOhqFLb?@=$a&nI!Ix13qF&lf!EGI~;mivw3Xl{(kU zHaf``0+RD_EZqn=p$8y{#a-43J^+Kir(j(P=C3fkV-GOQn??b6Mzrft5qVITqttM` zK@JhUi0N$!)vQ>jZ`C+ne+%+*G#Y$fsb8@qwnai`VQ{cLKayjFCnmiou08+00mih-ce(lT7|$ZDk3 zTxF!}aDgF^($k*inLOY1>y}EC?Q_d?SAhwmCOQc(uP6#C$!EsJGW0(0ez!1KUDnl( zHSR1a-P=(CSN4w+;Vhe5q3P*w-a-#K_tNT5>OeB1?5OFL+`}C!WVvMEXS>;73z_S- zvN6We+Q?Ilx0gnYTU}f-?2IZahZdVPS+kT8My(l-@oM+t*{(cA>4kIT%ZZi-zEZD{7k|&kjxy(D;u|dETMZT+#>cHaNYnLrFpv=afIU&0ANYA} z&T$W!&aP4E>I%60+@*0}paYCaWJQ)9C;&?|Zu*m-vjIFSGm6l?~djO(;>P?`zrWG%x}4_W4k8z?h%tkyE2F}OYM!v zoQVVC4m~DdU+PpB?VZ=I0%4F&O-6p5U5%m1$Bn_45R8T0OdIe=;>W5nCBW{5^-D^M zCA%Z<5a36Px_%?v@iepB=yPC0(%VZ`+^RROTf3-1KA5X%q({k3CjlPvn?@SBQPNX` zBjj|?&ld&Q@*{PD(bUH@vP|B^n(gJ{mSrUE>drz-YagDF{1l9i1YCPmxl%2M+J_#{ zhXVVht}aWxTSQ-GqyX zLT*0!q9fi)`m^Z181BX9sfW6uh^Z-P997zB7B`knnvYZ5d!KU|`AC4Z$jP>5KWW^9 zD`6M4yIdG37RA@i^VsmmKOK-leDv|VP*+lIJM|burC_sYGob_wN&zUw^0bF0CMH0M zVo;cpme$zR1ZFjdV60JnjOH$sq3;F8FbHICG2q2hTo%G1_HiEX=IY|BTT%X(=BfGN z_G1D@QRCvyg!0@g3-?>fCyd*S=btx8n-34I3E7Ogh-XR(&fHY)h!(I-{1HJpQl*ls z)zU0YOP^*E2maWRq#)Feb%hvHN>%ryAn>)4o7X!bUW4^sDW4hYf6Lqg; zKy7EK&Ae9sf|OlR-I*>Jr_f0lwGF$Sy9%k-O}ZA%Cp-e0#U@LWkD@x5wNy*ZnfQnL zLs9X0FivSEd2<#Bs=4zhD>;`sxjS6<$7&G=#hp`!fXi{HOz|wg<(j8 zK?ndh=(G3<+A0vF0BE`t@|2P(;?_Bw=UfgM?+G29{^hM7)0vtAGJ~~*f}$el{K!bC zhx1;0ENF_XG=FePUt33+_}@-YFPCJM<3K)+s!o0E?J71Jsplbyw)D}ZMSC1 zLcUJi<>PY!p)p6z`gia>Q_IzK1o6>!ES(}PVzxCL^t$3$Qo8fnYLsIRaMszs4?ttx zkUdXLOsnj7sGM?|jBYZkun_Qq1sr>RwJeSa=C_==@4`fL!@}mKr@FcjCw+a7PmDN+ z{Itk#JPwqY4Ro6tAK%l^BeDd78*#Qqo?cg2z`>)vwr)VrE0a4 zfb(&x_KQy@>tcLBkp_bTs(G;ni{4^w#Di!$twc! zFEYifmoxjR(GGOf&O9juQhcuLv~x+U=bReeP?L5JqjQ0c-dFfJ0qwBrAQ>64Av3+ z!x;P7zjLbI?Z`^YS#q{jv$E)+>Y7%ab_}m{kKmf=R~&BB$z^YxFjKf3aPA@jv|VLi z9kRD?v9&9rg)SKKSbTL=;<{Gh!PF(sgVf^_ZX>DFz@>qK#7EJhdiC$kPTZQTP@$z3 zkto!58wX*~;X$UkijS{J+FoOMI13BwvfPFNAywYC0}K(Nm!%!b+9ROK6Ly!9*bzZ6 zF#&T21?`NfdT1$H^z#YQdJDF71;)of5do+HW`fV5FCXUsEUh9L#LM!tC2eN=p1I zO5x099(9Uz5>Fu4fIp@MdrBm}t6LFD=mWi9}p_=P3o#!dWoI^Br+2X#wVa@_}T8ggWTF z{rz?i5epP_u~8_8>37d#ZEZm;+#1={K3`0KpprfQgL!pp4q8x{tkm>`I(_6q<~W1d zuV<+#8EPD3_QABmCnYyIpnDFaSDL2+f_yl1#MtE#JB@dhgc{s`{^A9oaeTkp*d~4L z8jxpXMY9Mvlj-GD*T(VUOwQr(`d|;yT_iAi`WQ_vDcq-~ z#=rb3R!FZp4cupl zPnfKVisOO2k>aOoXbvKxj_7cU7&SeNABn5VTCucDUl4_zz0!+BTD%L*V>)$p06C|p zw?=gAUFL5BwxCY6a%IumFPp_C7Au8gD49nt(W6$W55*`J4*>j1%{N*13WnTgAp zC&=vDZ<#c`?s<*<-8h_4oNe8+`h)kQIsi&?9u2XGnS23!-ud|6^{GnLHLk0aIkRnB zADJ1S9qL)@r3H9+SXS1aK+z#2Wa$h4avi*V%SCjspCJ>q?AsaNikuD(og=9R5rFBZc4JIt zvh^55H>q>ivBoLH%_a$EGmQsD2I&)3e!x{adMU-n zZ&+{AF)STA!p8h%sF7Rcp@CYlxc17qoIv=4U(epeJl9qIO~Zyp&~rMjtyX;d@$v9C zjnz<+oJi$Ds}FnA@1HQ@8ioKh;KR1mEO&B}X=ro7qYb?Eg@%*?X4-xwwv>LJI~ z<>4znRsa6^dcW2Hre!`|%r4WKaJ6+F$kA9FR8UI$z0qW!`@6r?LgeK#+f6#!XJ>6I zN{V3-#%(JrzSnOU@l4n@n`6RRqI?yV08r@Zb=)zZ`;r2drGtb~9c6@X-d+lija563 zglnQB^zSvXT7^bK;PO5PcqftYuSs^#j~guvP!@mrqFEqTY5~$a;iDWt4q{@8X;|2I z*>mwp00^3EX9~n!INBRXPcUEXtEpO)|A0N`IG@l9y^fnHz1T&oc~ouc zGNDoOo|#pKgeB`So=T2-35;4vNi$zoM}Va8^=k;%(vjwS)@&Mznps1sX~T;B1G-8b z(NKEk*QV8ZN)MdB00zaqa&%^`#5*fmiDrBive=sir}?QvN!s)9lsp|Enc)d9)g5im zMuHYe+*d$%V^!{ZZ?PEv*5Qt3?Xc5ZVH<1fXs)NgNnTXmS#lOnJ{f7DhHOk)C|mhwlc z+@%VBn*zPn$Q94HvoD%kvr75xrEFLGc}X=Hh_ms+fDr-(G)-1|z~1@Z-~F2`O}aA9 zmc4UR4!Pa^II!_HFg=d;vW;bh_p~3$MLiIQCm?4dgm6CXEBxM9=;@chC#^h63C2IwAgA>`zOgMxUXWUN!uU9J=rsI%VmZvZyb z579hDAhk-{rT=4-X;`WWHcBxu51S&&C zok2w(prXZc*kOmEVmC*-5=4>UqkyHY0tVW!{Rw%VZjUIN*gqg3%mHS0fMQZWFha3) z|4wiZQ~mx2-=0>3LO#I@e1Sujb0_E@Z8W@Pn)aS&<(j}jrrj7#i;>+gZkZ7sPuOK5 zWj|thQK2q??ylNX4Dx8-4|2N)L3f_!#aU49g75xMPt zJsUrhW`lqEXHL$*ddz6sVuM8~O~xfN>!^!jAC;wPjd`vR77#xG6UfUn7~Xy9+lRr6 zbl}3ZaYbGLCX+lf2i<*vG%JqJ}o{KM}3=jOt}A^mfY0vrCy-9HvHIA8HT{c(!?|Kn|Z=)q<1 zlq6MNOw4;9kFf!aM`~+p!P+*K;-6UWy)(t2KmQ_$8A37Uj!sv_S_;8i08Giifs9}9 zU%p4Mj$pnA-b>TqfrnjOFf|32F6*9@kHxsJ?blEBqBlp@HA=;veDv zi$4+jQU4bf+}pPp7f0$1?8B2i^cmKK0tlLet1ejf&iU?+K5 z4%!0UyE6!{VG3lCSWz_iOn~_wRQnh(`2_zX7}Y5t$WH7F8b{C{uEO9L5Zh;fD>pbT z0C);T2Chky&%HvJPlBDn0G4R5U0cdGz}F0XD^Evk&~&u4Fnz}Xym>$#4z9aQ@A(v- zXqH*RM6q%hj{{}^HCno$p#~n-C;+wKhF)Fm4op7y3%&4>U@|#?BA6A!ar4u5b2|jLUYLVbd~%=$BfkOQb#iiG z)C9#TxJa}g^g7$ylk?e5g8Nb|#a(>tRHnOT-~s2#+t%EC3mgzO(08+82~O7Qi5hM& zjmuMPQnWf-z<5Ym-<)0!%S9A zE>al~DF6xfkB+24fIRv#Sa=tlJ@^DAiAliz4a@>NK0m16U+{hZo(Gye1`U2F)|+iW zMupizNC+e*z8}Arz#{{*eBu-%a3&jv*}^l(M*!h~2Z{C77JgV>AU*!UzO51&$$$^Z z{;U|pokB;;#n42>nign*^Ld5>4fv@bz)c_PIssGmCu8A)>vZua;_85V>;m4lEF&$AwX+I5 z0W&6Wr|l%TE@22>yp3{|5CuVQUa&R+R6J1Ao818shDq!#YWx9Mb?iNGY`9g2i#BM$ zPu6g8sqnn3D<$Q3!X*yAgc_JG}=! z0wybzbgwyvK>BLzL7WXJ9r$pSCGQ5i0g&q?<-({4Hs9c?!tG`Q2KS#$-q`u2e9{O>;_3o($zb3cd)-;S!Wpxh8zYb z!GlfmZ9zeSK@dR$2R(fnoJ$laJFZ>d?|A|CLXAYj`e$R{uLu^|wAi_x`!B^6;2cM8 z^rB%R-W{F-G!c-M(?7oyrv^L9w|81-UY93e$DGXxxy`lTo{5_Z5?p0${i?tRb`SIG zq<70Nh=4Wa=O?$p$d@WSD!jn}bbM*ALST9!$*Xj6qe) zuT$G%s99uqndL{=MR>ga{{ErBKyMmVba?4UAe91N1;&5xuGkNLtrIQ3O2s|}^6KF# z$Ly1lm639jj(LjVVw2#)ZkS!+O919mwPwBa2(U6BzkHl93Sl*FX9Qvl#~Sl{yfZ__ zmZqlpGTbLj5{lR z7;B|36`DwBUodKJ)kH{$ER(gQLZKA#ox+r5q&-BIN|6>)p4U0g@+Zvm9M65saooR* zaec4rJU`3(bDpm^)`P+!LP1ICb#wFb)ayqTo9pt!aTJMroi@z*PP9f_kHcJI3_EJIew1w9|DUu zbX3wRjC0~|C;WJDX`MX*JSKKqfBE6Ai?oOH9)(M^a8;krb1qZQk^7B(X@@&|6VW?NGZ{NXTby6(US?K|_PH zGvU5g5!xGI%Ms$=;5WkY6#^}>d-y7I!L1>T_(r8&xE$1eCSzBcSKvbFr>_^Q9N*Ig zu^s;tiMK=2l>3*j7o^4)KFTGueZd>& zy}~#_DkUo(JANDrkrjhXeC~y#6fyaX^Xd5C$&9@I*wR;;>rqiX5^9y3ZKXT4tW zCAd`b0hYq5m_flhGxr3LvS1gltGROV{zp^yn**XAIc(|kOwP>F3mzK3M7RUKexPZ4 z?b(1WWS%R`^YQ&#s-&F zGx_V_rJ&|Q9@b4*~)1+0j003i$Gytb!wMM`Xs`Vzh}1gFu#9XLIPuqaH<$9zB|? zD-(3g&CMJ2owb$l>>5@-^^J%rlY@;o%IL-)k566L$jaYgVKyZi4`Sbc_3{BPU%b5j z>DG)Pii(Tothla(NsVK=i}~o_&h~bJmL=`Niqy{OD_@rseRREL+=?J8K@+DU95LnB1ryqNzm{qMQAhy|tY~vm# z_nN%YbBQ-*`0ACt>-g~DZ_3&E>AXw!KdehG;~Y*%u-bmd=?3-~XBLRb4}Nx-1l81- zdkWO9q6+n@ZtW|_IjIZR67Yqkr%Ij3JY3DW^Yh~APlnUgWb0~bzSz=VePX7+^zObH z9$A4nPFz;Ax3QU6wsm2G-fnFrZt1M`ey-S;;bd7UQVLbS!b8Xq3?j8pc5xg zR#`;9yx`yBB4Jf2-x@Yzhb2puK8o?zW&(%V+jAP`L4TE>ay&Xi&#J$cK-uc8`9x-2 zpID%xb`>xUTk(g6az|PIlLr28mvPV1@H5QA%mp@I4cuP5=Awz|HwBu4VF@d4cytQ7 z0Ib$cQ=H^d97TO~cO8on;*yqU)V5>qq4kD<*QMhChbCO$GSzDEi3iep*TWA!^qe`e ztK?>L|IC!jbPFFaJ%`bGD2uz8jbs|_)9e_GdF&#{hAPXmNhL1l-L@1OLdWKM@|9ce zKsKHVLZ<%yO!b-^UkTxKj5#e1;NV&JqhfAstPAZ-TbyO}fN-WdH=4*p%mq9LN02mj zBs6oFV@{V1=@Pt+V&ly@ht2kLcHw7ftcn&o2J`%SffWmSMMcF9pA>lyY2i>)-g71M zpbNVa1!ms3t&OPpf$&?KnJC%j$F-Y_&=KR=_(p`T)F|r8LswIK^SAFd zSXneU(_cX?@?PI;FXE9@N-N7DmBZGvH8hf`z!hdY%IvTu3VkE1*vhT|{&WOB4v-Y@ z`^UByudL+vP}UwyOw>NjF5~1uPS)sIPaDfK>VF+F8dRV!b5A&$oL25m5ARt^f`wO| z2f2r|&N7+(hDzQ7xV~rq=x2sBWW2-<(mm(HjIOl7&4J%L3HZ9i!yf1~AGo<`;=Hb= zWLby9W*={lOMhNN5pdl(`&QZt8lLE_+qMBZbK04J!8#qE?wf#Ct?Pq7J*ek9Vv12j zl~CX6g(;0N24X@x!p~V-ou#iaHKdE5u{M2S;@;ww3^2R%xn5W`;v4I49e0lU5Xr7L z2b5=(QRq#Q=mfEOeo5@O+x`PWp-v8*I?0{ub#iCN$b58Prm=n;kUi&uCbem{*cDx5 z>(JVHES+->o|&-M&`J)O(c!WcwCbNIMa+YN^u2?O-_qII>h<_s4R32|*OR`Pola+I zn#8b}R;^4q3YQ)6MIdn%)y_WS{qS_e_czo6o-E{W0F=_>5wREvD31Rr2o0=W8tmrg zMyNQXCby%QPeh{CTyWY+6T)Q0(<6&gPMr8VJ)c{KRrjwkF$U5nN1of-*>RWCm08zJ zh4sTa0dLd0YmOajHFKsje{`w6X-)!AYM$2Ci3OpmYbG34xXUr3b&WlDg5%_~BmY~X zDH!~>d(l6w-uy$-#y=}b{onXePrh+5d-uz0kHkMcJfoWW;beLtzmIU#z19ruc$L9a zq}RclC$ViUyJ6_gbu2A}2Ggxc+Gq zzprbz$THL4-aatrg!=&LdF-%=5VNeT=I+-|DOWde1;mJ7tSFRsw54#RQsH}~zk&>K z^qg8KteBS*i9bX>cvMlpWFUucW#w~>Vv za(#(<)-j5sUXwPdhkHI~yn`r;Bz)xa_5p5}Jm4#`BPBsKh2~ zS9OVwl6IM*87ZdR-n?<2YnV_g{w8f=uRQ$oO%4iG4RE6ul2bwL0@b(Rl;fzz63TNG z&mOuHRJfql>-{Oe|5M_m4xwcx?otSG{b3geHTwGmG&Q+s*TzO~ zt4{KW$yiRN;*ldGBYUY|S&@L%I1cE_P2|I*^R7>y8f1GZQ&~}o72aX=K8=sgKBfgc z82X9LNaBBTWBYdk!dmTDnnvU6k-q*rS_m`>N1hnMFw~+(_;%lo(4Av%9~&Lr{f9;G z5EYeR+~aAEs=3RG7(%{qy}Tjicb0H2l7Vq!Gs26-hkKSZ(VcB=w5=}$xRPy+vf*-N zBaJ~QO*((lGE%lU=r&snaXmgEVeDxD!F-G}7@|TlHS(!e95F)Bg305-q&~RnMM|HW z^BZI$9u!rW0|NSdx%!4$7qwJj&X;9JXKYXPPfS~Qw!-@@vY_vE?qN@;TFhdLM=KMC zQmIk2cyAjTJeEz=(J8)G7nVL{sSI0Ne!zg8aA!G;_aQavH&tV&{IJl`5g6L9G+&vo zq6)f|s**Z!MC->Te$`vDm!u)}?)_nb{$?uhAC`6>%jrAeG@o#>Ycf7B^jJF+re?gcCu6sux1l^dQtVlD%{WsT~~q=H@kF+PIDt zM0_P(4Etgt-g=hVZLGBdRRWx9Bhs0U1Bmhi_cu^Hp#VR??FS>o-RIktqh~r>ySltFbxsYJnSr5hz zaj~(X$lnKy)*Wu?`(o6{k!~D-lXkF@aDMW6x#sI3LxyCmERLt7uIw~GA4ZFlZboa; znv=M5)9O!rJC*seI1)9rZpHV6qMa{#Ce(q*@&#Nl!OPYO)q;iC6!l+L#&B2&k!zgG z0tff@_Ojl*aU%JhS1xRvj5W?Wijck9l8{}{EG#csGyG#nrI=(5A37#B^~FP)G**Jd zmVnqoNmr7k4o8C83U5av6BCmbYr(eOr@L1{LBRmneA1v9mX@(^BRKvlNeVY4+q3}S zsEuZf`ECs=7*$i>Li*Xx%#6 zDyFrwv-Hx#mHG%7x5|47$j#j-3tB=tV^fv{>g+BS^iWA~bLB;26@{|3ax&Tr68p-_ zL`e8Gli%l6S5v}Dtx&hUyiOGwqFi^()IH0#QZ#C&`#8|9o3Wg5@m@E~CmZwt@zIVm z#le1KFGMo2%HtetJTE2T=kK$*>a8;3`Idq1NHk1EV&me%iQhaem-;$~m+kdC1OC3` z5XHHkp=ya)w#FvyZ2ez&1++>XoX8AoD%U&vXpR1U@@80fvym+iArSuoWnfoVc`F*d z$e^Ssp*a+rKb^ndIB`O^|9i70~8)pt%$#ELiY^9Su z!$Z|;26b5S3ytGE#-;h2{@{FSA{rVy&_;QpTLD%n?@RWti(*ovZmpKkc&B?!;OFsc z%2Gd~QeYZ2)7tu{bIXiPOez=cFLepCke568qrx-sLd`{WqTLTW(L01`zmlKi2MuC= z6IybniSEJBuxWdp&tfO1_>5guK_1dkj933$HS3D^cWHGesL2I=q>io=_f@ByZa*M;^@Oy`8<1^;hjEOnDr;phA<>I`ZXx6yZmT9j@L?k3Ifs zEA?P2SdF1rND4ifF{g%Q;2PeuDe5Q2M|W8TZ`o7*H9v@p7RSSx|0SyvHc#kZfxZ`lxvE}&`Z{LFUNeO3XNS{ zk)HehHY*@Y-2~G{M5377jXVbtJiE(moeRDpdKRB4+}M1@6-QVwQYmqnP@1Lw2~W+2 z(qvcJ=^mR%AHgkok6j+QKA>r(gB8Z*wpS?6HG-S;x{qQwc!EU_%(SOnIJj!tt+R1( zpa;*%@Cp$UVlbe}>ZuLlmtFnT(QJwX49X{OhJhXR_9ygX9`2JcCgqkHQyknB&@0Jk zLKCNOFdRPd%N;Hw{x#o009M4rkO8iFXZ6G1ZU2QYG9ncJR!NS~F*Q_Gy~9UJY%dXc z?31R5c|Ii{D|#mk9m5qjej!6)S7C;&0b?^C@{*U=D`5}v(7{5fBAAt66{9%sZu(t9 z2po@xY(rDC${gQ%_8yZ=?pZ*rx65` zbW*f)6q+DY3Tj;zl`_V%Jct;VMtN>(sqLY8W;yvVocZ|v1xhNL_%RxCI99VbOHBR* zd|Z3{*tPEx1k`=@Kl;cNB+LEJQJPU7gsSkrcFIQl-T%N5l>ZMp b%@_HHw;!!+$StcDFU@w=+?nTQxNZFpNv*6~ diff --git a/docs/willy_wallaby_from_wasatch_files/figure-html/unnamed-chunk-134-1.png b/docs/willy_wallaby_from_wasatch_files/figure-html/unnamed-chunk-134-1.png index cfea55d320d3943753cca26387cfbd7caa0c15ff..32d413ed825524b4797c67b62b45ee89535fea0b 100644 GIT binary patch literal 71456 zcmeFZg;&&F)IE*~f(Qm6ARsCrEv=-Yh_rNrBHi7g0)ikQQc8CZ-7uh}bT8tzQ8$*TuhBf( zA-*Ull6mnnC;fw-J9wY9hzP3~G+r>pM8>ZoApCXTWZnAgR8$d3}R!3hdTxN3_4^g2dtr)ojsM>Z?vIx#|Ggl8{{P>Hgkcj$&4LTAIO?SBj|V>1lIRm65BFg~ddQM8qw{ z$7n2ZM11_@!RCx#?Rm%b`TL#P0=XMrQ{&;(s94QyI`l-Z z<<5+Rj7;aw&dz_IYqm0g=oCE8PPFVv7V{@zEfGJSLsSL1k}76wtG%M}(L=0bMA4sXB2tLAko3b&;j-~L}^N;0ROV~4; zc}~-zjkGxT%vH;`zHuIB!$li|+8+HYyF^siTg2q#deUW+K6`qSFsWFcoj8f?v3ULM z=n!4zdTcagCCCmWm3tDFL-bP3(pFh5{p4pTDk*| zwJN^bP|N*rt|jn_2P|zO|AUW_la9`G!b>0Ly;aps>Oafo;}yHGFHMb&YI3S+z07Jw zh7;9p*f;LM6TEh6+RMurv#8C>%flmKxvQ;)^2EYoz;l?q5YLW0J6Z`24i2NZa#r%b z?#t)TpM{kX>ZLiKJzqGLIiR)RU6|WY0W8|p{Dxiko!^JvuD*8Rx9}L?!-`2?diU>7 zm*!|lNl9AROAh~m=EsDqDaBGf;{Jx)p;ki?DV*q&sv-ZcYJ z9$sS58GGTOiWB{Y_eLnZrz$*Thlb0t2sN8Nb0_;J&>(h2qEk}(6?mU-Q7)3Pj3e) zpjwzJp2B|nCA~flqwMj#h=>Tic}>ke$|ZHUk}y%fJyPg&M;<~Jg6f;MK}WQhv^2L+ zLcDaEgruaTv^2}(#~YiQ*Kl#!drgwQ4Vsd&{%Wb17T=UI{QB+NH&}mPp8^2^fsl}p zw)RA#abtb`ShbrvPH;>NGiy*lfPm{k9$_LxDLxtJV`Cde_xl&(>#t*dERfZGCG0y@ z=M$B2ezM)kxO@(X;>X(*L`B8#Xo*lOz9WP`J~nps`rSPDQ)i@>#mJUPE|NBPEJL0y zCVhVPU(&264qEgVI_xYQ@Ho$m4nMJ*Cqa zZ5dlfOEn;%X3eTM8_6wdPD`rl`}?_gcm^f$ykjpSlq%o-x#+ZKVqh?sEEXc-Pi$1I z#(_{h?aNjx7C-47E4RTLw}aErNn=#bQRpjoJKh6WBj+|Xv^d{C#iFB2OO=YMKAew& zz4tCLJ^ae3c|8A1TA|uxkcDGc3S020)0VDm*#7D;hfY0C!_ItvmNI&x<~&0_4R<`* zcF=T*h1A4M>kgy)?eN9rKtDg{g%}6O=LAFMCr>u|m9+&N)}xnb0dC5lti(8=*v2#Y z1;1mxm4m|kR*B|xnAc^`ROA#woqP~tY-g#r=_E*kx;`mLF!JZmbi?v4-N|EgO@x8Y zmTo-Lae>ms8nFB2kk5AT0C)zl5oN;@4dn@hN z)cPt-02LByj$4=dbai!~^6@b~VA=`D)DQ?@6GK*Vtt?hj>GBVm5GNl$1+=%e#=#$wHnaHn7DRRh-TD)u5uKD3h zIJc9vST?KZ*>C{-8cy{i`i;|M;?`$pb?4X86*3oMUDhF=x9}w*e@9GU-@+321NzaD znJlDECli~UZc&w%)@@i}x1xU*ks=G{jf;yr%N#{}>3AQloORAN<-3{YM7pmRcG zl55LFRm951hR<^HIsFK0@cx8{HtJmruU+!-=5)kv?5keNnGzcF*18K#6hd-g2cM|p3rz0zer zC-I7f1^Po%pepKf#^U1Q7kqNHTbL5Hy|pzv8on6G6}t9={&;>~bu+5#N7gY4=>nL< zxQBcb$zwrDPHr+*^6HvvIVOLSf4;tf#W2#~PeMe~NUp=7-}cmbC*6BZqr!Fx4wy&& zMOMFbEPr|W*WB#NTsa(ke7CtU5;d$w<-_TGtWPHaJEj( zzBeBxSH}A(*f#}I{QC8anA>H(?H2@7RBoYOYj_NwLyJVqZr?WwKBcpscTY?wsx@Ur z(|Xs23q`fqop%gpwt7+|=7PMV9QIa+fpDr(YB`mn*Vl9IzPQiKjF5<6feq0t&}|+r zH1LmnmG_u8#(Zb7JGE|vWO=&5Up`HmOa9LZmwKZ_s<@_A?J7?K4XspUa+zgYnf=-b zJlr?58x(y=w;{;qX>>VY`kj#jDNz7Z6r+JVZnFMcN%DMV_da8QbdXkewK@YMN7<$dyQZs`2xAV;}!fw8DGQs3pu32At zu&Y^(`>CIC)6N}Eqn=lreLE0JtE;QCc0iuqYp7pHn}<3Pp=@boCHt+7=R@OlCvUb7 zjdzt(Qd1dFlK{YX)wX0WDn@-W?=3PYaiEZJ5-nZfhtG8r&E+F6>Kno5;g+C2g? z+RBK1A64Ue*Cvly?-x!+%-d04fpr=|A;g$c5Vr!@3DsJ36vusi;UBd@Xnp& z^xWR1Symfs>$Xw=(PKCvdiRF#-fHn#;E_LSRwrs|ruX6rbxsa8AwFxYX1?Ul0ve&a zlYEHkz8i8zPPiXSX>EI~)73!YFMEncSZ!~_f%}K8V*jduj(iqi7%WW3mr?WCjunq0e&*Wnd%gW1D z@2WeMvUIYh}0*5GI&~fjOQ(vSUAj2TeZLfhuTQd5o)l{wb);fVQ1cCipWYtz* zhCIU;>IsLbCcT*Oqh|4C>5-ec+TC_%RhZx=CfakVPP{KO!W=g&V@ z1ul(O^#stGHCMMdVpjvv*(!#8W99s@In&gdRczefvW3M|s@F=~U?ndpV7JsmBjSC2 zwqN7To$~a`+1Z)5((z(a&}@G+5S8ow3~7z;o;*I2hl*8YrReHf<%(lVMMZUZ--#@s zaR-%=n~N^9nt>RXUvzZDov&kbISy}aZ`Uj}9|!h+>EcB%y`#M~MvE6`LcWBxD3su{ zXT!Flva+%kEmgY%>WzozOf~wgj?JN`evIKnHP_lb-ngF~8Z^uUNoD3tE-K<|5W)c< z|E~Oa?}n=gp#1M!K(#-L16A-15;`mEQoQOp1yLv(M2UpP1*ik?ui9MMddk zE^Lh|e_gcYZHP8&4QJ*!m1XgI`^p-_YJElXSAd3>8uh-$P{y7rC!0g$0Bjh&^Ein( z+pF;4Y5JYS(1{FEKKtyGj#(hz*ROhpmP%|Vk}Km=32F$sgd>rs@-hO&b}LFwCO|6z z_UZ_PPVRf*T)VdPAGLOY^mHaKSD72Hj`sI&wdoX~+bsbuTw^&^3x$dmkn!x~5E?vT zo?)cRKISAyv|VMLGs;dW2?^ux);zy;_wE7Yw)!nO*?%lE)4qZJ=OHaq=rmghthdx` zw4^IiOG}Gcy(Cmf@ExyuvJ=tar^{Fppj14k+1S}Rne-$8d}65Srxz0w`vtHN&a5sc zC)ZL&(hxG5nq8KAOH4*a`CB0cDXCHO7l2vD*vPQ3VzW{1e{m_rmqI|vuM&%( zcxL@bP_PK+AC-!0+qW-*hxIG5k&iP_b&0&#PQdoSNZWfQjCwqS7{O zjbMe2@^C9r#2+yGDgoume3V6a5rMr^sdhTh`oSE{N{{oio>o4G^^?PeIH9wH8GxcW z4+9)({f=lO$;sZHp4luwbNR+6eifLQnO#^=j$xGUXyZvBg@g~T;S}=G_PmNed#!4T zW33@G6hP>XH*lcr&cQ*Rr%#89#>U5y&|5&3bDZ>oc@Il z>S!^EH>P7tT`@H&CSsymyxx|@ycJ+njVETP$VgsGE4E7a&*v+6#7s|W!nATR{G8dt zxbzjnWFM#5ND%-Surls&)A5R8*a|g-J>XGe?Z=N;y`)Pk{aGYrWJjA#RJp2!uY<*# zjX24SUgw;jEC>NqWz6M=IUI*WhuZiZipzDQmOpfE1F~$fKmB)Mq$8IQXj#1t!eaL2>HRHLTabhu^sg zIPbi@?x+7XH1v+h$Uv3Tw#WJL1mr^&^g{{-#{CRguT+AM+4XpH6r;G!+Rkx#R^W-_~S5-tV3c;?=GzA&x>fXf@4ZNqih>&Gv&4mVmStV~0(w`3s zL$zCFB1idH^fMb?`_+S;UKw>t^}}M1?RhDueOzb_pwIf1F#(9X)SJerzgO*kTG6Ne zVl#oc_2uMz*yay5fv-e14Pon1c`AO<{0|<~O3ndlw`!{$?Dz>gc}!mTZH=M%imqpe z3tL6ZPoExO2ne4w^cviytnz|_f(&8eC=Mr7l@kpa*@)V$+qa48RMpoZjsX14yW&6D zz08CKS%sYpYPDSS^7c?Ax^Fa^reE|U^bXfY_ve$j@dvzId_I#fHx)1d}k*paCWDVy8 zU!wvkD=9)rT_opeRyxQ;Cwdwh8q(6zvYm0QX^w=#ch9dddT*WmzVwl%uR-|&mV&R% zLPxW<;C35FYRwi=63=jbwdC;VsE?P|Ux~JfmyAL{m>DI|(*R$pY z;C8w=0c~4i<5dpDTrH%>>58_;$)>2W@iJ@8N#bAA_OYDo$-O8}qX!Y0!(NRLfG z39*#=%S;0Ck9FR`=mZS2y{!JVwEH&*2zX*PvX{|WZ(%#VPUawx@DL>c2MO$;ArzS3 z^*h@}p-@3lDUN^3`EMQg@pC~(5LR=tUWsPWsxovn;&qK7Oa!LP!tPf+CyYZv5&=4n z)Xyg`w9=K-k6e`8Z>Zw7}`w(z`|``xrjnmCORAeu)ZJpTRP#+ z=rIgk{jGpuT6!qMm_t|`%1@dWKUpp8oDFh+!%5Z-kaox0?P!Osw_s_)Y3}sX<&U?r zl9yJnCjY!Ynq1kKs*~uJhK^t^;NalE%g2X&P^;Wp#EL-b4iAAVM)mhpEzO93>ncXRadN=$dm> zS^3<*(o&#p^*@xja|4^oHGW7rgjNcGSwH)GXs9mlnGT>P(c0trPgKy=B|`dPnl8ty zlvtU50ISd)wj8i?aTP;bBqMm34BP-Dp-yNM-oA|=O70u4VYJrv#=UXFeg2mLv@7qJ zRLd-v6C>U42Ih*gHI zK7}kLv&F6i&}c%IGNBsLJ4Z>xy6veRa?C;@rlzI_HUJ;Bx3|~P++6OoJrAG5!ND=7 zj_Id~zbK&AvJ^7mL3h{3)wO>9`gI680!A-k2y5x=>}+nPXJhN$d5TAUa(a5c9e4g$ z;tc-8GX>9R29|gZ--cVq1U|`@@1Vb3Z!-sW4pXVI5TdH%}QJ;A3UeCzLfTZW5@{|QB zEGaQjyw5%Vj*U6?)vLsnU!~Kwx3@zq`qO0}I(ZjN05O+;Sn9IBF6@vG{6l!fq*#qU zKVM^^BPPHh*&jc5bxmDoeXN{j#2XZ#T~L!!5pFx(qQoVW>+39NnfH&IxlA5!)5tfr z`zr=JOyXay15M=4>6JG1YS(h3hoC$wC4@~lEl^|#H@E^bH$o%EAV}JNaVTBS{wyN}SSHytr<9@4DX_1RunQU~x7kQvg`ElBD8Y>zXQ%rU z%pQ-*szxIu;{;DfEozv%cjl!cIfrXJJc1|%B$a{K_wJ~g$gi)jLyu-OR%UhW;%OSY z7JHaWgDQr9b72+U{;E=%joIsBYZ2p_Cmd~-y zA&HBN_1Jr7$F||QC_LX!0T=rW^6LZy6U8RVhK7cclHcOuv;h@i3s_EL2Ud0;zW(;T z^EQju0j$@!FP+*;rO2>*0HGc*2a3E(iK*(dH4}*2E%(WsoE-SPkU^%B&(S+L>(HY? zsRj*bc`$bZ7!fRN2h^?f;}0~(6MaCVBch{CU6)o@PnxJae%6rCHMh2sKede4%H;{a zeEITSs)wZ;cTAz`Z%#KfM!+GaY7V!e`yNO0IRK#9Z~XwBQ;99pK4HiC6h2EDzzcK+ zfFR&jE-o$(4*OT}$sqdPcg4T(PE`cCs$x%qK}hHvP6E=U(s|eT_3PLA`l+6to<2T4 zad8;&9kgQX-Wmx%ohNrY;)H4-(ZFE zM%CQmhz4EWVnT3fEjcZ%vb_B8@X!v>O(st8!i5VB>nsl*Ov5s(K$Zlx19~{%{XC#8 z^`4tB%mcHV4+c|m<8AMfE$CQk2qovG$w^42$(b4Xa#?UGc&xzFy%{2i1AgJ;Mm*D9{67dA!F=F6&!70I^J~=so_3eXu1Kvc8YW4-J z^y$0xH~z;-C2{`u;{Tbf|2GZjRYpE66>VGF9l*-0-X#V;LdgJE>Qd5S%56% zIBqkb*!>KA-~C=*JjYTiF&)OV&Hth-b1uJ$GHaC(>vv=zl@O}PsQ2c>*9!LciF_`f zT}bXW^df|6g6aG8nqcyPK^rVZx;R&B01A5fR0Vq`talVpsfmQ|+_{7KTYi_%KJ^zG zbYj4}APF8O3=O3}&UNc+Vc|BpMV0JuKGsJ&h;yGWUzF_c5Cy!zBU`gUq=?Rc9N)_@WHBtjSsjJ11#9*5B2;YDG!Zn5z>NTcjC>`fkf76(*S| zX6@Q7~2emUTuNR!9d| z$_EPb++(?OhYAnt0;ulw-MAeo^>fVH+U|H+4__8XR__K$rr^IRjo-MdihISGgA65+ zJ|RAduaX8${ZX{6!}p4vtT%K`t{Nm(IXH98EUz?akqEo)4k?%X4CYOzFjmTCUOAYlyG>?YtR#o1bN+Fc& z7Qf3!upF|me8I)Zp6<8Fa&VlSk~1x1HJ=#&Tq*4PeDL?;sv&iKqi?`m%x~ouX|p%C zgJ@lj)@1LHc}*D~^#|M}uOT?sN*eT&+xHk*irOOM z&Te#=AnFZPNs>h#We%2Ep%|6QlqBDA>wcJhZS>xGWl4^1PyJUZONyrRhQ0BMf-uo4 z5r5WNc8ttg_LRm;94n6P`6~Zy20j-4SO11w-a03VF5pt@ckVq2sWaisymq^4Z`EpV zt7m^DLitno@W;w44*Lyjw}s32%J%UW`JNi)DOA(z^j<3I#TK*aoS$$w*P>An_U-3U zqMMM&S@qq>i^o7dmbgW|uSSe?)p5>(50&%mLKe#Lp}On~c<{Z*!B~L~O3I@+4{5WW zSL*auN~)uYrA9ks)b+hDWcY@k;pgx;HiywbKh>?SUlL=KR*+@PhYWKvacSGj`i6N6 z%Kq$vcSPI#j6G#8w8vX*b&5gO#BuUp#fz7G9neA~j9KSc`MNNU(+yTV zG}vG+j*EjiQfA|7@G+OpLw_kt_Ynxknb1PNR>2EDPRxjVSz)*r&cR0TGwkTm=SIue#Yu-6q zHTOtl#fIb~i4Yny!6%mSI*P95C!sU)fix5xuS|S%MoYEU8b8Z!X!KTDe1#lC)?!~SmRRCzy5?A)0vY}oS^a!{<3i)g^@_0fj5F3VO1A0=w=&Z% z#^R<^TF$HoaPbV{P=6U9Kl1H-D2yW`+F9TglhRC9v#ek>8_=7$j0%aoCneCRCc?b#2?i^{Ym{j!L5nlBj&VQ)yLzoz%owg~}w4XI9Gm$k@e)LWl-+cXmf6F{)?s?1pHMA`(Gf%s&?8^$|>eE6^*IumpA(LhDF4N zv(};uWfDn=jwm$h}oWSVYNZ=@0Z(* z59T0P+r9vkbN#i?g!Rc=+ckIy6jM_RA6deGVFYu^QR z8il%Ld=BCPhpY*R^^_us;mB2ie3#f8&Lb68;W}xgS4bszRbf$S=E1tLa9~hz*lop;IO$X&hnsE3nw@%rNsQW zKa*5CMto~4)9;yUEVs5gBn zw&l~LsOFUqq4)YZd2kM7~PoN^>f1=$!uIZgT!G!rTy||q31rb$$1{B71exv1Vhl1p~aJ3@!`bQ>FV znx&SOvx9@@p_F*C>_e&D&a2fX#OUk3)|>&VAyb*HLC(S6d5!Yg0Y zxF()5;9)ZEcU_h$M|-RQ{eZF5+1Rl^>%8k}vAT1`(e5CJ)Izw+{r~F7dpt%H{1IkZ zbuU7DhDtJNu4LJ{SxC;B7&~>7E^~j>p70Dw_O16B4rJ~;8aPwylUjYxZOw0(!?)74 z0+lOApP?d{8=wN4zkY51inQ8q{KaR`a*Xy&)FmFP+0TabwhP2>W_&V)t(KdJ*z^ny)70>l>TH(I75Y_wgasEtPQ+7Ok~6rBkBS?o}M9 z^spJ-a(Red{1QOHVSPN~;QNDu2g%-%{l+b=2y$agvNHYS0*|wVo1%F*g!sQ{4*T2D z&Pm$Z+s2!`qghv2KGBY6ee80*;fD-6Ybn&LDtK zsjTYV+-?{VVXVe7vhQHbk=m=qzlQUUhm)7u71-gM|6y8v+c&G(4eyp7-LYzEH&w3TE4#HAAK%^K`}SfTUah)q)+r7<9}%R3BBh$Lz|M zH=*W9U`!W^>d%j#V<-Bn(zaCGXYs=K+a&lHyq z;8O4HO_;owaX}_*WL!>OSR*QVZ5Bo>Xm)OtP`4#vfp|c)h*ynwEUo6A^T(#q8X^j_ z-iZF!_pU%py)2VJEu+!{)F7%$$+9&+I1@ux+_3@4esk z`IW&PzkWK3;8{oEND(0FWA*^;)Kr}9GL~r7koM|3C6ZG5c0SADx@eHvac;)kHmhLJ zj;J}?bMD;=b=mh(yS{gWTQiGeQ4)KW(Sj=OSj8EuI@l$Nve^0V|DL(Ve`k*Td@(F8 z4#c&-$;sM6hvuRD^6h!Or@UlaTMSG^3=AxE>#usQQG12N^vYk5`IVtsSN@vhoTXj0- z1=6VTpUmCf>botvSFDDfP5fz#bw(+i2}MVL$BVrIQD zOJ^)JWx8-LlCpx=b+>so#QRGhLZ7Et9Ko|FgxoMnr7e_sP6f3!Jv+U1DI!8!JvKUG zHQhe6aY)_6<lclwe?3l&61z?0&?#Dym5zX9f!v}D(f#(}Cl?On zeAN1;G7oqH8Km`|+!1Vh==UQ$+#Y1yg9AqxcZ2@O#hF-;RG%i2*KQdszv`2%)_?ue z2k7_z^-tB+)gbtTPL;EYQNy3*2bnu5;B@~*2GeZ+7Ro=pw#0@;SGXChMrwsZd}Re7 zlAG%W_SGi(YfP3t7qv($`{c&j*()VDI15k+DBG(K57!jtKcePIms>LP6?mt&6dn%3>J=yy!d z_;#{6nOt)zkU0h#xeIFOq28vp8k1&3oBH_qqPdt94|eIWnL&{0(;izzdSNb`GV%RJ zs;eP%GN`=wbuvPyoc7t}&u&sZu}x6!G2{d>OdW<_z=08jc|KoXoEZkB*7+Z7s>@fe z?k)8$z|_D!FfKqpin6G70;^E`Iw&w$CDYa;di25eHkK=jY0CM#;}l~7r#%wm{wOs* zw~S^9zPOR$G-)b2KDrMHr6K#<0kP*cU2$8hnmDpqY6yZ+_eIkM(~)X_WNRuaik3TS zQRh79!p%O0-IaE2kJ8;r@7G++x+5*-jO67!?BPQhtCZU=rf#{r+V)3V85J_SnVKi+ zg?`YI3`W-2>rMs>7w-3X8_!!R#_$AYe&pLYCnP5320;%-3&e=3sG7rDvsU|62B_(V@|&oKNl@?sClyeg5Sm8m$E5zy-G*I!KBK z>YD2YhX>*NN2)kk=HqqCu%cpYRGkxc?Bz2HnBZE5AxLTIz@<#^2kRRfrNNiNnOa*L z2iX8dZ@zeYgNpxzi%Suw8w@GXxRq?{MbhgpyPr7b8;7Y>YgFy2T|tv#Vcq$9new;i z!j$ABY5lqd#8h~}{@&WC5QGkh{7BoSo*y(4Jv}{`;pdYRS89wixd{q%U$)cvpC7Zc zvtaJ8j+UZ3;ctM6-t>)B{+C63LCmvK#XBg-sivMF#*O1 z0R`_Pue*Y-plvq61PUkUo-oF+0V84HpTcnO($W&R%`o?p#pXgHd~kS((5^v^j;iBq z!iOj;Z76>FbDvg**cq9wB$b@%H(bcbcOUZsj(OjIPdLoCky3lb#EiqEudbc1|9lJL z`cf~t(`OSRW*}YmWqv|wX=z_E#w=TwKDRuYBiS6(uAR_c9i>}cYuK9mEcj3JdnaO> zP=f~DM%ecVjHOIPiTq4(AbW)0ym^x_O4j<}!v{B4*KDGmg*XpEuxsV6%YBC-7O$sn zM@7XC=#{i$f3m&Ap%y%rNISX(=dv28I^DKR@?}4Ic2S6`3(M2HY0*l@YwTGU$JL z+%SNcUP@zR!rXGfa{S)yeO3(?_cdzsv9kBhO~>%yJ+&~+hC4o>@@H_chn|Kh-*mX3 zNX`hx1^PYN9@4&M(s&GoxL;WqK`e}8VZmk+PADuH!?3l@Nc-QL{to^ZLV!=ew*o zZ#^&{zjf~M-@O$*;fo#Sf-VUJptoHBBYUqjxLCtN17}~3&#qE)nG`Qo?9bd zrvdC9FG9@a0kzXEip~#B^%gbPNJ!Wm)`eQu)*2E;gmE^T=^jug$O%K|g7p!*LrEfQPoZ0mJp4_cJM*GL#pBvFa*2lkiz$noctxEa_>I&9xPwyA2 zXWZ{=y?D_U*HPnAieCgwd|4)3TWE%T)9;Jqe1=luKzy+p_=9ZT; zm9k*U3FWpw4%0goWo0l|umGEx8q#)nbaV#m>rvPs2+Phse%Q~faz9g%u}Zq`i%!p@ zghBu25RvAzI?O&{odoc0i2pag6hSDQJIEMoWHDTz=jeEVna+498d%@dGzcyTI10H{ zTYhl6gXW6*FTtZXk7)fmVj6sJ&S$nq(+_0dYzp$XTd7Z%B_g_qd0&3ZH;#I4n7e(QV+wiD*A91BI9C}DwTAFdE<)muW&dsP5vMo$p|rLD(9vXVzzVgk1&aI zXk#e-10ikrP`E#F(87v*#A{2KpRaRvvK4e4^AVJ9$3BUET^pqy?x7 z_IYZ)=b?yZEuv~;o7wsF)pXD>@~iBNM?5cL2N*t0{3XiKix{ymkNY0gHIlX)Sk2c$ zpC`9coHV%mZDm(ug~H`XMXyQjqTEnBf7BYfX5zO!B<60v z4niOFENydmbPjbGMZov{W!^y%P| z+ohxMroc0jvzo(*d4oH;KXd##1J3u3O0#N}rSN|3>HG8aXZqyiu4jbGp9>yZP}lE> zrV0^lDCqgflYwN?n#-(hW7ED`QYgxO&nS$W>SNv3edx9K4KX;7}G z&iTfQ!AS!U8$&EY1~sH+`ghb3Fh<@0!>HX`NG(J*}D%<^j6TU(RKl+YOl0(YmYv1WT(FX-Jx`&_20g2 zFDB`r_=CfhTH2$y9=d?sNkHo$x^J)|r0$2$^V&3&IM(9iWJKMB%efs@hE$DY;=$%& z{6Z2!u|4eFnLGy!s-|a$s6k*agIeC9Y2E#FXyvft1fAot{|Z6p+_(H*qY}}{9rI$r zaKW(Q`&(XU??l{s%%yUPq+M*D;lNRzl%#Mx68g`rZ>i*T}o}3va|p39$9w zzpkx)s22NsFjJ}3?@xU60q?PU?;JT0J`ktwdQHWIpzo1cE>gRaO<#EZ??#_OWT8XH zYkg5%>XD-&vqhBU8zi;M%pNW^w}@$h`9+UsYXwCcx-oR^3gk)_L&s?-)-IB0$!8h; zCTsos+unE?BwIyJIX7~sQPhns%00Qd#JbMUoVd14X}x?~0&%4X>})KzZ8o#ik>+;C z9+LO7xeOl`KH%W`b)o+fh%~t4$H4wMc>~fy(w0203j4xo8mgreCVq0c2>ltgsJn1y zmTXoBts{m7f)G%1nDCL_LMrxH)!rHt5zB$^k=)M_?h7= zz4&ie$YwU|I=LN{aJqq`lDI_ho1wIuJ!s}=!X8U;p<~x$i{d#J+Z1fx@=2NsvI+8N zWg4BIAJ&pLp+Le^_$GY*=NAD82Fsdh`#l{J=WbgQp#736KO24~AdmhPG~%YvUaeNm z;xRSnEi`d5{M0cvsIapVfub0pn;}Fvi3m$yqdRg_dk%vO)k)iKq+zs#3&0`WD7D0!FXwHgH;^`yD zB_3)0e43|~YPUR=o{m0)U=Jhcbbsu1u)CL=daa6xz<&<$hhb3)1GEY^5HA>Mkb%Zc ze_nWc6ZWrWLDQ&?lfbi>LiY=iQmyW0^nN%@uTi)9ExOd*qxpkGIDbifb4T-!a7Zs$ z?GFPqM(#ahEp;b#F*F6V>T-J-p{YqrbtJf}wd|6HIz|_tho88G{{Ae17dyP)R$b<{ zRWCPk{DV>A2)!KlGvaX=JY^+cnQ~!`kWz?0(jRy+d~cq64@D0N@*TG!pD4#7@u*qS zG-F*|w-!B|FMJUR|@;oUGYngX$~ zfq{`S{8sVRtjQY~9r^9Ey{0vpP8kwme$@pr0#SktOy)<&o4tetPk!-!eybFFfecJvJOqIP$)5dA>sjhAyI$o2N}4xu?noT``h^ z!)EHGMJGk!TFeu;?uPXPp$`cr&v-@v0L*3vp~omptj$ldL(Px2Qgxke2VUfnW* zOX9$7@B*HVL~m41K~Ozo=VNDM-9*ZxPrGKYsq2ps6CO3Y$3C}i1U!5C8fR10r*k%` z-UxrB8(VaurHaN2#~Pn(+|t%2cOG!8tEh|0{N%grcPpl)q4Y~hC40lhX$2hSyW_iP z_cAMwfw_*en7H)aR}X_;EBEHEe;X!^q&xx$XFBmnwms1}F|O#7Y?F{|kq~c_$X=4` zUG_;yhIsk!&N1WP-Z?QD91@WfIlHZuN36qlojeXIZsNbbi{~=1zBE4>p3)510V>|> z4?&Ge*K(6Any$6eW|vh9qZ=rXjaC0~) zAM+?chB+{LK5@0#C8vI+<~}l0OsPHop?Igy)~)m?=7d+EcqM~zNqK!~-FqD0L^}zo zh>EX0uRW(@#<*oKSMa&Q_j7uW4;wr8hllBS`HpnA#5|v*d$qXAPx!cj{Q+v%K~>#C zV|}H^_!%pYo6|}6CU^?kZ;*KmreT`|5t35#(~SP!gvMb#s%DcQ=KOvnN>*-{=B7dFSk{~itJ=b3u&12-1$~dxh<&Cgdo^ZZQia1=y zh7cxE$W&`k(KPRRCRH{7efts>8{AmE|`E}F@CdC85ct3`;%YqDiEs)He4KlGR z;$!#5!0oA?pO0jc8i;XmaE)R$Ivn6*b9wrFgOq zTlBoa5(nQ-(DJ!j2O2g0^NI4XwWsGxm>0TAMU_SIxuuZVIEpiOv@1TrkGc!wtMFR~ z6G`qE@j$7{gm+LavoAI;1Z{NmR{nemFbFM6+S!qw&7a^Ikl+j0u0I<0CXjA-9ad(l{Ehw2)I( z*q05z>E{h3ix-Y9{_5(4llE9OKoOL$`-4JZcy<3f3vi#h9%{QV4qw=K-}j8vsC32U zLG`LL6n)kym;7VV%H!_&mF)oU0bZuLZZf8S1!`lXh5YH;QnyM5#@5MIjjj^2e%8S& zSC_Ml_tqCyd6L?%a}eEeD$75w;ykO;D2*Jiim3v9Tc_YdM{@MiS6d;$j@4Ih^=g>9oI`U!B~0#EU;i%Fra4=cBg`+20#G&+h8Z0r zMrm8KZaIiLG!Sa*g9jHYE=YsXxIs$(&^yuAXw}LiVqV~U(Ly)3#A2nl1R*oZqgEA+ zXhsoLn*5o&0%EDN=NY(!_F2PHk(1;6+Z2}>7?mTUB?KJqum9ey9E5V1SeTTuewF5j zHLHl)Pj_eM{IHbZz25Z{G1h>G=TTcWn>Gr3;^#j@L-_XB-{1k4^HbClVwxKyYh&!4 z+kv}zR07ULFm&SEq>dD-} zore5ft39K|feqOj(K`0pz52EMWQv(k7jQy|1zX>xRZkrMZbMjvz3tCbVr1aUOvlwr z=2)ynf5D@Mfdc%obLpmRJC&9sE|tVk?aOeJk+u3@8g3VP5!x+}c~pyJ^#_u2H2#EA z33FxxaE%ALMbP@f1(b&TEUIsA&wo8V7D;*?Mgh(I^RObeY4`I(Zb7r?wP-wq4gH#2 zAUBa+i3iS}<(7|Fgu#sla`RpXKS~FS{JFU-f%|07(uwY4m4f#`SgjlLW-3Zny=s=33pn^Y5mwF-QCLD8&ETPan_ta{XXB> zB=7B%Ht~E)N(!@T!7Ch?6%T-WVrI2o$00N@6{w0cwX#wP6P1{8c(^)w_nCDvp-z9M zA_zPK{-lbUYa>u9skDc!F!W3e6LNUqI+BDTB}`Jh!7^?6xyrR^_#l9xlv?NI7E*UR zEQppej2Pw(9=A$MNRjQ1_ksLQTw;mo0G{L@)(hWm^%}d37qvV)A*KfYL`pg@D}{T; zV4zm#y2rp%>lYvP8-;zpeZD?ZuB77Og7m%+S_?WP%mm<2ulDaHkq2;xt8eD1Ah}f1 zIj^0-SY}~ip*JDiAfgm>ErY90gthOUtfG+WIPaNm3QPWC7s0UtH5slLsxj*-I29XT zqkeUfgb6g6YX~h9$t%z%>Zi^pxd#o{X3%?;rB#-4*c+(YFQcbf1MK2bnw{4=2dAy4 z{W6`HXGR8|R9@PNy`o9zHuJRaX7~7w!7$l~++$uC~yJDHwR=zzB69nXnUo z@$!0u2f)YT_AQSK^Y1WVJ#Lw*uj|>EgsB5BN1vY=8MiWDeDX}5le+0W@wa%4j~adO z^caU)nl{c==lC{bPOfj~>G47&Dn6(s@Mvo*;^AJ<-#%Sj%s7IYd(s36O+Di)oZz(f zD0>~XK<*|L_oq`EWpy*(_DlWyd$LlBo}Bl%9(r6LL(Y$`ObAw+fpnfJ-A^jf8Ow_!ja58p%qR7f18Kf)MtF#>&0& zI7a#z^OanlMTn}k(k$(8+E&R0|DJ0#DoNNq3TPC^MapU`@YVNS)g(Mp)H{11x%Gk;d*bV_& zA(3$m~p@Km>@`9?JbK-$aRmli0b5#bk@r1%+q z*C|?r^zKv3;eNd-hdQX2=(6$;2In2I(y_9tvV&+X=m>JnwUIyw`E=}Ys8J^??SdEF zXs^gv+gNf)mzF%WwSv$K(BjBy&xYw9%3YV8!`GeXw%c@w7ssfK{dvvk!TO}H9r2Y` z`(HRMqSaEgMx2Zu?t6SY=Y#<#n~Vi0yB`WZd|9QKo}U~5ml#|fxt!6zNLe6ma5nS} zvbZ(mC_NJrr>~9iV#o;Ndr!1+pi^i*ie1rM)oelMh2|8?RcZeXo7awukH1gtrIjlq zCdNig$?y0U2Q%3^BDY5if)X#UT2?|&TTDj{~UU` zN{*wV)x!AWi`X%qzR4tM7GgsM)b4c+(f&koQ~EP|dgrVwPmRqC^hz@)80IL>G~Ct) zOw?-vPPrMl7_uH^`6vfTz2rfncJ0&|)o+%Ex-5m*#63Z>|1Rb*vUC}{0w(Am)v@5;BK&NJFFS~wUC$=eykbE>WLA~9dV1hG4NBZLQkeAZ0+uv1|ERHbhm ze>YkGGR9+gb^tfhX(fN&f8gZgq+VhgEF_aj5)97GV{GwxQI4q$YQ3dxc-O8%8eS5AOIZKH+q zyr$Y`nhUGQ>ist(!9|Hh**#1oCJ{=~h>Q0w32O?wIPJ1Gy2v|}mW>Yn;pwC4E6Pt2 z6=e(v^OxePj8Qkr4h>4_DJ!b28+iF_MQ=fIsZZW4&iTpYAM{LhO{fb!3+uYvI?YS( zggBWRo^|)S!9S%D%DG#9)>PP3y%M}Gw)Uk~`*s^#W~MVe^MlwPc&^gAjkOrODz{aDHc1J)ENCubsOAYlN@nE zOahqh+iZ~EdZPOgjl~$C82{m&1sc^tWw>0cWsvJ?*l%5IwOg>gyx#AYcm8f>z(MiX z{s&nrp1zRO#KatnA9usT!r&ibXlQJ7c5-q%-5rDheP_6pP<^z_coTD52+Y<{bGT*= z_XcHjbP^GWH7E*5WWUV8ALgZj|BtdakH&hB-iAq~L3Jo(E>R(r$V>wYk<2AjGKb7l zhGvl*Q%Z&g^B59BG)U&mnU$GLndkTVx^;ffv)<=j>wWHZ{y3}KJ$=8Qz4x`Rz4x_~ zTL6gJFi&HsPv&D%eNsSxgHv|4nVjrRHl`Q&`T2M3@WTsr-qVpC_9JGkSC(do`5#ok z%tzbQ;}}@$2kd|kX#L;5eT!MK`MEi+0Z9poo0yMKi2Qhm67fa~Vly%k?(?QKuC|{K zt(K?`^9N5nk8~M~hnU0jcOoJpH!(dU*5*5R3?du+TwK)t7zj|1tY-daRXnUPuP%>* zpPkXS7vm9+A3bV?ItL;Aa{0`S6#Vsg&2wS0{)18`?ju?^Cj2V6m7C*ihjrn z-&5jK$>QjP7YeH)r9Cj4fr=U?*qSoUL!8Y|?!0Wl%X{&Rb52yu#rM02KQ*DLOFHR+ zc}rAG=Vxaju`wOFvWRihsE`mX3_Ma$P+;sQA-@~_-b|u4*OnAy_jL{N#K#8&*%`>L z-jL&N__~;5g;q#}sp3WA(G1lCD+hV}RL<;>RQWv)J&6i@7Yoi|^(3m^A=eH3l1QoD4{ z$kORdR-yfNXQRKuRq5Qgddr5noB3^0pFl?ajJ_eQp@D;@dkUAYT$jP<&hA~ieltrX z)gOkRZ8A669wU1$H`JrP@uOo=KF>&b#=7uz9Ya%ky7O?;N}?h?M#gQ|+wz98`i14y{z3}-kDCC!O#~E56lU-+w;-yc&DVdeIHf zVn3`1sP7)s>$|;ktKQ&bO`t_YotabWbVFFV1VbRp&fmLk7DohYglZHo%NojNE;G&k z`4!S|aY+73V2nf9V4Q7|qc1yqyli}Rd^LMbeTF%Fhf&Uf1Oa2u#-XX6o*rQxFV>1} z#2FNn6!WcYOq=F)xMZQrli9y*-d9UF^xdLUPjNi8TE&I1&B<~@qy74s`Ef}HUyHmx z$t}Z?EMhV3-*nILUZ8mEr|xxvbPX5H3BUHF=1|m1O9`FWD7zEL};tk`+&*tsvg$(8AiP8J~8NyV5XRhb?Q_ z2LnG$lLdW?%d?fmSyFH6^^C3j*{u`ns-AH1VYJoB+Qg(E)8wySy$YJ-V~5Lwl7a49 zlk*scLb8SPY8Inf{cU#ED>yhRM&ZJuKoG}recmm!wSB*IJ@5T_tM5MF_MPziaj{8? zak2k_vGgwg$du*D=C;%erb&lXHC(f%OHW+Aay66X&(xfFfCKYcQK$T|_u-X%-?-vn zFJ}tAG1BFi)%Gpf^~C;Y{lPP*Wyh%dudT{c^B=E2`V9&SSFpqioEw&Jx81vNQq%NF zAeg)A)J)qU>J$4eS#YQB+ry4#tl^stb0d0f#`Y2R`-jT=jzmV5h;=2|PQRP69IBaq z`2E16t!2@4(aW%eoeT$T6m2P(DL%D*F24H+SVJiPxwXEhVg8d1q8pO^bXIDEBWo0E zSnMD34(r^{y|-frbILD!OWH`Cd+NWH+H;SZ8lQAOc~5=4$VRGdyPF$hQ)%8<1zPgH zmu#2^*)dc`W!Q0PzPIQxzvi%7h?Q~0sQEG7wjf;>gKIOdGuV=SbPupcPA}D!RSs#6 z-VcQvcgc5$j0yWybMsuNBCtwwr=|_`^kiaW6p8c{=E9?*x)HJ5f8L|F6VuGIa#&pK zp0(Dy14F84o3ZZTp*(k2VlO+aV)#A{t;YTd__4thR=H*khZp!p za_0`9l4|_)m^*vqdhrbfwpHRwZVOuZ*NU2GBf}Orv?6ul-AV>HN9{a}Uvv~I49nk> z=aEz~wmL?#axyh}FMF;H|ceV2aORa3qp&tT#m=kkE>__p*b)g`il zpAS^!nH#my4{D9)=$7ZqUY+zQ4OlOi=%L@1YdGpRvK05YO2hbi{|{VX*=uk`3gjO~ zI+Sm^=MB7{39;eVEzRd(N5zYK1XuLg$pXLy!`?;Y?TImtuVG&N+Ha1_pxFe1guRI; z1B+Kfkk+46W_fXT_T+FKNAqeW)9afu`u$?qG`72B{ss!9G?sQcGAPS=uX*UVxU0i< z+ihxc)m;-^D}N#jl5+D>YB<>@_sdmXTp>Z0M7Ick!nKptXtaK4Tr0Ej&}*JzFyP)NZ+rzxOkU4?JbqA+KE^0 zEK+i51qX%X?Oyuz=V;lAL#Gu$xI@aY+(y6OJ#C5)JG3TJm3<;&k_z=p7I#kg4zzVt zOV@F#GtBcP9@MdRD0-at0`@F5)F2h<4aPe$K z-0Z!r^g(wpWca2x0=(h#BL?@qzI^#IIjUcw|NQ$~F7Xo113K@C7LKx}e+o{Bs@(6O z_1h@)f&45}=G{rH#ndRO z+xuu6OXB*@fIk7K#lC`yGxC!J?uHW zMC`S2KYyRzcH7)EhL5~%;`w>5Q-4M>4>(sfhJ2v&U;ePfX0?<(pf)(6`nukhj$AcJzB5+K<+U$*a*h==>oaBxq|f;OzJP|pR{^-+{<^n zGvZOqNw@!T^O1!6+cg;z?fpF%4K-p8n#)XRj~RUZ+EAH1`X_rxy*K`*+wwc}7S_!! zU-Ik+c_yX5XdQj0udVs?($kvbpLCy}8fVz$!>92wBz##<5`Q zQco|$ipp&B`yt_tVXtj*1$7Rz&zoTYE1t+h{fAA%V;E|4y+kgE^NyrBb=CU<6&*OL z94ggwqxlMTwW)pLeDa?+Q-|N_m?uY?E!2%fwYBWb(l5;kO09BCcSxA`d9<13XvW0Z zF2DBa_6)JA(6PwP!tVESI}XPHL zwY8(8mKGbMkQ_1-lTK#$#cGQ|r7ZTU=b}ePtBjd0u860`C4HN;???zDzw#PEi?2_Y z&cB`4l~QEm(+C&A{_}=RIc&c?4!e5tbMv_T?A$6e!%**1cY9q*m*tMf2BPHrWXI_f{BcMfcQx$#Uci{xpi+F$&zxR@0WvH9rb_yNDIx&GplTB zqLrLfp&0e(8xR!JKi$D7X0np^El6OI&m%%Q)rq6uf5QPf^)p2{9Tzw+jJ$Jn`C{IZ zX%Q|dwo)*?QuWq)WooJPZXxyT{bf@W>5LU)j|ypy}tNZtoU+Yr-EPr@vl#Gp^&YR~Jw2zSKYg!xJ?a*8(rdgVr zb4U;%otNbg3nRAn_7x>1P@Jno!U=(F}s?naZbgq?cjM(J-D)OAP;p3&#eNqD3#=g?O z8On&u;y=%PGoo9u0VusdOt9#IPV$vC2$m(1xA3F{Q`3H)K?XK@1X!Q&y0wpbX zmQzxyZdY&KR(a99?Nf@Zi?z#eLF}`)HhBX^qbDj)l-^3OYei&j1xPOS`zqU(klxtj z+U%^^98{V2>(8$s3L+(Qm~a>0)|YGfTYaPgS@I-?`bU4H?GqPYfF_p5P;wSVv}tK* z*x9YEt)Usf%aeO#=RN@e8)#%0CDQwPYx1Dt0QZcx=W6FcqkbOs%he`x=0v z;h*mEk+jZYDXAIjvkqrZx_?4AJfSRnH6i`hGq3V&bGc-G4#(+P1MU>xhF;*n=ev)M zt2Dn1e<;2qnnmon{kw923e7j7S0?y37o6Q5DG^)7Y_Q^2<+fO9HF~hv@|dNgb8G#_ zLkT1-_s>;qu6&$pc4whFIo(;P0;8C0LTwWIC&_POwvzZJE34Y!>1yzbHPy1Tu{q`M)cgm!t+#hL4~}H7{CmTRnig$kn}ne`NVDG{If_IyJ%g0QNOst{wvon%(^c3J6g z=l4mk$`uR@^DfE!wx3>7pptR_+Y4}WerjUCF{`hQpUvfh%rwX7`E`p`T9(&K-o043 zZKl~=lrIA*))BzH7cyszT@Au5tAGOVy1e0OoZ;gM+tr1upn zSF)o!qDK_AMSB#CIrcc&eKRr0buI|MKdme55bUUst${HXDBDRxU>KD{7vk|=FI5`4 zXx_9uXGZ^7NANKz4aFT#rB}3v2I+TeNz@H~_3Q?TZk%4(`ml64@ztx7qAE{R)nw*R zb7y>%mYj*VcC%3fy<*_#7yj0xhMxYz;S>(xg>Szfc(^(JXt*L8EvmtPL{8s8U6Dsw zZ9BU}LQ`#Nw!{L9E$2$9-?+4l=t0_za%8Kgfg{S*jawIit(FH|Eu`hHO2v53<=tWUMuwejpz)7@au#<0v~#;LeHe zkpn#W%zE0IB6Ee^uhQA)CcZi;$fW-3T~+C!IkjxgpQ5pSWc>YU(M|T-HMT$MYKbYz zV_y0*_j{t$FUx8%Cm>5X&t9S{+f-5EeCe0`ox###XHCREcBIApl(&y;sU42gY$(@f zD{wa~SEyB_qP;JBfS%=S{NmhGStNBn=lN5X-7>vQgqk@eLrjD`)_-0R|ZACcSHW$0mseJZ7UBPU94&mmE(Uv@2l@st>q3@*Wms_(!3Y2&i+Fad6 z-{?PBiXPVg)b^>2r?ko_Tz6?<(0oIqOjLoDM{RN$xWcLWqSR9EQYndrE9z!-1|R(p zI)?<`XXj;u=h+d>O5?vLc28I0`lXxmra~7_-MjM9YOBpu&Z%|xvzxMYOLA?*h9+bC zr15!~)y+YS3bNPQdh7!~?XSd#TMYV4}J@J43ZAO zn9bH^$-?Kyb3`wQD?NNFFgh!7>fX=fE~{bXX36hXmzhG2A#JnBe>bHvxYyUDIrp7k z$;0#GiU_u_+gRz%1KZ}o%mqF>;CNv8m~yVu(nuqZ#Xm1bT8@11d?W7c=+Uu^T(at8 z6@tz84?|y-l{~8|DYARp6gOFJ2H&xG*kaTw^m+CH1t3u>>Z%D?Pf9Gm!8bps+mObp-jnOgE}A34yO9hb0OK=pQ?-%o?uud-j)WYzGe){{?>jPUhaKUe(OL_#pdVdj3e}1t&^=kcz$YX zb{_W3{3`8Gcf6=?Kl`}{k}S`e7V4u@Wv+Ehxh<2F= z8};VRn1b)Gt}YK$L!h+ob;j9wZg@%Zy9$?D|7P&gFXlwTo#$u5I%K~zHH+3rr#CF~ zHSm?@ez!3;u4alIDZW-*E-qCyseae>OT>feMczfLnDeCy2DH6&)7ANG7R>L{29-^B zww!Afw5vT{a0|q%!D~6eTgta=b1Nl97B9ys#+j~!Al*yoNI1r4adxz$C3oiiH~Y~k zm(g3AJ7nhTm#d%o8JQFKoh#0>(qEsgWU7wmn95|JK32b^ z1&X7>t~)2gxG#0%O`)96-7bq4g=zbD&0c?CJR_BtW#*z4H|v1g-I2ZEYbKm;z8YhX zr0AZ%15G8iNS7tr?hXj1LS0NRQvAgTiYE?G2|kUoPcd8K1@Eryit^NKY~zzblXi+B z>&TAV21i7UfW-Xy6Ea<@(C-;fUZWg9kN;Z za$Clb)m3*y|IU_pmG=BbQ+xY|;iqdJAGA7@+mh39eRvXvT;o#|g>hM*UL7Vm-R638x9=FP6+S(ggA@*D+ z)ZD_bEnjvkynTvNLJNvPQ46j+cJVpV z!^X4>zx=A?O7dyP$y8_Gw_-J3F(dZgB?&e{n{H6yzwBs6R5ONHKZhlSi#gjROhfGQ zjF+vKsp4H%A)aSEZy7DN73Ah7yXs0YAb-F!b?Ig&cy&s-ET?RDQakF*B-l<-gs;PY zL+(4pd2)5;g;g?i<{e~U26IMB8m-$j>bVNV#l$K@h012Tma*Jqw4(#b+{#?!O8ita zqzipn#2XqKG~Ikn%HLibANRkzKF3}6_8nf`ZA4*6@S^>eG4xPC&o5=XSh(Y+Ep!mE z@FK>yuwr&j{^V^APU_Wq-f>AGS9SO@%;`mAUe73XqY};5YNBj*d@0_Ua`~BYgfp}c z%k{Q|a?`znrz7*6Pi$m&faES3b+FlBen@{s3~NOU7Vd7|33d=`9u7l`+_( z*!8pB*AyFxmhq=1!>#i@%KSdxbXgrQyr+AiTt_*J2_z|ql|uiJUDeco4#Yil$Dv9( zfB~u{XvRS*{CtN%_0u(fL1vn z%C36;ynjT5o^2I0;Qsuvp{_O&lEiAgax(u#f<-3&qVv-Q5#Za;7huhT$kB}BnO+_I^X=z!( zfG+Ozg&dn`w69=%^30jIW(&;u0n0kt+b8S$n@k2#I}zv=1=|>-E%~2{iVOjXP!2_( zL=xJ~2Do%|T29>Ah{4t|s4GsLKlab5iLXTck&LV^u$0>AH4MysqC$A%oxOZ~e7w9$ zsg15qPIvF$=hmqp-@;iqu{sDq7N^-kmipZ+^X1EzUx#}<6(l6SwCmZ}a7gRp2CkXk z$-&s~jnz=6Lrp9fVumeCdjxMBk|AAoYgdY!Mn=6T6G4{K4*tJoevjbI)pP?l<)!-; zHylNx!<2<3JRra(Lkc|uT-@9=)KK}W$6+WuNxH#o+!TutC;KYkjvyaQP^38)=`-#RQw?D-+HogVwcC#Ou>r90QZPs@V>Fq6X+@Q^>u8s5Ah-np`Xr> z9ps)FF*=NIjU_dy_3hKVL*M@;{NbBD$RZpa9Tg#9Y=q%lg8`aD$Ccug6%}_4SU+tr%S5nV@3&!_S}LCi(+E%9TYWpGc7GGp-BL~6v#1}}ZOKYSDc>s(5~XMe61I%0$H4YOLMPb#Hv<#X^UzRk zb_mKE9i;PR&VVo?oQ4p1)cSgRh=#J$(9+UUQ;%a|1wrg&F7tpymNVDk3j;se;%@$f*zn~aR3$FQ8M7fWx>Kh8)`hd#a6dP+(tt>-P@ zd$qZf1>^#Jmr{k)*cK0u6@(<9$w!Z<47wn1{OQvtDHL@!>M9SwD%yb zZv;&z1JOrGMn+pl>2yysaq7<>0SgA)W#}FcIqL|@pM{_#9`)+VFDcbk)=8+5uFJh< zibd(=@g2T$wcz;;%HV$Dt4-t{GdqiKJ+-rYl%GA|smXvC@zNL636h1_x8U;)cBG)F zXyqYkQW_TarRjpVH-~sjO*HXG#|_jL#K&&ky2Y-lt}ckJQ&>iEK}SbF*Fi}J1q95k zs}~nT?sJysbE-)@s*s;H$ScqY3;zuS3d7{?aM*C9xU<;NfaC0DbiE_j(FzilgR@Ve z3!IOgN8IKzJAVzi?kglLl+=2nZWej61B@NUh+C_x$G5BB+3#n+D(rVWT+Cvhuu=Kb z1&C8))ED#d?CH3)*w^JSSU0e;Ji3B+e(zWR+SK5HDDf*)VpA3h7v|=C8HE%?@m<=X zQj4lTMun|(u9$@0bd%k^dbK@uVvX2lL7hOa5UQ-Tzw2L=tR|musqsUb^5Uz}2*WDb z{5$%E=pIrxj+hzv`g&4ieE|!03_4P%rLO(=M{F9)e$CAMrCvipfrF=P?V7$m=~-Zi zbyskN@`{Q_cT+t{i9Uaoju2!ci#zU_`V}6euPQPUO*ZjvP=<${+u7KV#FQNz9nq>k z3DIave|9GRwFq(Cb0rY4M-NLt25Ndh{_*qj3cZVsj`oFwXkFZS$iW`lp4ZpgyS&h` z0)1gZZx%j^SKX%T6p`=|=j{5vMs*fR7rU{}k4;TY8s{XZ>eSGjcm8pBSQr>)eLX#Q zcrxtFE0oiz|%)M{o35ciPb$=aRQ0mIuk+E(7|)k z#phLD+4A?&7NA;Suv%8f?0NWzpYT>_+`=)+b^{r1-nuo> z_(Da>FxF^6Fk#9C!U#8k5h^rl5Ld8U!GM+?8_B%R)N4#KZ6ZoOur>iNBl zzR`)RK3MLyM!%IwV@kMJn6!r!7GT-^BX#)j_pe`L-oBlLhBI`5Dx@WOd0{b-+#>X7fjt2oE<)?t&(F_h zu+6apsklD>rR=d9MSeHz(2G%Q?dj`VnM(D@zd4ugg~;j=FGj&u|7RjF!fjlLU6V74&%{qSfZ{(c^Pj0A{AKeFXa#$Y^lGC*=17hS}z zU&Y#b9H}z=l~#P!!<`49n%<<~i}p%f|G?L;yTTLRR_|N&E_l2AePh^p1v|_^PJ_c? zHFT%EsJNJrI3QHBv9U*Y9(u+3kQN|huc@f=+A@VWORsvY(uHJX_Ob5kr8qfzpeo__ z__4CQe9nbPa2KeNJUp|{#YZo(UJ7#+K7hp;uSQPL?Lvk|O?_p6 zJ9IzEO|^0dtMvoHCPz5}yF&3fCpVX87!6`0HPL0dj%|iF_ud3%Mnh#>*CRW@ou#{6 z)=)?|@`76zxjLLk#0cS{187sy+2w#aqS*@VJ{4{9b~S-Ko!^kIH0xxtmrY*=7{7- zNdKeBmh0EQ)T))1a`0BH`S5)&ZPx3F_P6C`rH!ZLj?nm{Q;|`}POJ@LOETBsVmHcN z95-G=!C5j&p6C7tbp>%vLU-xdu;}GNGGDpCfB*f5mR2KGpt_sXS=#voUlEX<9V*r~ zHa3rZ6+A&`hnb!`aNvN4hX=K&{h+#Af=umaQrF}S#k%#gI7~d0SOIe67_dZ3Q)+aXX-eo7*VVNf_ArzTnVi14xjD5n%m*IJvh_^jwyj%t`a_eu;m1+j zU`XZ2KpXXNC*BJ~VX!`(OHqF1^vux_=*kIYj{b!Vq+H>`HsrcM9&-@SVmc_EiC^y9;n zW6MfQ2eE1jD!$m=ipedRYgng-omjmIyOk(N=mr;tY7Bxnuq0xO1ytJ8(vV<7SL3lw zSa5LkbryPN<~%I5n5XyNf1i}-Wy!?OEBs$TFC}k3KaUHX2h?4@llPf zllEeA148CoLjr<5*pb?2QQdt(N708aAe5VnOIWUpaJyteDx&vT&3IUdFhjdrvWdxn3wv6Ef z<5;_+gojI^IS>R9Z{Yh#*AY=1&v}V;O1syx=eGFBlO9ki-_9a-HdHX)%!S?`NiR03 zWLkABBj=quxI>A7bopM4s6Q@5fryGlOz^;g({o1z1kzn+Mv(a7hNE640cwdjN8H>% z>JNZ?SyhzJYM z;CXD6F?QoY8yC@+i#r7;L5sVj}L+rYrEHBR*KVHFwVLH5|s>O$uo z^I+|h#D&P+;onM$T|asG`Bn|d@S-@O4$+Ea`$!@4#R$%R)QVgRs-JbOwLkVg3occ; z=jW$=LH=TB|I`ycv?ZduMFKrA;^I1@Hyj;Pb7Ir(U~-G#Ei%$+An&uAgR^MOsPDw? zFJ#JckcN-0ki^2%;5(05#1owU36(G+kw`4)%|nNi+ID2Snwpvp8J1u(NAqVQyw-2n zKzJoxH$RtBWtY29RJE(olV7+r1h0qGD_dWf)x)d0V-P36oPFRw2}MN88yR>&Zcjnep|B>>Y` zm$GlXtikSBX4HE%ss^inaP1pAWW6del?juhhtxk>2fe+w^b|vp2x3Kb-<_*p$!fG9 z>wDumV@Z4NwA!mjg*!>Ck|xqOYwI-Yy)G}Lo}OGy^6kN`%nY|UqI5cS^Nm0UYX$RK zq{ff*!@AW+j|5%0Ozy8SOs5ZR-=X+tnl4?KxkWVu8jN8CXc+@wKdCoAxmIG6Ep- zZd(7&&dZ#IOge+uGV+kj&upBL#DHv2>%Ve^1WMz4nooDeq9Yz^U)0qDmUdXgLM+5| zQ}jb!{QP<5Pzrs0YX85?9@#dXVe)qDro!WG5uk8W)7-n@yS?|;j^);v@1H;YN6yqG z>5+fH9%<|{#g2(qoR_mLh(U*usbRSZPDg#JgrFO#L;tvL+>S9=+&1jJMCT^4krlaM ztr})nK@)~bxw!y&?7(8PQTFN}F7k*qmd^;p2<|g0>SUnZdrmVkG1X>(7d)?-=i}=O zgz16uwq&Br&fXr*W5)uXaTOK)J@|)VtoP{ku)$`RKP1)#jvaHrmXsLdkV6*$()Doi;Gjo$HkpFcP=%@*v!n#**OOJ|40oQ-Cnzd_{2pxLzsxnH6#UJ!@Cno%ge)vMn!O@RLmkO_76?|tesx`f_9LI zO$>Z{_x=Q}fX&F&pkazfvPr_<*Mz?>BQ|}&&to72klP(u z7nTJNBHvI^`4{*X_#Zo3nPPr7Yo>{zc@P~ZeOW%;7}v)TC`E;KHcft{LW&bLbwBw3h#oQ zfBAZ&KYvm)GA3LXqO3!b_N7=+HDJ-vXEg^XjI0wb>=)-a>3|Mug}`CC!PW8A%hjE-T3{zDU=oTYe#(>zWixcXAXAulQKkwf*AS< zxohOjMWRM2a3b5=-4S^Ztg-xLs5Z6=o~B%nE_AQq7D-Zas=_PO@(rv42;mz*(TGaz zh{&K*f5Qo3?GHj$`}Onl2O^w)5$o^t<4u08u^x=Tn_zaKU-x%J=i;^yJM0uKy7N7W zgh}zUaGkR#NM7IUh0*$DBwio`M|gE#)&B`uS9j8xQtC+c8ce){cwsKc2>LN2d{|qUlp-U@w#GI*L&A)_17jN8>?j5(cnZI(3l9w3NzR&im6Y?hQmOS< zdmhBi`io#ct1WC=#5UQler){gm<+3|jZt!)ABeRYIKr32{aEbs+6&MAAA^A6-(z@O z!)*H2>TuVOZLGaTr0CuHDAci&*dSUv0CO+b+ZHOimiasa=EjO1YV3LBN9yh*5?0?;{SVe>n}Lq#bIt0huOr{&4d!=r=IbRDaFVyTcqETklR zE5jGS@{Ui7TSDO&4)ibCu>8V(%M$1cF}r@0`l1pBbO~lNR_S>Ij0VoU?zM#A=_G?y zmP{j_`Ig_qhYVtNCQuvwV*n$^n9iwFr!1~tCuWI&g26&+s;eXU!h-HYVq)*HCayP; zL5X?oW@#XpPgF^#M$KTwq1)nkG0457x}Ts65LCeL-mtVB!Se-R8s;^e|G>ci@~UO5c4QH@4D%vZ#{By7@M~MA zQcrKMa;TswoCIb!f#vGW>VX{S)-Ez9_slK*OtWABY-~LP=`?{B1#FH_17buWft`k) zeg-u&#Aj}zH6H6t8E9y1Fma>ch(&QGDCbEgo>uGWJ78RSMm4~0Vhl*DraoX0xmXLsItLfzG`8iI1f>U0+-TdrFk3@Ys(qfUdo7-_B%& ztlPa-G7NqrBEiURxITiE(8{fM5|mjBRGh^BfB1k4t>7IL@|QW^y=lzE>-zO5*{xhW zfo}!>f@b*cZM#V2lfqrZ|C273Cl|uTPqd5XWAkT7NF&C259ulfu6eYl?1hQMraGL` z$-#ObGLMD-&6gJLauHPdkES`e`&Pln%ZJs~)$7tSYBInNELP=)p#k$*L3;06EW4aA zO+D+e*u}=^8-qjp^$Du^YOkvwF^m3jpmT_om6IDB{4KpQ=jX*NW*67&@N@$ymF>nEfd9)XVC~#U@~ux3+_tz~zWl-qoR$(;G}MI9N&pwQlgNZ&e)Z!7mx|5G zQ(Th{t^i&rlL1`4SlrFZI?~yBBcqa@@6xxjGFv;l-``)}G0nwVq2c6o(eZIq$y8ON zlFr)Y4X(!arbNlz+fZU|Y7*t=AI*RaE2e}W4Fl2>v(GGSN0H46w>zCO^mY6K<8(yZ z212k0z!$r!p5AiFfk=2D$$Ezm068GXYccnXt)wklHCVJ;u=qz2oN!Um!!sy^=kUCL zZ6`MVCm?8-yi)!V#?|9dabcMJ1G0)pkL5X>d0iEi8BqzV)-5qg^ zT@n1j%|DlPeOGz1t6*ieSQ-K{FXsOs-8f2TK#kVd0Csu#H4|>(o;3LSCmBm6tb8OK>EoC|Nv0waxc7BmBoAd{(jt|%@*^Me0vG1^m9Y z59&IEdN^{Hay_L}xQsK{Bz_gnh8^DO39wcQ1ea@Mh4#T%*T${A;*~@Nsr0`)zV#<0 zdi_u%0vr96m^%gkgGVvS1Tit*XRw$HbxHuS8Ens*?e0Z;b3vV9qanCeQxil12PY;>YZ8JLyuC}k}J2<@aD#vC(l<|b2=j}R! zi|0iUmbWM<)kC~`ACa*7@t6Ax^|NoB5Oo^+NNGMIdi}@Yo;~~b%N5-rBT;)~_rOR8 z2Xmssg>YlodVx}lW+E+I&RN;9t$aI}aX0Q9z4gCh@h)q8JOa`~{A7xwshQuSQ5ASIf1R3q8Sb_jPwab@*0M;RZK^#w&eo?St&FjJIiNM^F{FB!#6H-{W(y zBQwSTL3^6PX$OaCRBo0VbgYOSZyX#lGBD^Ay?nFQS#-y&%@0&O3JKBHQIxyCi{2+Z zeC-yMojy<_i3{=t@U!Uw?a0LST!%B^$)5dmjEkvP zD0_nH71v5+zPMjvLUV`6au|iTr>9q(2>D|#iYU`igC|CJ?q}I#8-}*ME!U3^77cp! zM5ISWvpp%#$hr2n(S>KBN|ik_y5x4UOm53XBZ zpK)p;dCuB}{&najrNZgso*YF*qN_8JWH%13ld!aFAgiKUrLGQ~_AK^&%(WPVI6M^_=+>NCOIB^65Y@}=Kykcu0fBI})6`_JyzNajxSCi*KpZTA&+ z?JtyDeY4}9#Y&!Tt0nZp`Vp=2Z{L34hob;xS+F#XB1B-xwVA!0*ME@cOK;jvgrqX` z2E0QN1oqPK1P6Voi*4mc0facqaD9mGB0SG$`Rs;e1LtJv$K4<_94{9f`VGuO#x^F zkOttz7Fhv5Z#wy|W@5WHkDNq zY%ou3(5`UtzW!C{Sa^bp?`6ofW6yciJk`w}GV!V8Mf zaEOv|1Md(AlBUYgcF>D_Lz{OOuATPEZ&Z04fhCh)nXHIZhiHJs;Q>yF1Ve(jx9GNx zf`ZMd^A=zf=02KVtBVMkdm3XKHV%HR!0yguKnwJE#PYb?A(CStgs$(`ard=jYO^uq zJH7Yh_J#C~vpWwklO$P0&ZHT2+Fm&H1VkRh^|EtvHfTFCQT$Fgxxv|mTT<856rG87 zLihkOl|gTDiULGnRHwJV_>;7ch=@3$c_IKShl!Fdz){QVlZ7#68V@yA?Jq2NoTy^{ zKm{2qA3Sggu{wHAy0o$qJFG3i=)ZRyC0Z=5(r-TVdo|xI%weD!1LDC)Xy?0RPL<(a zH9*3C^p#OY$+dvDD0QLk^Y70q^lN|r41a(A{CPAphU{Wx{eU7f_&joojot2pklVvf zcEoqZWVK>9gppW$&!b0aR^>X}W^{>>DI84HmYX^j?%unXtd}n)A#te&j0x_o zR{2Ttp=mF;FkFGoFdQRn7vbRGI1?)2oE{t7SO4DG#f4bDfoo8Mzuy+JN&CoZZd3S- z0k*T9C7l)PKzo}d!T4KA5}xGNEzd=gYYcnlPh6jaCD#Nr;#GT^CG+$1bzvLf9QH4N zBq{dKonN>1sZJ<^%G3F9?%$7hLRAU%98PTPW!Syxn;*lfF;teH@wBRbb#Ll2I)NKU zaxQ;dH*`W<`zi5hdHEV2@+u>>6ue&(QXXWUs>U3J!WPYfh;!(2sw2=|!!Ytg7HRjq zDD?gH8o9p^d3tyCH@~I&K} z+(N>ABD`7#vA7!+NKU$n{i@2#nPGXLD511@z_cP%2q~wJpI?^tp@s;w3fsX%erxiF zF|8Th#f`?DoJfsQE@mbGJvZ>9(9H>?2s-UTt?T8>ohUjJb=cfpTf9iWU%9-KDo+Cu z7ga3!L)v2?cY)3KVW;}lG^`I1LT?7_73{`ara>JDgp4gfvP@611zU zNgN^9)yI<8Ku6ttlh7-mV zqp+c5^)1)@lrrb;P5|D+*S`HTbcCFBI$(j-$)~6~8rPESIG4a5#vFuTV!3Q#LDysPN9vS|zPY(K-GalCjvA}S2fQYA2i08VC|p5d5#P#6ExIp(v>4PNI!+0bAjq6E?Wq%P6G;bG%On`@00L@$liTN$B4Q@xBuxa!ML$7hc#c$y=BK3 z_$@r5Rkog(Ffar-d9zN_{0&YiV@QHaEJrc_#ufSJ3QHf6K_7vLriXWw;4$9Ti!2)dA?zli3Yir|og3^|hluYKuNbrRMcQ>eo{Z{3>Ttteim0!4N)nz{r@Ym~Jxv{a`m7nxDcHN z_QYXvoj!zSR7(S}J}6&YlSHH8yANV~2M?Mc#VO+7l}$x*|7zAkH;vHZhi;Luz(9D$ zXRTjxjzl>)y3qoSM%;L`paRK!eZ)Y_LHMcOTDNAr#5JX-*UGUmL?uw(C*Z`q=|FPy~p9wivuU@bk=f&ALGuJis@ud%#b zT2dXa_L#&iD@1%W?u7bw3>Xte#%8=JV5~GJXVC{ka@mjEBuxDon0nKu8%yRTe{kG3 zZe!eqP89gS)C(D3ucrOTl<2&;_FW(NlSyOD1kecvi2GT6{g`6uw{Np()*x6K(AC(n z#wup_t=)c|7O7IwDFJ++svcj1hCj6Z#iFVVG6%_XX|p>5FW>-zRThxKR=PvP*vyi` z20r@#&Ez9Kl0_yWCF0K;`}Eb932gupI8QB>xk>>=SHS@3BcTGPT^}3aVus5$eULUg zOsMcE30yJk5-#l9vSkY+PzAW3>5(>ZH1i-uQ&2c}j)-J1E_zy8!_-@zm)2sp#BaJz zTOpJ0m+|$`NBR^14pSwWM|gP3MkvVt%Q-6x>pz5)m6wCu>%oHeah%BsOVJ=fQqm@} zV`2R)N)-gz2|TfchU}W!THBTiBU><5;6;hd>WAT;Jq6qv-Gd2t*BctfU%zl#Yt_qf z#%q(KHPH}lKg75OZdoK+9amms9Bc~SA7+O>JYKzfg&NN)5>#amC?Ay4%QO=u@k1J$MFL(^Qc||WpX|K2Zk^q)yMH5Eb`d}*(Lsc|&GHi* z6I`fV1aeeWr{lLCWSm}Y+v?=-&b<2p+ogzoP^h#*Xv;G7w&&2NHGpu}pKP?RZ>Icc z6>4p<$jbHymW&9I%i?~nVcpH}kEiZarMa?r9$5?HG!E`N7I_P=rriXgNeEpcJ1q<5 zX~cN}x`%es7-Q@qy}VhF%7K%J=E4{mdCProbsU$hQ*hJR228h14>gXW<)YLA4XbUS z&3=GjA-Z`|bQiphhY>KbvMS*Ic9x3@_kQt{B9HQ?-~H<>_nHKu;fo+qF2MOfOZcIC0ffjX8l68CK0Kft z_OBfT)m^ph?y!^M_eOkwU~@&R-)gTH!<-aB|G@pB{y9$w-+{3V;4}SncXJ2!*8Qx7 zyZNnM?)zvh1W>HYgOp|DmaX}~R7BJ8$&;U;lUIkJ#?PNh!)pqc2=||%eW=m(7jg?X zWTfZ>50TLj-ed0R7`i}9c)YM(*`E^<=Hj!rUHsbmOy|!_@a}p_wV0F3E+*0LGarLtSU8u-@>Om-5tkK29bLmk*#|WN1hm zZF}unH?kQjKaDLHp+12=MZB05!)Zt}(IP(tmxIKnqIVtB)bq13s2Xsf3M(o)2godq z?x=0cZlV-xUWDRKs{eB&>*%cXdD?D8H_=xG1Rk;rWja_HTI+&&v@CIMF;t5L7>+_e zUX|0QNC0IEgI~k!f#aQF8YHG)TCk*BOiHr#UaR-TzmhO3wAD-z@rOR zfDo`X@RNM9@FW|(B%r5O@{K_qH74sc;#Xj9i6GYj*P?Z}wv0`v<=7}FYs%ekZDS!# z#061-Kf=+(q-OvGXa|TTagC2hghM(zJCPAzhpC}!5fBN^h)*r1oYwUiJS|4I6rUU- zRD9rrG)wK-r32+P68ihkhfBnZ*)#tyZw>QkxP#EH!lo;b`uX4U9X8>B(f+= zV&C&0hDJsTVz{Fim>SmIae4h}j4$M^I^&te$6tU-49Zf(6Eep=za*|&T5Anbe)vpy zr!z#TLyB)a8sItfbrn*0>O{abj*KiLT4C{mF-qg`G;?5|e(TZQ&?QAzOJ})bX#@iX z8%Dx>6!);dzaMY3u&^MEZYA7Mt(Kq)Efa5$k5ZRGSaebSmu+L`|nnk=(nh%g=uHO@gNVT52qS>)N@oES;=cmk6=0Nqbac?XAvhOWFuuO)Sux^)B4|+0TCNx$l{|=DKFwRUs^Z6MI3k-;^yz zG89I^P6BV2`7fG`XwiGpASx^j14Fgs937DaUpToy*8KT28|p$(+W{1V!3b<0453IO zUm%(9T%p*IyXr4{lp4mYA;q15DALhDGzftQ?CDdNMBwc~iV0+GiCAX+cDPG_Rq9t+ zn$`XF^&f8z_~ZW4~g}2@1YXgGwk+pSAMi04DV9 zn2I@HIO*&E=ia~40Zah)J@^5*ni|ze>RmntR4>?{;JmWcc1u^xRVYeX9q4<9=Db8A zO8ECF+UrAFKt(0a93!ciqflut0}2vlWo0pNVb9=WLaM6H$NFN#{8}uA;&OUv;B3_K zkk4~C46Jr@lc`Oo2)Y)+mqASq>O>UCloky|_?hlU9UVO4oaZ0*8qlgX`O-6JnR|Qm z{~|}{;OK!|`<0YlvubaY7m$|Myw|X>Sm5jMGNox~#X}7`U&Syxfg3YgQ~D8k75RPk zy8l(pt+g89(rUXkHuG@~F$luIv0TH(hH}%lH1H4#xk_D-r@*c*t2;ek$xPfBDoiAy zyK1dUPt2vj$w|y6xxb2k^mAlCMRe%GL}_h(c}WJJNn_RTW%-WgZXs%v&icvUBQ{Ql6PLAjOQ=p zbA!n2AuqVtuWy{jCXhd)L8|@dofDW73?#runiB&9a4Z9@0sh+{(14$>UD(YZRb5Rb zMJspNlUiz-TxLGd=sl!n@weo`&J3F(zkJfaq%Jx4C$Jk~Um;n+Kgc|4VEGSOV%3u0e*4T4S-a(FNx z-rY`?bQVT)UNVn726YOfwmWX;cCy{yR47*ZX`~VhuH%a5u>Ji!C!7kv(}tkx42mHP226xil2&!BQYq`V1l@HpswWY`0Baeghn|Ko`Nid$-bTjLk%b zkT42}=it2{Ii}^~L%JP^4ShgEZ&AL9CapWp0jtQU^~}0pBoC?>nmEXt56Ao@$dM0_V?Yt+Fl}cM8%{k zdFn2N-y!?32?k*yh32yZ9G1WDkcM{@D&|J+zlKmAaBnCynf&a&+&BcuibRfjdBMA6 zq!}VCrC|vHzX3KOxe+K@02NhU1G(2TsC0{g2pO`PQnmyRD`cn|(Vz4jjeH;V!1API zhXM2NZjAH%n*^}~Dk%m&4XWQ0;EsvWrhRYIwmArN$dhGu0TUI8)%D8n~FW=+Jb_Y#TT zg5Hfe1{%SG>Nq?!RA=1@Xf8;VM5sIDDzI>B_q1KsS=Yk0JYSTUWVQs@ZPt!{ea>pa z^IJQ%XKowo$p{ug?04N58{(MQ*}APDf{N`h&QPgY9$V!n`5^^R$6thN|*U z2}s^$U~M&JZ`Xw%ot@mC(W>Un9HBZ2xrA-}iU<6z!Dw-*}{AhA8pf zWi|XBBxd5jLQ;Wpe7aV>sZg3TnAZ~~L;Ltg5BlGOXkdh;0gWBlHzn5Tv z9sa?gmUiAu$=Zi zp9@x&b42?4CP<7af!(2s=4vj{6TzgN0EKI%A3-O7SYStr1V`4_CpfqNcOz3@y+}W& z`H=Q5{!ifMIB0%o49wgiw#*_%Bt-8y7}pN#{c_% zZ56>=cu#acIv*&A`;$5fhz4=|vzMa)H`JOI9WiNXKI{3H0L^PMUgPi|$@9hs?$( z$EWL0dI!R6AumP^%YRn$1<)f8Ab!C~3t^dheIB?2U^nhC8neYoQi;Kd2#kBQyQJ*o z9v=BB$)1Swqg3a^!E}j#H{fZ*LNuM4k}_DR%|rt;HXzjkWehY^_<7zxG~|E`zSPSj zk+dBkp$w-hC55A7m@J4NpBRrYpHE%>7qk6>U=J!=u7Cw07au-S02r9ZCN2R%86YW8 za2A6+mJ<}SPqEPNr- z2E9tC{l>Uag-}Bgm+kKqVQZ_NZhr26;Da=j`#`nzc|Qp_14A1lb#5+e-+`od1N+MO z1LwV0lAj0|tp5e^U!wmHRYo4CUhu(z&$Hd|M4@P`san4iiv` z{G&*1smo<2E-M@Q%qF8fyomK+voMmuZq_S9+U~Uxa%IhF3R-xkpt}y#I>F+ zx|(8kifn4Zc(TZi+OuyIH%M|MqQr~;y8UwC`n(nr`3UD`S{C*3kw)p5?`=%X-XM}# z{`!7#Wc+Sv0w9)%WI|r5sNI;^>d>az#P|fO(>DrOi}3I}g~pp&fkeq^X?Mtw=Z(M$ z$jgTq*p&J*JM?=0j4%@(7-lImm0A?`WtuVhFhPjs{O74uT!yk1&B#0W0D=RsfeLe` z+O*xJflL6C>Etd|i|1!k$|rm4+$dEBM9GZC*v7tZ)<;bXAm4xgJ>r%|;*EQI$4k}@ zLuq2dybEPxd-GxK?LhR4P*BXycHUPMt2(z|m|t!o)GrF}SNUHbBQs(9^%FExk{Kx% zhu<+m_qy0mpx>UxZj51nO{JEbqIww-_9X7()qa`@r_8}T*?}CjY9)zCytOf88fcjc z^|+x-Q`Nbps!C1H#LbaZvoYY1821RWnyii(mRhGXYEZB0-y-cFv*|WaJ=mG zN^m*+a?D2Fhxs$e^H0up0_X`oTW~JvbOeD zfFP&J@;TboEMY7$=T({s_v#?Bp}m!ZvsR>|2+WjPZ z3y~O~pZtPyrE~}n=kt*JQvB+O{N206tT|V5(cvSnVy|sANd1X-*jq}>k%z7AaV!e! z-1H*REu++gMD{N7y3lC_jXgIF@42lnE0nHvGFfqxJR8Slk|3+1ZqQNPYD57v%4wqLUyO74L2K+u5b7xil-!*5=pO zH){3`W%$#G4w)mDo$ZNj+fAfFa6E>e5$3cj)zVD;49;kR{cz+iTFxyjvDUv%z-AJ8 zu%W1vqUDl*MHIPUUm|l;u@IIdckbZfQ;HdO!yghk4EJ$(0wcuktMhGZeCNYY^$oO0`=c1-gGp(% zwI-@_4Q1@soxRK3360^bNzWF&1s=5(8}j^8lS0c(xDA9S$XBNg?vNHl6UFnIZ0#8j z6g=r7B`x6I>~9TJD|s0}(Dp+VuB6Olx01rF?{7O|HdbPp+#Wd)P6OqV00N_PX9zY# z$yik$BQN*&zGArS{>RRQ@;*}yF!G7Mbqie2%0j9F4{l}%xYh7r?U550 zib301dA)=|=Vu=uvvf;gb!yXTOOE>6d z3@k#vwO7;=~q#_!J77SufFB{elArZLdu}*k|*=!IvK!2y`=up~~x#0aViySRY7DMB` zw|}8D!>E4Cp{(b|UTS;fvlxEr4@)q2QJn~<0f)6UlcOvHWQ?gxHfvVh-d33`S+YSndvNmqM?bCrzF9BcO@L%0ehE{JKB`n|3b^QbiW ztks=jdNjk=AgH;JEC0K^P&%FLME=N2jkxu zHNk0xJ~Q}~qgXBIFBmkRZIn zP$(~;RUX_~5SJN>WH{;eusvN5?XRniWgBPm%|nZ-j)1 zOZICzWQzvVIS{dm59AoF4c-}NZJ=f6>Hd<S|hbef>l-$eut)FD&LInjWjCS_*I0j+!eUtBUOQOW&D}Dk zEfu@`4&#}tZp_Sni=RWsT&GI`xSvqhssZgILg#w7c=}z$c&sm0Ik2y?BH| zz57Gfn!^MQjc3RxAe^Z%i+X>}u$p{&UDL6&^%H4+8U<|mje2c4n)S)3sRGTDX|&2J zz(4};Wir+M&XPSjnv)~u9a`yAD-x`A>FCNY@d3P?n&iq+?r6P-2gLgTOF<^$1l@Ij zVpBW(U1ddygOOGHL;&oY_71q~p>tiWA?-4C^DEwBe&jx;ygd4OP z;h&SpJo;t$jRt7#mXkurceS=ia8vsgSWSPPmd1~KeN*Kq+s&JqAr;W?u_y_hIF5H& z-+oprsjyBYU?@20!G~f+03oZR-1Adtho@hVls zxK}7S8S^?#uSAp8?&*mM@7`zdKxT34ODWVM!Gq(KV7m1eu@2l>(X2-&#+U9;Sb4dW zuMQ4WLla8dlif%9p<$&4&;nKN!_}Sb$~v8mDhd9}yf;nHbVao0)Gb!L2^BJfUN%0l z`k!0?=&8BAvYallznq$*B71r`<}M0By+uv(Vi&kD=BPm%wZSPov4wV{A;y6MP4T%! zh_1^gGMa_fM=3ondr5&x__!Ymp-@uR1Iz=d3rjVj;k|N%ie%nE)=vg}ma1RQ017U3 z#VKzPs~bfW#L+is%Ss19B?p(>!lKk7aY(CjCxR^0!K&wP^a!T8{#1i*&GkzfF{?tI z>pOYD7YiGV+Eb{3etzOo48P^d@_2alnX?7GXWb;e22)8veBCN@d7wVVWz+QSTYsA0 z!6ux+&KJjz;&+;HA*Q;;pO-FCIR0k&ca7NrFe_jB!VXFmC3IgH-@1>R>W{xWT3kGk z7h}KC-=FPN-52H6De3)jb?XiElSy!D9dx{?A!h3@(uG5%zfk*nPy%RG1Bq7}PzLg{ zhgCp{aWWgwOT=EEEx+GR7y9h&=fz(P>phe$KXPTrcu%sf-m-64hM`6gI)!Nz$jY{X15G+s>1Z}p7a_w$V{=%LSI>y?W(@No&L^>9ER%7ZXuO_M%ylY?+nqRcC#yCbr$YJ3Ej*V_u_{i@#cPQKjV~6-aFKBM6nB2EuW_#|r8G z2b)ZRlne41fUp7}qL|yC6$0O3onB>2vu$MqEA8%vNh*_=hrvR`GRvrT0@d-4Yz&rF#*oU4>XH%Ew zQ&5SgtcA`rO%|V|hRhcb9<24}8QLDeSHi^Vh{Nb?ms5#musk`&9}YJiW7Dii`ldgS z{GslwcQ86OhnO`yk_xu&i2Z0sdvB$UxL6VT>DMRg!*ABe2s4RUF`a}}UVXkvRz?C{ zZjP!ZR|o9QD%LBg*&pm2Uhc}U8p*x9?AYtgKvExBVQGy*hx8m*%oLAi*?3EHBqzlD!_dG&$+SO)f4j zsjxl@))3E?tz1|WFz!;Qqr0$}As&J0H7cJ8KkxpZwOKXVu9b$!%cP1>}$S{q5mP}gtW zf5i6qo$4CBMY`N4R^rM(UPfu_xhZ5SF*rGaD|#%tz3m^^Mz7q*pI5oI!2U`){#$MH z_c8Uly1uDr&@EipiM&68F*Y!1`4K-6*j-vn_QAUkiq<~|UqyOU z6Z^kj7`e>fcDry|xnUL&nQ^j@PgZ!7WN)`Sowq9O%Y6t(jvUQKtdFhf5DoeMh|PYW zQtb`v28$6x_aBf0YVF}dsS+62_ee;fKC}+%foSpFL{sINPbiZ?%*^32G-QdWw>Ay* zN35*{JjW+?N4pau?C?J(YQLjdnEQcG#uO&Dze@AuNugWQRY1<>_0%;w@k#RcZ1wfB z++0e{(KfGuN9R(8koS?ZcV4fg@;@;%q%Ny1E7pG+host8&nev$wZ9_w4F6wHr z--;n9_!S?P=CaYVI#6~2q668cGQsa%QUkegFlkmipxONRvM<9wsBE99{YJ{Af`{MUx_AB3OQ{NgJBLsH|LBy+(HVyT)lyhOwEl z(tA+Wxk{4C$q`TGVQ6#~CjR_pVS#4E>5#6PnoeoPLNBgnd-MdIOSZA1dUbH{O_I*k5>o6m|5(GG@U$xpdvoa_oaEjj z-JDm?Bj|uvgv!3dcQI_w&qxoCcBwRxi_*fPvYL)Quc19^%Dwp9={k02b@lvel}jP# zY~qFi;qJ~l>sfA8zP9%D>nG4WwRa;NV3!vE9Hsn*smGk0WlT&iUO%Vae!7v9>!LR2&}_U9BNTolU6PEH^$828Md z*Bi*zJ>9o;kf~p`f7+qrV%y|=PgriK!08rIa|_LrS5fFk9VMnl=U26z#U!S9oK?P4 zY3^QLOHLjw?QpJeTX{D|n(vVROYXh$db<&ZnM~T|q~=lYDlx06gaO6*w&+a9Z1+Q z)ia@&w-az5njv|*#I`KyRk_*Ba4_!)QD?Qo}sA-B_SaMJmGi| zVtj5r=^EZ@lNla+zxP(kE{u#YZz#%zQXC%*E9E&l=a{OFnUgE#WJyHTFyKNCUY0*Q ztH+jWy@1Mo|8TQ>0>^3VIq$Yt+6N5$L24HqlUlw%gn&fO2y)-JhwS>6j4oF z`r()_X@s8}>r_qTewx4jqg0yY_-KA2QcAk)JP=Cg@7hktMQ!)8C?NP|*54`PYisXq z3E0`IIcqhyFYE2KuD)?U%V~FhCfV&&uioRDMxuNsCJ#9}qMQe{Mjq&tYakXwHi8{F z+1y{V0_jIm(gWla^(t});^JWt57zNGFt4AmT1>hI*fA3{lQ0YVOdV4{bXouLcJ zTYG~O&3Xh+WsY4>v~4ygJ$UZ`^F7kx(oR=@M^9L|Ej+eY!6Q1trd7eH@!26M??sg~ zoR;6C{?hSs$2-;zGJE;TnewkEaaSZF z-}X_70j6Wo%FU~ZaZ`m1^VRO6@gz^-#gJAp@AnXos~6*(>yr!?7K#O?MniQ?;ip3d z1HQLonO8yquv;Ie0R@vsdHKQiobkAd!z5&k3}j@-$Jx3+YF2tWB`4g<&EB=@2fx10 z@qmd5N+;6s&x{o&nFJjxX6MY#FLN$U8-;g^$VV2x;Tt=}k=1Bh)W?UVLgmNNq}S@u zd1J(=%2rIj9j`)GUlT(D6A>~Lq*~hA!>CFfn%r7P(<1a{h09l=C6yC98OmA1D zL@1DTE)My}F#P`Be`I5Qg>99;I770pJzTX zzoL48>2gDyT-12}r5K-F^}JqX1>U#a@gHO+y9SDdlO?-`Qg-`GfplVpH7QSZpP5Fd zq%MpWXF4C$=yc2Q$qK%w-BjFvrdR;QbcMPa5IV8e^PW3|3>GLZv=j5HLH&8QrIHMN zeelJbUM;iBO-1XsmDXhriUc81Q5U17&y+M*jI$7p9UZ+#HmUrO1$$?& zIq;r8W{5vNew6b$v5}E)zz1*iyu5EhHKoRHmZZP%29#UK);dQfUkN7t-g@=s&E9xk zXS`~Mrol4l*`LQ{x09K)Uax-_FO>T8;r)dNRadoxYzB6FR5vqUmNW@A>8D&}lq9)y zdWN$orl#IIzka=);IhdqXXmnQI6r+U@!`XZMI7wu^?@xchjyXsQ6f`piR{GH_HRz@ z?_amH(98_d`6)8;ZENf2{mR{~hk5z31saht_tJ2bPj0?8x-Keu4P~W?m*#q8 zPbjoH3r9ISPra?EV}UAwYPs({+nQ5pOTz*k*qScRW5$S2*UZMFRo~qqj^)%AySeMj z)ke!Mq-Xhrm=(1^^Tq3=R=pSnAEAjFKX7?sg-SD_Cf*8jt^UL>5Vy!u#PM(7Yn7F~ zlvX+;RrGn#(wroYPeG}#R6*^cH(rTWd4F+nd6I-qnYQ;a<@X;LPoE-Q)z#=^1;vwm z3bfjOBq{mY&?kuGsa?*a^Re=`gOWXH9bF6(3;pR5)bdyLJ5ub{I#>V~pwIV!Q0wxr+O`pStBfU8(W+4lHQ$azOZ71f3Z>>*Uw+1NxAWH4mMFfd~W42KK@M- z(czDuQlpDi(+F4$QZ$Fg0Tg>^3=oxRBl?M`2zs<&VFf_@GP(hSNFC%b6sS%yK+{_2 zMcSfC@QC(x{rN{vghH05dbw@vG%q=tyL*KtihPEiyG#F%wV|z@#=gv>bBDdPC<&bf zREv(Nr6V^!E;Y9VR|yHPj|O*P*qk_~t1Hf3KUR~ig(%`XuIH>uc>-&LKZjmX#7oA! zU+I&r=VR1T(B7_daX34Nf^ySlJ)zPDf|U1tIlG*koujl^#J}0>hW<(5^kL-K5XK&=)%-x!>4C!Ona;7j~|a$ zdE*yrbtRZ67=2TXr6bTFr z6*XFyX2kOFZJN^{ws6f79Gbd#4X%B8qwhif@6##&tZ#2rtI~?6J-lws`Q^*EGjtK5 zhE@^IA3_EX>rcdG6LuDg)9P(7*w{qPE6$t?_X+m4cAH{YKG#YK``mQg60ma1ZWJa~ zz`yIavmjgaNUfxSeTtefH2exbKaU$`<$X4`7#m}RWjoO<$-%q?7%=4|l>FwO$6<{nCSQ}v{y{u86pZTFE8AnVbb27EznSrtvM9(BRW-~lMu*p zKuGI_1N=FVfylp!k|)HT_7kVo->R{tvAxVX+2ps~z^0?~@@eXlnqroW_EarSI*Ki5 z8v0m&DRQ_$txiyGVR)>TvU)gWwGyfmdA`*B#@PS-w0O_B+k90)GA@JDF|xVkdmnC5 z(Qqxdja4gZ8g_!rm!BHd0xL5wj*i}5%f}kbOw<`;b3M4rX8QBnWjp@Si6GUBw{K?{ zR}cDRh}f{9=|^i|r&cGYA|2kxCZC=%22(Su)3Qg)E~f;WH6OSgwsyOZPSzT`NCd<& zjR%`QU>xl$V-tIY<2ky_&AiCdRJSc2A>s0}UxyQEe{E*7P(}Pzdaz(Jw;OR+TvH>e zy}jcB_C2=wnH7ZxS~Z@TnJG=sn=<;D*b)maZb(S7lWL#W6(Lb_8Zj${5pr^Jv}lISpOwA-d?YeHk$f2V!Dx$CP?z8o#69)9M|lI2}knU4PpV8G~Y`aes9{ zz9dUst+AK6xL(Yf`96tkGAg7=Uu1SxU%jR-hiTytr$o;4GI}G7ky_DCFl< z@b`Y=yrZtpdjgV20>xzx2{3O7KeLtF$ zM(aXIC`pMSEo1hN>N4M)Xth0u?X@N!-EEy(Yh~v%qRGjZ;*EZ|j+f{6!x&WG-zI;R zPu$q}4Lej3mn3kr?uBvAY=F-Pw9&GzUN`mB1l}TjtdqUAFUrzg&3@6GxOH_L9$Gf| z9}+H3Z*A7!l8_j$$xcqa`20|%lDVS_2Yq8fqCe~1u*#qDnDm9~4QXQ0Pti8(0x6w{_4!EuyFB7qQ`7;_w=7n>gCTc zKB7xF;S$g{WXUe~VZ^Z|8_JkYYAmZtD4ZTCWWNWml;^v(^+>Z*#<_eb&+i`lYqP+u zt&{{&%ZszlW3mAE8`lt4R$DuQI*$spy3#l09#*;#nvEt|Oz{-liw(b0sbQo2tzUu1 zE$59ICh{bXNa_uvyy&gXjcbf$dsk#rNw3Ke8urHGmWjzeC*Hw&Q%k_-4%y0YI`z*M zgiPsqcwXy!pX{@czCj7X5T#}+UvfBo>G3F!0cODA@PSx-D4(`FAAioSZq$YN&nRObqn%wR4; zF08QBE^?-`()k3B-Mpb6Z>V5_&r*@F)_JC|{&%&50on(Vm6a@QvY|rVjq%RHp%Jr* z96mN}y94&XaH$}63xm4^IkcR2`S=`oP8E7#d%QXIcS=5>6jUM}~t+$Pf z+bE~zGtSu-8kKbaXMqu_N(*TC{M$a;^&S^qN2Gp?xVG5w=urL(0fFTCo3FQZKHbl* ze4o*<_UW1P&J`^!FyQUEw>^SI^4Cha zp{NyKMZ-vuqlDwU9n;;N!-Ev}@gkSE@Vo0bgg;z`@TxBZe`V`N+wPM8#u%BqFB-H$L&Yyf*zL5oH~3{tXA~i5+c*WE-zlTk;?3hhi@U zM6Lp#uRrdi4>`q^yLT;Zy|QU0Yh)E{C@7P4gu@x)(tUh!RK8t#^9BcXpqC=`5Z;NqKA*lYa_!=#;JrR8;z< zrQPqNS9yStt*vt0rHaZG77Yj>6jq6@aF7|Vehl4M#hARVVc}+EG!Cn1)aG!0|9FF| z_P{kD5PzZly1<{2)S2b|wZ>+l=xAS)AxugKu|z&p9A4+F&0E)Ri-7&vxQ4EHsWpCtE;2(FkoVR(8!gCuxR5Z9b`Ni=Wr<)pd6IVy*Z}HgW6lB`2 zQXrnb_N-H-J3q-q2&IdUzBd#T`3UI%LZ|VHSz8bZuG^`<3_pH6cL1ANvk%##+y@*o zrZ4g64?s^8NzWkFpF@c#di}0U&teMG<&=H^V|0z1g3O&7vAtf-N%iw5&Svdn#8Y=y z#;Eq6IGHkZlN3wcus(J<&xx3Y5Ghh|_i?XW#S|zpk3UV{S?atiEZlp3N*n~`Dnw7Z z#G3`i<5jvF><>9rE2^;s^3)WO54`o$ZO^y86dj2NhKbh&Y&W%*;g4mew>_z~wQ5|m zhgqyr5Y5$_n3Xlun3l)ip0E*QWIW>GDF&g0|L}0iPt8Z?Pu+I|quFH;jbrLHZtAv` zndyEGleayOE?j1o;oc9Ygrll)X4lnc=ExDU$?p+Do0Gz^-1yvAx6)MA)JA~@53|$o zU=Ag8IZeJ~|1H>xNbnU0f*;=Z-L;{r1na|sC+vhGBC}`5sMG?2?`KVB7rVI9>m6l# zh+IzXQRW%%a>^l6REx*ygl>}Jd44U%m6aV0MZa=QTU)j`BTG$DA)of_%-V=;s^|rx z@%!@{$7F_>Zx7W@O`-}iPIfO&M|-g*h=)trdvcxkX(jgG;rSUojtUS7yz&4J$1e}K z)P_gEx~CFP2X;iAa0|4R`C%p$!YeGyaX;s)H6iDvKCe5ysp*vmK_qm|u zXq5>(Rmux2ZK7VTUp{{x;{yBpZS0;K6F9r6AoXpXcGi5hTa{eUy^*6Cp9PNI9P z5830#D87qbSV`E=pQGYP#VsbYe_u+(60luULcoHWPvXIz{rOU(CQJS#nD36~(Q5Ad zHqC1xv9!khIG#rdR&(^x-NDoas!6 z@6aDB!}lcO4!46>j4CPWg+^Bb17c<_RbC^<)R)gP={2m5=~%{c`hDCm332_UNvGWs%CS<3 z5|b!Zx8F(cIGp6vofjH5fz=i%`6KpbrH1VvYIv;B0$P=&CnJ!EhvU!)3u?g&R2o$j zp8~yOVx&nhcb=dAxg-39_-tEO*v90*-hINyVws#%b(`4ZT=D$(^mkMovoXkcLpg1p zojnx~_I>yQlh+~HD5#%#cui1SVd_kUaxQ^@jwTxh9<`JXxP~!6sS01R9{APD}C}dkg zEl*{^+vIU3+k>O|Ah5_{(ie)s#uQQaxzZSZbtV?+qJO(_Huy5C=xTPhKk1`(vKYzn z>df_T?Qvgqyvy|3+XhNpE;O+3K0-xpp8oEisA2dVCTx|OUN6$QCAiY(q`tSMH=Mve1TO%FFRUa+fT%U+l-Qiq}VY29K6P=qIYK z3%Fi5%`7I!qp+FN`)`vQr&ZoVDKe7p(HFmSbxT1IBAZ9F*bg73$RyV6Pby#R{h_%@ z+Fy>aTU%B9kaW4%ukD`wcC)wkX?LJkE28Uvp!GI~XGvUIx@bKB#25^;P*z*@r( zB06&0_#(Yx6TdEVE#&h9Fd0H7fB8%}_Gf$3H3~R&KT5MdzdLPoSt}8EoWs9a7bk%d zNq_tI8`Jkeso!Cc8>@*wmz`H(;nqGrgZkp6x=`%%-;IL1*zfOXezcg6`*wbLWYu-u zRvYcnR2qNluXxt`#C#e_TaUfEjVL69yuNZ6#%ZsXvsK}$3nZDG;V0-4WwlT@VYdQmO|{% z5`0j4I6I4%8sxCvPU%1*`ay!^m!by&(_o|)=ZEPuvC+^K4%ZiqgwUDjBF53t*PzQ< z8{=Un{*{QyI>pY%nQk|HU8`-y#^k)Mto*Ofl$+6og9rAw%wK+!Y=!f~Vd}cT(J{VG zlr_T(qmrpFsL=Zz6{nyeMT-~3mqy$!C}L{nrP%Ci2@Dq((QVVRj&gOTuX!|(Si zVebxatjbP-GV^!=bCzr3ws-3w-K-F9Co@^oaIYU##!=|Cx&fh!P`+A`Ul4~#w%0TXsR8s zmcK402JMU{CpCzEpqr!AbdPK2kU!nJQORS|EGp#nQu-HV(E@)CiC|(n*yvioGolDMhz;3a!{4X(^Q+V_>mYdg z<~^kSoe2yE-2_!vbSiJA_gJ0zUi6<3@2yfIdUBO#48(K94G|Wo&|N0w(h0IBskm8s z&tjX4?$T24(o(HTtal;g53j_swJda0xT+4}PE6ds>(>@mT|tnOQ&7C6YbZ91 zX2C!ZFuWK0^`?0#QTNyyzKzpntLYXe6(BCPE-x6nzZF~Q_Gcw4qkQ@D=~49^fp+?e_Mq>o$C(s}_r9(mS zE%jhQOS7M)HEjZFI2%AuAPTgGd4E=mr1w8fzeaG_V?)Er<2}V(NRspNcCcLlcb_e^M2+t3@*l7ovW;`{8(Z*TcEo z{OHu87V~#4_r=9MfYEy?ZW;2r`2$j*^WJ30+GzaOz|l6WzEte=o{^>r&Ye5c3*O6b zO-J6t*a|uaL@-&?*gY{Z4lBXTtCOdPvc*(#@>V+(lP)q-<0t^2%92M%PoLPC1aRzMW7jn#jQqx=-UTsd&B@mf|l&MfcSqb^{=w zL+C014TGMqN@$CJMNa@DTK0%9wL{r^H}oT%=54P~ZFzB!9<%G#yE^pGA~z@~XSQZ| z65|pIED)DRPYIZ=dj9@>-(}8(f|Wj)1k1z2Ely7~SVZm)$WBXL>?3H2STXr8paZJq zMNJ*-@0q5Mkas<)cK(wnoe6>Q+(l;hZjv#2ZZkPC2rT`2J4JeWsN0Trc2>OjjN3ko zo%r+!dc{X!5?5;U#Lri6_WL|mQPUZu-&^@`9rt%_b=jMn=#O|wntae2mU~pH2}FZ+ zh8Uq8bwx~$e?U`-5n}7&l*8%H^XD8cWy@8h;cEZ(2{j=b+wx#O2No0r%qKgSDhCI* zcKhfKn}u3L#4+XMc3ho}X(#Kvy@fBZwx&FWzo8>OeR|?vQ(#!@egpCF;nA)P6e%vp zQO=GZODBFTWgRUOe(76aXt(Zv;~rR}px|rMJY|%NBeYDJrjfgItt-y$jouvASLl>y za|>T(27SDeD%#YC>#zxvptk2`Lt&g|1fA7>PKrB4^<*t%*>KY=hB_M>ngJmGJA1!2 zBwI@BeuNT1r#t-Q6S4QfI@asD>m*V!?}t@NjJ@3y(xsA?1_El<^$}WQN(Itp})WBCv?|Z zqgUInD58Xir$yIxkKEO5ou6smS=ipA5ee)aw}Mh0ughXS1(P7_n+J&Y_SK-u)08QDr|+j)>=;YxZ+ZFnRMX zmUX}Vyd9s%2^FDSIJ-K)ljwfN5gLZIIK*y83#4D0D~oCFZhM>1@Cuf}ev{Etk~Nm+ z?$f7G^-6R_6}bJf+&nObW&3!^(uyX}H9{m%P>mQACq?waQ*|npcP`I;MvAxRg1r-5 zjF3s=-|Y|^fCZVxFI0M!S6?dGjozD$;oOVs?yIfAq;ihRD2)=-Uw1#dJy!OnoAl3A zFf7dj6V{ykDIi`Y+Zg_t4D`b-GLZUim}9&oeV+q?;nO+tG#!7Rn6>de`tp?akl!6@ zvVZFZ3&t3OeR0vE>3eBma}&FtCq*kF$fW0RGZ+nnd8wgLi-gPr8v{dABE|GvNv#5q zh}Y-NuKW1Jti9`s3ln4IMBD_344H1`rusrAD1KSgzZvH+ozhaN#5KGQQ^(-dD`G6| zVA-&@!j(?;6g45iU*z-)k4I1z^%uTCuw66all>2jCIXwI>60h<3i+`*1s4G9I+}Ph zsd;fbfSDE6$?-oky!q$PS6z&or1W{?erpX;0=-;Y3&2~E9}ZB)(ccG6H!$~gd6!}+ z+mXFHU>*syZ;lDVdWMUC|Dl%#BRe39v#8<<_vfuS5$+cRqz@lNy|_knZ|O};$tUcV z5_BpmPHA%{uSdO+7ZGjNmF@+Uw^3JC5Iaf=6Z-Y;A|fAg8bLST+k2-WR~d@eF?VT| zI^D!QczL0@88pk%z(~MO@bSU$gr;c2uD9?X^}Y}h5wRkCCthf~VuCJf^|@nV$5n<| zB#j-u9i@_oU;g^yN~+k8wI#LCGkiq{vWh1BQ($rsnYoJy5KjhLf7UV~T-B zH^(IJ{j~r|hepT;e@1joOw`BqljRJdsW)=$BPJ%Y9l2GMw|G3VFBA-t2XX!L?O%Js z>A=H?zk`c=d@Ru-cyfkp`U$EAe?A8Z35lpj(KeSJ*VU&bfNXI%%ZM5wBDYj)&=t$M z%#4C1`);(tiW~uzC}3;iIjCm!!$^5FXhYUePKJ*KT?Mvwqx}8$j@Oo#RiI7OETcqr zLx>u9H00~5C!kD!Um++E>i_$r9^yZvyM6mE;_BAFKmPyuR)yElr9>CnalCmWJC2^* z`}5bYn*uSQc&vV)KrCj#Q+J+Kf&7)X`8@)nC9lbj>pw`j1IQ5%4Gj=J%PX)1(`Elh zduRGqy_3@h@U~l3M@1U2;wy1vfscoC)Sl4LJiv9A zF9VJlylK-WmDTXhLCX9Y}X4BY)c?_oIiiQ z@<4HQJ>b?m$L7oti*0heK+tgh1~!H(n9Z@n4(OfYL`6G4EG;cvd4Lz_1UYH-`y0RC z(LQtFugQ2`T}TIe=d`k6k{fPaVgjH+e&w#Y;ACxtLE zeDaGM!GkLXU|r{p94{v)=h8k-Uf#3Qsi?>C*vH#rch!G}!B=3@P05Bgmml~ez{aF| zw{Ev$#n!u@vpH9QFhOM#?b@@a1NJK~zBK!{gzy@8$3R+dt6Mra#5LC()ma3lQ5J`x zA$dSpk0sTH@WufDo5BLsqvM3_X2jgj=x!7cM891LUQ38|BGw1Re zyB;IrApikh)f$oah*t*veCN&`Ng!tO#&GOGK_wq8@Yta(#G2}3jfdP}1$~H7r?$)v z(H(z-+pXEzXlirCPXK>IqV6GLppF?H5Z4wk57-%0`L~P;dhp-@WF7d;^CwM-9Mkf- z>NEGQ+2eGDVt)R2p^p*|U+Lh0dhGZG8><-}&u2St0+$1W5flsHM=;koNaQ6aD+Xc0 zDkt_cYf{cx{=HzBosNGMMiS5>{J7dUy>N}pQ>}(xWhvmqS|T5)?zpk7xvu?HK_3wy z%Kp82<$$;(oFLub#S*g2F=FqylaonsVjJm6Xx8kT#2JhvY10arM7iR8+%VOW18p=Ia7cO996#lUXzU z+txGB7nuwehi%~SS`15(mGAx8`_}4lwUcJfw2FRvue{uw1X(e-?`TZ{T_$p`x{^9` zWn6vnbPR*DOoiwEM~4@a)pb0NOU6dV=&pU=0eqTUX^fLq_3U&`A()8&5R357 z!cnMqUPJ(MLR}`oPqnmd3_z`(tAR z2!fB!+Fi=cReCpLti+T;fdF>8<1_rkh$>63-(_T!+&DdHuyQNuUAJo8+2uL%d7FYg zTiQc+QX%)?Xp4D$Q8Z{Ij$aw@xz zPQn9jQ5Wi}?A%-zgE59heP0k?v@MjC)ZD;%?s zOF@ppMIVNV`N)y5O<%-FV&(+eRvKyEk54@IUQQjm^++Z2u50nM$jr>#syro9na$8) z&$r0T3YvTW$$+cimg37668YI%=IQONVv!b;*?&Hvn#h}6I{SFUpBI>AXzJRTaEe7% zZOxeUJM(1ZnyJCBPH4@`y8Ja_r@(C2Du^=~htsd0-BD<0MPcJo zcE0g!4Fn_%mmClo-cJ?FdmYYP>nVIm!Uf2yiN`o#{MFM`l$7);?|m0-{_1E=b}}$uo*kr32(*uJD!>N-a4*pOIwqx^Yw=h zrzfgGDso{X@$UL|S3kep$jEXyH)tCmeg{cQZ{Q#VXgE&(DO;xBxOQ#+>syXmxjlRJ zYK2-AUq_Jpv!$Vnj`kR_*dK2jF22+kAm1SQjC}=YYl#{coGYIOi;@W&;qij|Gp=gv z{{54Ixx$|h!04^OwC(pTLik_A0$RR&q)1=Clv$^MZ7W$|7ue~Ibw_^w55ScmQUhD7 zhgC<*Yi&1djWpA5I7Qyi3_Ema*^pInq?oG~rm&Is9i)cJh(~m;+>vZR%I1Q*m~m|AM5eM3g0d(>b?1wY^dwO9h;rglLdf zxmGVego_V!m|S_@q!IhqFND?P2`{Af_g=Dja|hY zJ2kb3N0$2-ijO%+H)g_f-pVO0=~XZFhkvjA-8kTJWn~*s-S+l(54K{7Wp`S6CGqpr@y|*f}C&muu}Ry@)xKW|6U>8n;idchk)IHP#eXamw=MfE>9X9#1Z$ zsx>*W@1y_78Rs^Bb;$DW*D8vipU$=+><5&=k)(*)z(QgEbg)y_p27m`A3#yciG!W; zv{+bsg&EWX<5NE_SB)80!|HXnJbbZtcb& zL^_M}4%#AEdIu*g|3l&qMwdO}l9D>&0nxbsfB_ys8r}H0 z%Fgc{RyRk<>o99u*?Cgf(czRdfR@H=S;Xl_6BHDNdFUoJ?k{#U$*%?y`$lcUf}A~^ z!BpR!=BV{!-`$0huz@7?E9`?ii?fHQTeogx%hl}a)95F+RP{EPJs zM3Gq>yIN1Mu?oJp$Ot5+Kow*sJC*V6mfys)XUDsb+uT1r7p9^*;m)Z|yM zUrX_b+2s6@F5_G`Uc|mEQB^)~^cf zB{wHPS^Z8!?V_Aq|Mm^|E5(lp>Ct!iT5>cOqQ>`jP0bqTmn*2Sgz;p(P@LiH$X$U_ zu90B8V~GHePVpRc0`#S%kcvuu!VHuX5mu7^(SvrS>th5#$F8p4R9_!{CEmrri}WCh zjg5`jb~y7JbwWa|J@#<^xKD@rzb(H^P?x;_N;9j+BOW`~uWeqnhHkvEi;D{oIqEaV zsTCBIz%AWYuTJcYz_;TtsN?A9{xWJ46V*yQc2qt(tWYPcj{xLXdD4uEN*za$A0B$a zij>HlFpO3!d+spT?A%Q1p4(}cDOo?xdq9~^Mcm^!`!M?+OR~pu7Hr|E#6$s>P1RR^ z$cZp5dFqV`thV*csejDx>Ab^t)~rj!@ka#(+EE#fQ!RPZYzHa)1l?-WR&IcfEWb@9 z`)1`&xYy3aO&ulwA9N zx8ir0G74YaHr9EU;j*$|^v?`0S(LDhT8pgI>jfDPBDk@bvQ{8(I&UHl6d27=s@=h0 z(@j{zDM}_;)R9yOUDtYf9U(9o58K=KCIwZ(_$(#PGQXMil(G}f z`3JwGj5E9{wAXE6jI!Iw`>^-omufJU(M;P5rP9+H@1~J)XU%1DYEeywAH)#npjquQR+J>HGc$iQ@o9zbw2pXHObT=iml`b;9Y z=i%XTA5mTs`U|-LkZ`5YZR$t96~&!AiI3ChQKR}@EhwmlgPICg)nEco#*_3_A&srB zt5^FF=VqJhhs~U)6=6psXKcM~sX2+o)YLTO(jK$8ZasSFe@+{_V8@yzz9iZh>73RV z{|pE}a3JK=RkLe&^@zSu$cImuGR3%5B$Q`dzaf@eisypb)NYy7z%@&GM*`FxRhyBK znfVl~;kvwuQ-^36ZcQpvIJP;??-M;PN{6T3$2V`*gHwOyXEVM#!iDko+ec^p1t7Vq z%FbK)d~!0pkW`9<>n_27{%dmYfL_CQyZ2UHJR3r@4lR;6m4* zv-V}zMfs*_jlS)z-@XaZQicg!{HFO+0x)d`!Wwpom?ma6ju=}~PVa$Wc;?I*ndyHg>9>LJ zrrj*Kc{2*z8)|*t&32plN+a&XivJANy3c;RexA|c4ZMLRv9r%#xUj*;=Wep;?_tq` z3-d*f4~HI?l^LrnweUl*qZr{fV%PZM(8Bcm9JJxYPgPo8eyZp{^ddTs zUtdyJe<{RVACa(C;OE!>|LA}B6)1@d{P8{gGWv3X%^Qb-va+WyOK~QwMBSf`A>jrm zkd?N+s-}T7Bp1=?pl+i0<@}Uo9VfrNb5wTta5|71uT*uU(caoe_bufC)RQ~#N>wCG zbdu12Ya~j49w^>@d~!IbX@bwDjT=!JP8LnQZO#Tm>V8-r<`EaS;z%%rEe~2+@s{@X z_Vl13mxk<&}zmJ`cvy z6o?Y>`HL5QjEn{b(cKE@XA(ks9AvzumD;uY0J-HE zPKLPja54ykgRC`W4umf#w~ss@Qj_fLCktqlsI$_;!;Z{M9zYMWI6OvE(>D5=c*cz3 zKl{BO&7{aW6Zp8Y)0_;zmZAbEtEfDC_KflqOUDSx92(X3=Hxql197WDf~<554ms+ZwNOQ zFE6hF1NK3%N9)m71ePp#eWPSlR8}4&_#oE6S~1Yqzi$#xR6^u}ty}f{IrhwjPo8Wd z8T{q({DG&QiHYP@K?sd(E%q(w+svKGjT;HpGEoW>lQu2Z(n^RJLjV~{7L_b|gCmFc zoDW3yLQdpNuKw zZB4&M9hDDsNqUQ5&eJ|RySONIwqT)B+SF8d{$5zt3VL zNY}|ogfv=RCQq4Cd+)?QQBhHmk>>bw3Y`zdAc7!Eg0ANF_3I0o&d{p2hw=uRuT^2(!JcWvPdZJSk$y62MITCW8!e#BLkb!p zD?5v~j{obsrrxbz$}Q~3NEWyhlba}tkbSMOLDSPKfm6$=ZMmUCn^*&w*qo)jWFUc8 z6IOOTg7AqEtd}a@J<-?GW3k@nww*~$?d<5_8!zf<{(WIV0sV34=sC~t+)h#;GiBtga>=lTeP#zicG7D4F6ckG9^t4%$P*Ly;Vf$7d+_~l373>Wx!5{7#@8Eh z@_<)}U$A%YUWv{oUB-Vyyk$p>5NaP46+_>NxJDxlNGz$64whDzENFYmrU1YYG2t?(^MPGC0&)`J(QJ}~qhlkOf5ijIa?vx^#XZ>APW4CGYpdd#e7=P` zMbLM%lDM||Iy$1s&XRZf=cWcCud;)JoX87#3=5NuD5d=K~HZ%U%Y$NZSR$pePwt` zH@x%{vx$leZq~>ET?j3>FIf1lK}zHmyG=#`Mz&Ad`_H+oEaCCRSWNVSi_HQkjkVY! ze0(LhpOL$^^zL0mV7@jM?H^mQ&&t|bUpD1_V!@3Ye?QG0G?$(kAbEJ_86Fl$`Fzds zt=HOw=HGF#DnQEK!($E(wjX<8su=%1=*A7LTf?N2#MWgtYB2Aq z&Cci&_cuz>bfFtvM`Gp_>>;uiU};2h>GB0(Gjv$r-o4Yavdo_-9w_ExjCJFmTj%`1 z@2so+iWO5hQTj0|yx5VZ|DYfwM2P3jdnSAw^BLWl$LnN}|8&0EF6P-7=Qaa`v4)1T zK|^D$>8de_?{6KRW^&RWzXLPt;7FOj+|T&ednIt?nNP^f#rr0sM`pNy6GK4~VK80Z z-BT*b;%uzyh+2o@@bR?OyMu#q*)bR164AoQ;?A9$Hx=3!GoE!`@VdyW+BnoRddE4! zw-9>JnF!C@*;uJOWy)713{gi%hqAJ=TC90~fWPwsmm~|*pq3c2w;)DnX+<*)qd%Rt zB=IT39AehXmoF(b$z4$2Lig@{c&MH!lVk^!uGng`>1D3T?nZKcZH>;&CtXZG)Skov zg?aLK6gyF08MPXk%F2~a{`8x!8Y$~?HR~+y*I!Cnnp#OaN>)BL+t-9(MAN79XMrgLbDVa^KjaM z{unifUEjUlrBCXIWWv5+2STe(BmS3f{Tq~qp)U&h>|;pcX?*cxgw$&Q^GfC^5s8Do zOALCr^lm298!HO3#4a5>y3i_U-iKmh`W}ZBKN@S*vuDpHCay*{CaGX1nbG4*$(FCI zeZRM9BQG2^b)6H0C6d0l**tNxX-QT;jk5u`B$9z~g;ur7Od{Bl)o)dKzwz|_SK1(i zjvev4Jg3G?TU%RKmX6dY=N(z#D1+qYKV2_7ZTT4q`~7-2TAZcHpD3 z`rW-_maFdDv**mx)ab&K^g7)Mm-0=A)_QwK`l-dAIz>djSkzmWL_c!1&{s%sw%R)6 zI6OmvQd4o#Cp7ONU8|BCvbfRfRa|HrVjdR>;~||*U0fcesf4y@lCaZf<;hg*ISC3$ zK-Si&wBu9@QgnQkEU7GohmVUp==fMw+fA}ZA~Y;`{r%oOBgG)JwJYmSqhaOc<#maf z<)f;osQcQIHAp%_;I5(m>781{^2q(UJR~7g1IQ^ch212c>s;XXRMH=cXuZp`Z#||r zOz7vVu};Onz#v|i5z?!J%9i2ce_5aL>UV|g)!CV4mwLR&F`ZPXKP~g_p(zVqOFl?a zK8mh!MsT#??Gx|*L)!_+qh#}Um%@FcCq)W2fh9*JQ@y$UsJojI}RBL73d}h8FqO2zr8T-pnBE zaRp&X65klkSc3qf8&a50Ens3TCIWo_72E76Q(AGXrl5w|C)jes;zz%v67iz^!xC`4 zmsdz|FvI2w;xE2}NBDac`|n8g2)07%o@9G@4H*Mmq`$wvp`jt)Z~XRUMQsVLD(lv+ zHFEkxrC55@D6wz7Beq$L&RCXGYd36&Ww1+;Mj3|ybcKN;u2>2RiHsny4$))Wyfjs2 z$Plv|K288>1Uk*;*7U-YJhcs9zA(l47e_^w6$4^molWSad_S5qI6+xi-it>)=o1qb zX3UvsE|%ap&k%;-RmF6OfGv?wbHAoMrS_TEiWGtx-s}07otP2LT3>Vg=`?as zbuf-uc)~>%_A2I~4s5Gc$lCK@!fdQ!jl?t2bYi6EyFNbmlJ{MncERVfZEX#gx4|ze zqTgHUpR(9`%EXlW=TJ0G%dyGmvgbk{b!LVX$zsu>MT^Q!BeG;!rlqmZs96(E647r5 zWKx-lkZsI-+FDz?d?Ro(?_1^W9-H?Ep05?mCJQZE7>2Wx$yqrNBM()(D`yVc9uwMu z8>+&7)HdwrXF;@LVFUyp8=Y~p^CEDuXHwBXYH9_aA-2~-oO6(i!DTSX<7f4^K1Q=EqhFn zM96Vtk+{3tl1Df(jJruIHAaE@@0TbRxMsz*Yp@^sOd^2wz@`Dz$Ay)ob;P3#1|a(4 z{=0YZM(8Ie*71`CVa=lu7DfZ0ubEUX7uuV1}XaEx6UHFUkLc)YP$7(=i~g-iW- z=Z>#J(9aJKqRtnli}B$Zz_A!{F^Q+R)Kwuubj`){ z^&45!rUiqKV+uBo(b4=ibAM3!dxib+Z1d^Yu6n}f!&g{LdE%x2UWz>Oz&?Lfb2YH$ z-&@+3#3wGCzms?2@y7XgtO2Z>j{jbmV&U@rd%^kPGR40aWG^lfp1=5T{lWjc%>U-K zsX?CyOibF}_n4#Q+=rQi!^18vF8%3me|L8m-cO`6$|JbSCbYL zV^P_R$+LDB$NJ6+LG__aLR#8wtC@x|D@!gTn!|+Cabu!n%(-=ZyqaN}PCbUxESgo{ zck@SDnxUGS+JB$B*GP%HMk%22_PJ44GPPv5O3G=0QeUnst>Vv(9LBYN}JEir#m;`{e2EhlTM&oJsiRrpi| z1qI(I1qE~)zSh*#WXPu(7#b>z#}Z~rY)L=2o@wf1#V29PN*Bkf5!$J_^rF70z1@AY zo`i$6Ei)Ib_BVfm(#XiD!hV&NQC3_W{(*P^1%Hd+W0ld>;gST`t>(zcyOMOvw&nGH zB$UrBCpv!!_x@C*Uml>~AU>A9OG@gzIVCbJ+$u(d^+MzaHU$=47nw!njT<-K{(U<^ z3Cq{g(ZOul|LHne3-{en@8kXT{l)Ia+Lh6TmN5ED*aUjFMej|<5MeRABILtT(ek`{ z{pz*5513!e$`TS13fjy`nVaWS@(BteF(-%gN?A|GvEHU%Z0wNav>2;&J+#Kg_SyP6 zKiOSwXL7jO#>B)ld%x!66@2q?boNtIxX}awH{<>sGebj?TeohJl9sz2+t+Jxr^&<} zVUE|ekf*!W*ZPV#rE##7!~kmg_6KwW7!j2NMSPr`_9q9MzBh){OU!2tO{i~>le?`9 zNFyYkF$xE8*zqD14-uYh-k^^Y&cZ9Y!^U47rq@c{=?4wNbx- zfNFUrR*(S_qH>s9jp=&N-r&=lscbP)YUROU$D^sBpkQrny*h+~ zh|Q79_*3`hBXO;7FWM<6D9Cqhe}BxSgWZKhNFf(-nwH=$Atl9UJ@Yw#1Jd^O>(_H6 zw|f?C;X~l{;o%`X-^-UT4MfdCx1t$mspa1)Yv7`B8H3{E<0B&@gM+6gCnx9U=a-k& zRoWI77I+=jaLxmMW|oD8MJ#}VSf~dQ4pHUoO>SOK_!#SU%mu7BJicGWaac?Tp0Vv} zIZj>?IfV?8d^@u*fJxCXIgMY3p!_ zMr%0ZteH&g^Uc+g@$iV!^76TvnMk7_H6<@Hdzmy!-^8B|4GwY`_q^Yns#nYIlQL04 znxM}xE}IrM^X-fqJ+K|VAvtO&iMR9ZalF$F{+U|U=&Dn}(+e^(GX7*-?%Qo#Nr{Pg zrh{8tIeM!xhUIS#Ldhqr+?qaeYPrIj{Xv53R_gMfIsc>W%J_2aJr_k`YYa z{D441#0M4CU&o?%c5Dp|4Mn`J2n0dww`6kj@Ica$?i=Xq`%?;zo!DX#5D>ufNXp8V z@bP69S;Q}rmFv{wvF&zub=|=X$MHHyWl~d9^Plc?8zd@bC@`ZBbX;94r)s7Ug`ea! zSe3X(uF<>SQggI3K@BZ-r}1)gM>+A(g)PDELjJ^`)J#3u(BJpC{9)?hd5NG~>*#9^ zp`l+QJKkX@mF~~cY&$DpHy$j|cXvlu>UZl+O#JdR-mSiKp9RCtJX@9(T$8I_A~P1D zc$bT-MA(NQ?E9&X)7k0%?Df8%ijqs?Ir3@WDqR;ke;AOyt-J!6wA`H*aA?T&zVm^| zBmJYk^WuV5PLYc;2)pX}uI}@VE#nS><1DKPrA(A7D!bTh7(iZ~0RBdzZ^|W8cYIn(&NbJor=7{I-0{-M!O zk&%?t*UbSP>J9y)^XJvM$Bp&WB7#X%F7z&Moxz_-tCIsO{NXa!!!0O|e-t{6Q6V(4 zOs13c%ki0)0kse=s~p)%^vp$tg>0I}yliZRQ0M?g`mVod*M63zRgFNf>eS(6kSGMT ztPP>)A3geAYXG@b0ywp82TxM$vTyG6IttC9+xRU^kp1p8h#ZBUzl*lVd#fQ4Un%&9 zN~=rbp7YpcmftJ9<18;?ZqCLF^Ca+$$u zbU&#dw>drDk7&_y-4xM%r;41o@ZFK_i1IR4o>Tpj1FLknaqmy+)vo1TU)6|ccB8Fs zSwYP*Yic^WrKx&9{5uS9z9QY{q6}yJ>7BQ4pE};m;K~Jbb|Z9!qBBE$a(W6yM>zw@ zN=>c5Hd5vrlz-{!wWmtQ^*TIeJJrv#5V7U$HvAY0EK{NWT~Ny zr|Q0>K5}miptLakku@HDa^hN_Ww$kE_ho9`^QU5F3okLT{NR-PfUBD@D#MEiK>wO{|Y((Vgo^s5a@(nVHes&d$~- zllv?5;zdn?ertOyw+&?5dEsb3ahPekF+CZs1$m^!n!;Bp-goqbU1fuc+j8Q=B0h&s z-DdzQ2}vxOX!hovdX1v@L5XpV6fRPTcP*i`sXxMZv2SAedP4=2Xx%YfhE*@bFmiJ4 z)qS}opC-c*=i*xHg)7IaoQnvH7#`xvex}Ok^GQA%gedd0`4b{ zdU`W}53Fo#+15F7Pd0qjgB0n)7M~e(_lUu0PGHdn#Dyw%VrnWeGRmUi>%A>J;8ZMY60cvs z<@?2`Ui@?@;PdA-^_2y94PCj zM|MuLkQH=e%z}c|fWgR(U)T*dxA(p1!n}(=je@liTDNt2-+?LR_b+re;M*b zT25hkxMaO2LtgNBg-!Rhsp-n!B(XMGuJktq=T$A<7lVBP%QW7oI%GCgHo<&S6^L~N zgKFa4X;LDh2f4aKG0Fvl$&B2}v@R3sC}ZmJ3hiGXFFv}*Wgc6p24$X+u*Q10*i63C z+&3kSNei8oz5zA=&OqoCYGh665Gu;8&bLV1T}V_<;2;O2F-?Nj>n4 z@S#^;Nr|k41S|hr2>uvWi>lplt5d=MUt|HoZ&L;*ATg>|2Z!79_}6GTI21PtYro!O zn<4gH@nG%%mZb8L!VxR|*(1@f2e7^PBl^$oG69(Zz6<%USz*t}&dv_>vvjqhtc*6? znW@5t*iR-+K5MWT7TkgxArDjwsdOuW@QJHH8IpOwiWx&2Ye?6F8?J8er(1gDhk{hpYVR80{t>Xlk#8zK$ zNC@%(kk9?#CC*=n0*HX!rQS@S8>ypBB6Q1pvl<#2=W%LOB%+})0xbzM-1lNv3iLv} z3=CN{!h36@$JJ-2>$4GB;S_Y`5LUb$tXy#zf^DAuQURUHs z4CGM8y_{x4w^h1ZTQi-u=eCBirIh{HbNK`Vv9xi4|DKbnYqVi zU^-mP$jx0AYWQ}f%&f(Co-Vwoh`pX~7&@Qu`;iF=2~kl-qmvm?QBlJs77-EGNZFfg z-yD5^={Ykvl&{-3H8o}2mt};u>fY~nEqkk4%>h^wM6&^aktUn4Hjtn6?4BTV0uoW6 z*BlJ=5aNb-U@MGXIY*O5brfi9zxkeh2^X_;qKR#D*s-9S?M?CN3e|ch`Ut3o2)K*63o0YPZl$5h` zd0JW;(L;KAN%zo*2xZz1^#Km`7D-VSBOxn7=1pSdDafJQBw0Q_K2a^upKSTSCf4(7 z>L0wg_~FL+QfNV|MI4f-`jDGjHK>@wY-k5U4vLE#G(=2cu0ZsZzU1+B=(pG+GNH+f zIU_btR(YAdgLhxt^b6F{2$73g<`!+RWW=~9&=$$Y^F_tSn={oS05+l9`_LBj+TUvYuS8C-g6k{Ege>1cKq`D!Sj-D$GQSu1YS z4cCo|K28a%N5|B((a%!r9z6-~AXYB0sC`Ily|uAn$rc78$|L0a$x(Vd{Zo+ET7y6u z&}Xq2<8vCibN{|8=EO-PKtn+xv@NbR6oeFsl;8lVctNCle(I;0vUqtrGWgM)oSP*)r(D81fU(2l9bGpOYw~ozJ2Q<4i!78f65O~QsgQTc$YSE+GzU)Sx2P$+h#HW;B)(?}Za@~> z{Z+)mnw^4x+w$q)-;Z6)b**2K=Copx-`m>*;0lb0Tog~ zx^e`R0+-uA6RnrXdd1#ZP=lRaK$V+bwIRxscSsgI`Za-TE{&ncAX zXJ#`M>*x?cQ;7KdIn$6(sKbwyV-+DdUs_SS1MqCA3qB_GHvkUhTMPY3jM3oEdSwCT z#l;u-ik(|_SQ{J9sy6MQa!|idgZ5}k)Ya9swzl@GOD>}7=YsGpk?c$35c>tZ%C!9+ zgt)kGXP$LgnmfgfyB|xVKSUAfK=BR7e)G6~K#i(SjM7aH6X-DRrfelHF8;alqql{} zLz>XsOIidL{%`H=ix=B^MWG`}^|?mQqkw4a@85^$^yLA13|&x!^73d!iT7j}+HHHs z8B)iexT9xu)D7q=fQ*xqQ-d);Isn|>{=Tw8Jjj8SFS~=fEYOFkGHGnZh1$O^ig*L? z-X$zK`Bz$w=yU#u6yL*a{3axxhuHiJd*x3IBtv>$>l~pI$?JJGp!~j3c)EP9T9E9Y zr{7;7+iZsIcPUeoli4lCPJq1a?e10qZvdd&+TJGi33WVhaEi}-@$1(w2;-4d>6ow7 zes>s$s*y-g?V{o4f`bbj*2j{>0$K=-`g7v!Ea*An(B=SybAxX_w0Tk4Gm2Rv;~G8@$gY&V`GDZgP?Pre~zf= zr;~INE!V%bfi2~3VG=0l453UnHRl7uNk ze}+7bZh&pjc^pXsDwe3E1%d)Z!EIevgpw6FtgF2Ms5;AprsAbbp+&nz}S= zI6^8x;1p7xzHe&|N8Q(6v7b`YXn7x zlHZyA`E!+g6X;if5`~>;cpDAmO-8X9j=3MN)++THcYgl`DethiqS~z|`2of^N9&mB)aE-NboF`{#Mw;>qXj%l-* zJLoI$Pc@-K-TZN$Z*TSrTug3*me~nt;nWNpkQS*}@xr+Z0}VWAjb-kq&X7z>O5sq5 zfF}ZdA1bjBw4A89*iF~^n_l&}J}e@_VQZ!tn4bnT8vOkHAP51*{QmvBqhlSYqschB zYH^&}o8|5`BoHasKw)9wkdP2aQy}W+3Qew7^+lT7MAYR5)YPSVS&oh#=c?vNho!Oc zlE)?NP7)j}y$bD(sdU==$JBt)+DSO=3S> z*T&&LAF5*gzcK6o&Jpw;F*#NiGYFCZ+1-n_#N;^VyB!+#lJHpq7S?}W0O*foVmPy5 zia|^+en{nk#9`DzpX)E_|DtSmtL~Ppf-Me~wlu(|=}74&j1}m30=I-Oo)N)7Na!(qQTGBC7BMp~?{Q&=0LZXip|tXJJa@42z6oLugmHor614t$gwo$U@KFMWat_|U z-@vS(P+8%y?!Gr@*xufLi$*4EK?SRZ2N)m-%5*%wTMdux&u{sEnKj^iU*Jpocknt* z_2xWTct53*!$8x~BVun7d8T%ZOK}nT%$=fzO%Wx(TuWl>E^E%+o}ToqbW%lQ*U?22 z$)(2GI_V=5xpyx7aeSw-r$b66$F?=yYXob3__y+I|8q~gv34r240%n%m^=Qf#M5Z< zXn5J*KURu|o7Pb-&?L}qie4iso}HF!QK5^L>Xk`CA4=0iJ#%oXJk)tn^vItl$BrQ@ zA+wEfYxHDtj*!6EXH&XC)?A?Mr(18=r>9=&?93!|g8vK~HFztEdmKOeo)$#=g!m-B zO6oTa7-C``BrUVE-q16-YM7WM>&$pML`E}4B;>kVr(7Hv#FI|`Q#pf)xw(HfZAA7j zX^{|)KcettXzMipm}_rfVSY;asC)3J)!SP5-1+e#*za4hUqH!nCslP*v|9F3(~Y$D zxNQxMYqLZuhndq#^9vu6sGo1VSn+T<-1j2#$nkiq`c~+3+hj~Dk4J7`$WWjth5z2} zXxdYDoZqM-lbGLJxwz^{e@$xFysryBckN(v(OA)>do0fPF|v*THivq@PaPrbqoiI{ z{@J6B)PHHWzY5{p>@q9SzlHnflj*ks1~VxO4o%WJci!o}*`8Ly@z-W?66qhSJ{lHB zl^sWo_Qq!7Qkm>qnw93$|0E4*Y_cfk(`;99BBvi62FijyjN)d#g5k8=Yq-HWEpqLwH{b zyM`!Gno0dQRPooH`%w-}LGW8qUdm9t5qF^7knq0zWzW#OWUNl8?c!E+} znHpP%!Kul9#CY*(b>7+P=b@nZ1SWnBd%MAgVKQ$b`tqy|yFFr|O6Qf{uP0x_#~5K5 zjD|4&J7kpc$8T_0a+*sGFUGTETXOGi1syz+!^+vq;mGXET2^wf?9?U{o1M9+|I-Xr zEWAuGdLONt*^jYaj-YKz3uWkesFkCG`s73L-va9Qq{9MMtUI^IEi-uq&Tq)=p$!8H zeSAQL3RTKxC%&VvkD_%ADPl<+!=ZY0E8vMgAF1~!zGRCuPmp7x`mTA{d@}b?Y}CA= zd1=+J2bidO?ZN@R{izXUvZ|wZF@&p%dqST-XW<;aS}S2SA8je6_G;>$eNAl`(vdos zGL(tm{)&|&s%!GnMeL{c^^l3&+_SHyaJ!QX&iIedqEE+%+s2C*4%M5}kCHfjEba)c z1dQg2PcRaqnj$h)vL}C>GYl-4CO-3_abcFzQ<@o^DHQr#Zf-ZMWI|6&fJa(1v2~HE zE_{PgKd2^RB^eU=t&eD2g^P@{3Ty;_i8IR}ixsyOkEglf!q0)&fiO+wpI%E*9}F-4 zygE|pU^AL+C-j-h?i1hKb|g`4_uoSHcZSBS*lXmGfiVxf-emUdjw~JW92=MvK4N{u z%6<=4!#je+nb)bhXCtL6HbrGf*snY}t|8vC=Yn$2U3n=Wgg81~9Pc0ysK}phSYqn+CaOquSoIyF_F7ozx6+ulax>WvBUnKs9hWvtH*iqQ zFTMGBh+mC9LWS1Z?<_m&OngbY^v!$;Wwg&$!7S+DWPV&`9Hxeo&uMCAB=c5d&-fTJ z-HV8{YEGshF4$Xbrd1I);W-@<|GCL5pvdcTMP+4TrfQ;mTIK#&Yoy?~8{Gcxa;BHh za7Ec73A4=pPLG(V+C0LXkkB(NZH!J@Rnt^vCGXPjKMHyBgJvsbZxIi#*C^rkcJc`O zd)tjJ74Mikaqn7(KQYnzgEsVxN14Y`KR0rAKk5oh5yBBPtd4WDcjk7w=U}(FUhs4H zr-utR9#!9`oK>xgYP%#3o)A}2oOij!TD^8*B_SpGBQIAsTG8xkNR!S}-smmLq%xwq`Yv$Jprrgft~8g8ENCp8Zkjx9OEe_< zq@ii{1uyxRC7!-rHMU z6%}NEDCHQglfM6{gfj3{(#tK#eT#z5b_qF-rrY)MLlWvo@R9ATtW+D4c}-f5OcqtL zcgL31N!ZIx?-)H=4qzx9M8ve}6jjGYxm2#tjvup4cChlhyes;Iok{mRBkrqr-*uF= zfR)1S^FsAOOgZF9wA*oTc5n0vdOt#|9VP)KMeMs|7J896jQekX{#QrtDD{yR4L>Zw zr2-)Ajs5w$u%)@zF4G0pt?&CN1?sf^6SEq1?ac)C6$GT$=&!20FAF30N8nf1O}(J4 z-9v!R@#fN7apdG7iuNbZ%Od;??iF9SN+m(g`JRS`&8oF8|swJ`0E6I5>&%mC&X-Eet? zjS}XVH+EPbXFLn6W&54tzohuHwKbc}nA&wJ{pmk%<a#T9xOb`#=ccKSb59zHfF{zVibw}peR}6x zQ9l_TLgGdFIASuONif-eaMgh1Ib-!sPm4TS0BN^iY;=6++(Gt#)>TC*#_IMtpnSi_ z#<#a(wYSxEwx@n!utKjVJ+?zoRym_?KA~cLMAfo!{gJ*+o#&4gC~O?^*~nGw(QK~u z(QB{qGipNivPYGNRP4>77OB*#>|J{Hu1Sk3`uq&nTc-!%Z|#4U=EOBDA;EpRA#`HmW1&Oy zKtA{Od>StgskKcjW03?MC-0uZ6fKokNL)ADg^tJ!Rj4e6XYE~kP2Zf;z=W@xJ_%wJ0DmShM009oc8IPbzJR(~L>>7Z~)T~(Hd10H>;q?t_0#sf^<_X{>ci68t8|(p3r9i)c zkuA9VT;1Gw?UrU1RX21ZiQg~vWWYEGv(?HB@;9_kAV&OapN@`>oLyYB%B-OYbHb_D zYve<(T2Z~aw0|``>(W_Ids^~vrLu#j+5KolB;0mu8)HiSZ_d4uf%8G&X{)%9KINR{w8=GMZr{aDJJ8ywIRy6ZwrRJeJZL5q$?epA*m3Wzl1mZ8J4^ zYc2sbW8bi9&Rec3*WLEZOesbcfSi3)@#CAeN)2k4seq3#-M1Jj0@2%@l(%mr`ZM9Z zPo4Be?9tF4GN=~lbtS(Bzau^&A=essEMOk{4s4F#eu4?iEavq1`0((28VUo9+?o$S zjJ4@=k`X~SjkvH&;!)0bRL4hujXr0ps!LkPw(C+RtttSjUS+atUtq&}=bQm`PIwDq zV`w|Zyooq|_8gS9HMZNgr><5ERG2QJ^o2rjnfA`2iC$*QRsY@bcTlfV*H-9OlTz=s zQ-lR?xXBwfddpqbbfs*w^_!M}{Bcw~ho8H#(v1bugM`#GZq5l6k3pAaA&zZp#dw-; zEgvWjkhs^clY<*@9C>EOkY8A+p0866#uoEucN@NuAA=+sNzq&Bv<=!_A1Ge;A918z zbeZ5Fy+KMB&BZ+54NbJof(Ntw;nB%<^Sn0^(;id6A)f)-TF6|;QM439qvoV2LpI5w zfVOY`ExrYLr(4`cIh5cqU^l3d!l(FJ^ZIj&)e`*rsC324r-KqWVT{WZ|rrO zGb=wgAb_=Vx-o!62pp*UpwPkm1Ex1Vgp@L`U*8Y{iD4Gx+F5cd zD!OML$Wyy~XVk%-I?FpD?u2_Fr}(e-{>8Ar~{z5S*Nx%%aHh#S)A0+!dF;0-V zFYAWTXF|c-k3?~pZO$6bPh4R6`nXI`4pbx0f*y%^5rJaM>$Eikmfla-?rnjgL95d7 zZ(}13%#j!w+ge+f!Dq=~unyc%uo_%aki7U*Fw0PW`%C%iRnJI1)rH;KdXbSTdV{u~ z3m?v}KQSqE@vH`XN?t_v_V(bDM3tNwY74;#AWasTN6`TBg2-0msP}*gKv(BcU?7pv z^V1}wQ)jCwuZ`_n*aSE(E}Ly!4<9Ed>Ym^52Ok{a1HQY@z3MC8k^@ zUq~wIA;JBpB}D)3mrCs7*$bGk13elYwT*xgx{mO#<+JhCxR@A|`oM>e9>Lm6x;nmk z<+ZCrTU+V$kX5e8oDt!!ibG}4aiyu29g2xi~rX zf<4=pDno%T_a>xgq0U_q=Xvs=I7U(bo0=6E`R7)N@5eMo(2QtIIp4Vo2CI7q_v8`? z?ZHpdWB8oT4U@?t)4{>P2=IRy**jC!!Dyx$48Qdr;N#%4Ux8Y7urX=8`)J=8vm0@+ zt2{aNP~hxDYj$rBrs7^iQ{A#6uCCr*SI_U5FiebTlHqGaM89Ez+Y_&*mKNt#RKJF% zCWydm>+5G=mB6DKbDZ{{k9U#5VSzaX?yV>H1zev$c`^q>y)I!Yue7xJmBB*#J=##g zDuh%#!?mlMV=fP$ikFob%CcyGX!>)T%~0u;<@u*=SiZYif&2>R3V3*&$10tis}PyJ zPo6xHn1zWyqUilwC|2O90{pG825&pPGTk?DXQJ&aK9t~^^nJFOfA!*n<9yslde_~s zfoiQR)kh5hB&)*_7XNegAJ`a5W7N6-3#RMKgN13CnVD+Y1UNX3`=bs3;_FjWa+ylm zzkLW}m{`I7U7_1ZU_^T$=9_QgHYxNLW&F!-sh`B6ih9SHMc?-+#`@n8q7TT)l~c*go=;1x7+b*2~0Ws{QTgmY9N!J`&MBtiHBGG@HJP7!_s5Xug{dt z=A<^aJEn50>l*xZW}3F}&Ltr%3ZBOs-gK6tJj%-b4r5O9akZbm3bH0I&h{E__9_dd zuqt=ANxL3pIG);Vj-UB@T*6%U+~_NuQ$kR;Ei#jN?ZemP!iaDe+2%&sruimA#_;v)`SxW0IXN0lnbc?y~a9ZziV=Rdw0R&Mp>>dOYGD2@WO-IMRmIEZi`TwFfi`5s!U z)o3E2SOG?v1g5=7GA`fsY7)htD*Q~4igN_GK(=P*ZkgF|Z^nzs@Uy3yQv)7=jzodD|E!Gvp*+#v_72h#C;IB%#aUkOvBbKr_ z3$rL*P)8)Sm4#VJ;sj+hgJ%6FLlM=7RL3YIf^%DCnG&|EkzBILDPbEM4gSqA|T@AUcxPJXz_rT<`CpCR;u5nKi8^)5+-M7dg}_m5Itn2 zXMN0A*Y>NeqdwQZcSv#d4D}(e@4J7~*QQRr%L>;wZ1(6kR_52@yBFBwW#Uu0NO7ou z22o{D34Q*L-!LNA)I2%q=;;`l+j_@P2EEa#!ekE%A5Irc=cH*xP>i*d+OcjX6D4L_ zsEeVjJ74OU>j=qw42hEmf!(jlkNN_{ImFQ2YoCBqIq^EFQ#dudcI}1e;x;B0B$^wJ z2;4*_i?vZ)Q7T+(mD4lH!}}+wf3LlpmP>}PC?!gC$Q}biS2i?K9gLCqBh8=`Z9L{S z>SiQt*cp{E_uW3S(^g1BCRU*vb2g8EOIGS@lkb~9LT8g@gpfvyikbZdGOZYwqiSfN z$sD)`x%rjn%VdY#23d z7_->LbkT$}x#?tt#D!2Zn(Ra?e?vb!XZsN3TIOf0drbq-vQc4wVyG_G5ZHACwPU}% za({K#!tadF_ub^IcL%xS+KZJ7PCB%4Zi~Gsr+Yh+P11o9v8%B_finjLBQu-p$LgWY zfL{k^LQ?ip>!Vfkh*j;Pu*P(<@&|0xBIbYfiK-5qKR> z63Cq_XO=p9m5^xbj;jLnX0fjYk_PnaBT`hdf6*=cJO4`8@N2-+$jM0PbD9bN>f{0y zF+mQSio_=_9}9|cy-HO3u4!kX$?rZqDyC!&N9*~JTQSlrWSu3^HmI%=R|ZVv%e$_z zZmFCJLw~Dv(M(9Oo^KYpNYE8zXyVq&t4xd$#x}Wg7FVi?SBUQ?+^Sv~DoENC^Ct4k zVQ~I2iLcMzcjQ>RSx@EXO|(aM{KL*iKyjK@%W&M(%zmIMFTvvUeh_g`V9dr?)V})y z#*=93RB63>eChPjrdY#$ibq}tmEg@&ohrJzy6`*6!f@n{(B-nL{yBo>8Oq&!nh0^@ zCPFTxc!#WDkfTn|kGIGj%x@WeLq6kO|6ak7lb1tlGvkXHq_P|v?=2Zuht?KGL7q&| zIH;>V6O(9wd@q;W1!1Xfqz?WIB6BvT;C2az%9Vc_mG+gga*aiT$=Y;;_5fz9urVD>N=_c(qR-d7p55l;V@Td|C6UPO&F<%SW)SOFP zYuCTfoFptv4F9P6QuOk!u|h_EW(SjC1hzwQ$&R@Z(I<->w?|&$ZIW&U4Iw5KxnXP3 zIm=2VNm`ADdwez=t4kn|ce?yo{fR1UxutNRHE@>fpBsN*EFMSIu!TT5S9NZ9k5#yc z>5Ih4@R$)@4SN%1VJy8sO6x-T)}pR*Z~N;P2Pn%3vy-LL>FM?-2~|*djyu==k)?X? zS;I9nAD)Z->m|-eox(42PU~t_G@LQdS9gkFE=H``r=;9b72k6l7q~5Z(V{-0ekuFc z&41!i0Mqe?<$e^KBB}qIOtx`1uCx3`I;S!?X5fNPLb^gf9WXGkqOwS^T04-r`bbQa zl+xfm>t5puob|KG*#n8lfA*VkBaC|v8Be%iknx#u9aS)d_kyzIu+-bZkD5v{S-)!c z>S=wb9}GNmO3^~ua$NhyV}Em}YrcD!xenbfnUx)K)%i0<-O=kNBQXIZneCyDkPfF* zrHq8k;}H;~g+4d^X=0U_cvSc>ZYB0a<7DPb3ZQ*9%y*W}vs-)g#;TD0{e%5dypSha zh(kG?zXOGvCj{HXAtj17ImJ6)QurNq|NNb6x^-~AP5#+;)eO^FTq2&QlcSdodJku? zLrB8O9k3mYrV$u^-5)4Kc)|PY;ptz;-2ii*QW8*6wfhjf9(Ecnzjtd+LfzcbvI*J? z&wzkl#7Eyke(Q>&oZ-OlRKB#G=Tp3!2jh>B$dcX)rIo7tP|Sd5FsC7)0LkRnt&Bbws(y=Ii2nb00TjB0d+2t_+QuB$r3RVWSKjYG7~*q_&+k-YmQ*de zD*5l^K3};t_O;H{k&Ko460=x#|C2Flsc`?;fo<6 z2b~Sxwbs)+K6ftVDW${nckt|Y$hB}5e`ix1fhwvjin+%{SVu8Hn%`S1%rsKN6?puj{{?LV*Dg?q@mu;aHHJn7$=7w2UEfJ;6ojgqyvf6a zg~Y_TfDdV4U;yL+4x_GDDO2V^+CQ{Nq_FN;59>MUPpvKFL{7DSja-Rjdu<3EI9-J- zv#aq@%;0m2Lx^o9*$MxZ-aQE0qD>>V{J}Z~>Av_#e*9d9ukF5lWIKgt@uRQ~Mjj5^ ze+Y@v(={}$sb*_`3oJO1_38EIqS?XenKz;!Ku~vDk_mX zc}Cr-U?c*I5W?v9Wr$gdhaiWG)0Ve?KW5vMW#cFjGds%~HQf8jD(k69uHn;5-dAmZ zHcuenKsy`U_g#~$*xudeP~UBg6~sWn*_qq*Cd<<+^5fZFJrh&ihO%nmXr)*l(8W4MlrKqdkQg zR$0ec_=OCnEv1#?9`4vMFjKJ>w~JVkTX$Aa3XL+Rb)vn~@aE$DdVB=N8D|fVbFBP- za8!s|Y65PaUVq&hj=saulwhJn@B)G>OaR=`(29bSt@>cfPro%hj8ksln|#_Uz2xWa zvFAVU;yV_AqQrL1amB&e-(y0Qgr+uX(J<}Iaa-&kmZnHeK!<{I&L$FC3ghT#{VO@t zdCxY5JEQkCSGc+k^jx(0_wPQp6p0{CiMsMEZ^<*h3E&$4mLp06d?mIpkm8MU*tzQF zv2LMbp14V|Tabs@b2bvjH0MvX>g@Ju|M*zV@JCWw8qU;Y5GBHC!|Kz$;e$=WJCXx% z#GRQ_nh2{FMEjb3#R9>I`G8djh%+aMV-cV2WfBDiA zuXQY?s;dDV>6Rk3N9x2LY!Ehl8%?|D9(>r+cV~usjtlB^;cAlqtdgs0))*;+LhxnGdi7a*OUq56b2S;f=jP_-LO7nk z^57sUl~?{N92WhS+f;RM_7kk22B^Tdi@oVoUPPN3dGI(-ik!1qhylOTqoMTUw@B<6-H^3pixVR|v8N?A%eD$LDRAzAYByEPJwy*ONbY*WMlW@gS7Jn_5 z^_onM)=Ei-y}5qv{!=pxT;|e8{nqP6j*X3jVmMYji#zhX32IKd`Wo$mg8dLsU}b|UYnamhU zD_>x=K8WTXf*kOhkbtW_G(Ipy+qsl)7RKyQ1HFhq-6kQZ&Rgr_Sv0tufr2I@EiU?r zmu0aH#jtkPW;y|y+_`Us*6VYN?@~p{Iv6VCHlUaSNkzLH85py+T|+;O%5!(db@iJCk9QF1qPbB}TVY}h;foRSBylg_4e9&t zw*JRuo4L(m=C8P6=V+yayc(BGwjA?vQI^MX^&B#K&|uCx6d#xrC=j0eg!VR;nd^u% z`vx=GezS?O@4#rG#SfcM#H%3bI}B*wy6OKi#9FZa`8zaGVvtypKR;*rKtsOWy#kl8vBY2ukUG@?`vY+-_1n~l+ojrm7WfjGA8HYWVS-J(0#NAcq2m=z*&jJSjwuHBZI9EWT?*&# zZ3W{8?}|9f)0yk{FZm-ZPPTs`m0H{#8c(y=wMGo{J=hkS!ui9JQ-yGL!M{6+6 znsb_;-8~6#{-!ppI+#sXB;({G*ce#2dQ_aNTm4PS{|`g^%5ty6vatI2%5G|QSoUX) z=CZB+ZKR^;aYd=aTOmiK*M0r@S{C1iO?~Uu_ubBRDRSwTGYxMWzDf}#@1~URa>4E= zSf*4tvyAbf@9i8*Ii1^R=QAr2Zq7w4zDx+xXZ(QTlF`k18Oa;D7Pzg#2`Q#g%j0l(_@( z@aDu_qM6$Y3CFT)?5g(~D#IPJUwZp~{+t#!AB>iq+mEh3o5ScH=^hEW8N5#QLrIdQ zabEa#fr^RhwCX5k%1vaysJyZN7grs1vW~8DK#meAJxPqQJ#R^u3N58%ij|SKPqvSB zUu%12SB4=E-?nAV@9_pV7wUNl4fBz(So#L*=?4ifEk?J)b3HAPxfuOfeE|XiV&=uk zzrRNtM*Dx2=;*pc>bi8 z7CZEOG4@rRs7ayIz+4BjyF1mg(3kvVl$%NS=X_CiX74I?0lkyGJRCr_E(5RcBXM+~+-%{n?BIjYTmz*sMi}P0zVEe0S?v~VFHQFvmY!L5_ct*k^_1V@I3hkyF?siC34 z*~#hrECU?+aDia~`d|wG#`%vh5&L#O!&XvUycDW;NN}*JnHl_aoT{o`*@ItZi=!3$ zFp34Qyp_Hdmf!nciLP#HYLSDX3s`xjaK@FHnVFFA|KsaD;IZzb_i<^G21Sv0QYtHD z3u&ON5=C~&3L%j_%A-(HGP0GdtjgXRQnL3ZGh6odyFPBu_xJlxujlo8dU3nmpZ7TD zI_F&HI^M{!t7X%Xo%BYthBI>OP)jI655Ul9n?@`>Q{8|iln{1P=V`50Uf6|ZAhv-# zRIiARrv+$R@?b*=>@%3xuA;F8UqMsF} z*#ASkc>$@<>*9h_rRivDAHnwdE~JLhdw z*TJ<4#7S=c4f#OcEI$7Jkf$;mwX2ld4aPo_`MI^5=5Bp`%=W4uA}*zxyMieG=Q+rj z85kB8hLWHDKz{=oC`vVHf;MVvR~hy8$-W%SQwPqgZ7JyR}HZWtKJx0tK>0UmUrnudo5ZfInw`3qx|EHCYPWS zcRKRLHsnYBoqSajgIAAc6|euvniva8F@AyQ#sJ`|UoVS;cH$R_+MbPv{HcBlnTWa-}IK(qpyrjBskL~%hcYhBf->0%mTSHx&j14LDrOM<= zWv8shHSc!>xMyHsjo3=k_&r_dfVmjGc)Li7C*LC9R zp3a903gn`P$6>VSkDuUxDNb-+&DU~_7g*6AUJW#*31krFGEs|Aim)AA+;uza~)Nb@UZ(Pk>+GU(L?OVxK zlXKa+R#+fYiuz^km)h9cVwH1zra7oF!#*a|cZO;OPH|tA5GcNze}6-}bpPJz8`sie zWW1t^tDj?A=Q5@`ByYd&2%O%F=Qqv-|f!KZajXOAdYWW%q(tod(9X zME+Ksi8^L2o;9+oW!P$lj*TWmcG)BlayN*j-2VL+O;JKTJc`m|EQ$&W<~Z?CetGbx z@)r7IAqAw*PAmV++xv#9w2Y^(zgjlI;gW4(EKn7;$txZzDV+zKmY3dO)QR!%O(~Vn3o#;Po^D_L>`n5f`zf5~|+?XRrnZ<&&zofcGmP>Qrpsb#Y z;C?Ffi6-aGsG<9*x@%yd34Q!%lCZJw2y%HD8Y*T#GYnOTPYDU7qD60|s|LRPl6@v6 zE&b7w4WEV1Ptzxl<4%Bk=8YWDz2lYL54|}?I*J?(;(RpjY7}qrzW%(zAS~k={RiK2 z)6mtG!Vdno)Gwz48e-_8)UsJ>pO&yqF*}TZhMtbZx2mdp;0&P=^F31M9m(kDLb;h} zutRH*^TI&PF7c&tKNYQ0SMm%M9UNg|q%YI08nqW9r*6oWe#8`KjAi0R=k0XmbnmTV zE4QYeuuRZcZ^C88t{}4=X@N~sV^$^kvkUp zDm+qKr2j$W#)~=4*R(F4m1r0G``xW{yJPM`>By;BjlQgcqNNdyK4F8<6dcyA0Xwg_ ziR!7{IFWX8`pLh{m!rZ<*nU!&Q~l{`*V7mmQs-##wB{Yz{^8SV$=R4foFQkqxl@Tk zWmo$foAa&qWt@m?ey;fUbcSa|qDIMVM`MAA?QB}Eajzm%=l*ep!V4Q6x3F(m+&R$p zes<4DSR-RbR_%?2I=^In%h0Q7R~++0cLWHS$hQVvw=mY7jr`*+P-IH;^2!#cF1>!M z9P0}swVP}ZS?d0zq5hbb)|O&$r6b?3p>b+rLJfKj(AwFulktXz#?olcSU&L7>4O~Y z8A47@7K3|-8qwY5KiZOM=I0voD4gac*6exGL2f;587?_7UyDB*!nblNUcW5;+@kZY zN*RNG$2Z56KecbC-cEUXIr|lC5#F3=b3(d$sUBegE@8@6&1`E5743sXy+NRInTk z|L%~Ty*gM-&fMaYsWaq6y zt6KTmc6I}`CzSdv!4c^+tk*bj0*bY+~~SBvJ(XEsDR*- zGYI~ni-MC-BiusC2662IvTS0YPfyAnw zR1X!aTl3Jcz0J%vs~y*+s=6n;R_DgVO|!G(s#%$(oHKi-XjqZfoSU8;ZTOn<T}e~d5au>MNV(a}Z9$7lnxyx+U$zEsCqeT4hBbN42 z8^0%97;`+qc0dl64e2 zCo>-ZI_6otky+bGKg)f|Zn=R~Pj)zAbmnqtOS0K&-=H8%b#wmo>fg@W|wJTD5>G(pV;_U(Bi`qHGlT0;+4=m12e5&~+@$GB-J-n#wA|t;tv{RZ z=oR>Rr#oI7VD@9+H?0tItup6X)^#v-fF1HPjWxN+G`cT8p@WCno|*dRsh*|jCG7@@ zWa^h@UHS9}cpkt>G>a?Jh6lTIie{Zm5mJ&^Vq7}IIDSif%n2U(CZ9<`zN3C^DYD1S z%*|&{CniN27vB%;hett&&>PbXfy0@JW-$fUt1Ioxvu%yDy?)%1x*x_mkF|)Y{$<-S zpLEqDhzu$+Y%C03(a4FW%|df2QaPTV!Hb@r-st^>J^e!*UzD*meELyQ5A0_Q-?s?F zYQ%FnjeoYVUzBQFO_F85#E@bv$8|7)PdWIDvtY@4xvTZ=eI60Vr<}k9Ebhvd7pn3pxrcrN zzx00PQ1|)oWdo;*r>>aUk7bGVU0#@dr|$ek%q*}v$9-ycbhY5nYW%amS|$0{_zR!@ zIDD6B-m)_$I&R`b{fYbbrPKGv?Ux&eSSD7?##Z^dV&1FpkKY%0ZDDRUp+yakG1vP> zTXaNJp7~4UUFIE@ln$1~kx}18a-sl=GITxNUu)?EY;hLxdp&E)6%s@?*4KBGrMI^? zfLnX#_U-Xe{h&TQlZtfD%*lz5-z6W!Lw?0&D_KO91=K(@2KY@S!$b@&kB(Icq_0?O z1ECc)IL0e(`BxquD?7;ky2O@Lrrn+U75>%|9>3*>BbT{vTWRu+2R)0IoTqX4+OIjb+eo;4mt!|viF`ZGb zwkq1yuXFoe)N~`uBbLXDyJkO3$G-j}=Nr{s5|#TQ;X|{+tYSa_dz$6spJ+y#7JJ-J z?E#Z36=54~dz30xc=kJ&X`rDER(A zESWCCspU8QR&;E%?PJ0dyO%H{d1EsK}yFMFrkrgrV3kM14ic(0dl`9L~7?wOJBAfs!DD?2|s_qC^& zUfbWOD{3*;?ss@m@lVafH0>S!W&@EkBC~%qsDoY#SYem0}~_H^xAXM%#FyGHtrJb92^V+0EB`qqRkmFTANM5P;-O? zn&h{-I;+mYhdbE!$nQ}?Kc8VwNp4<>a)M#@@>u>q{SnUM3uATdY*tgv~;1 zzJ4$Hp1Snf{k4D4;;*cNrI9Zk@AKxT9}Sh9ley(-aq`1-A@$1P2S-MJu>Z9CTlw^U zV*kwSXh+QT@OG^{`;5Uv&R5E>XrQBZc^I~DC+mW8uEryVc#&LMvd~LP>uNck*~RO!PM6l4?OWS zi=pewQ~!m4u%)pP{qqJ8`E_%1yYQGQ5MoD&g*7x+7ajY1dJ+&wf@TP@I~E>;a>fI} z%`4t?W!H!5b@W*6^7d2dXxg5pMc~VuA!utRb?>peuU~+CtKdifu#O^UgLX2rsE-Vl z+217_!YG}Lu&;X0moN534@128BH(o46idWUdksNN@p zQ_xH%m|?Ym1q(2#8zinB)4kB`Ve65$NjSxx<&OuV0LE9e_qoQ@&TE~Is7ZYN-sI*{ zmXw@|q)E3PCZ^R;Pvgk6`TGIlS9XRvNI2_SSC1ZEI9wH274^M(sT!R^%S*q~w4)yc zo-XKjP+G;?Ew!7Cu@xhe*(R24{a}dk599tWn(CU!e$Nz+`l1Bher630CI6)N5vgZ| zV^g!wv9g~LG_yDB4fmUgPwsg>s_;U~H%#LNCrjdpI>)auj?%>uzV~C=lEc4h`wheJ()C39`;7P9*6tBPFDw?FjvnEUJPGT_-0z`p=)6kawy4)Wcxmmspg@ zr+i?G*@ukxPRmP01zam!%NZ7Is|c0v+^QMXchhXm{-#^xe#!1$Hd~>6bGw?%$gaC9 zyAgQi%%;|>?J^NFzd7`fMe&B3jp)dn|2MASBvW>N{}s6Yt?5icjn*IBLn=-uA343x zI3UU~;7s+*%b>m&9S2MDE$bBP|M8^^mrJ|mVEX*@MQ#yp$Cqc$2@V)fZ+*+C2df6HhIFzmqb@a@*J0x9z zEjuRYExr52l9_C;OU$%2#%!7kr!l?zGd;7S%SfeDpPywlD` zxnLsTd|N+V592%T>SH~u8ozR)IlpWDMCOM6msiGgMy(E8{d{YRf66EPiW4nnAnPHqjaYvB&szlDIgb8|I>75o*I{#|^+k@8twc-7;MBiOn z_O+ps%uGj5uNSqpfq^@EMbEY^1$^4Q>?I=f7x76%6|(G}G9EP42vWs2Exqv%OB_~f z>sWB=oy*A_8wc2ADTZgQYEp~S?zGZyW12ItvD$-_%Y!F>Eme^w^~3KAd!1)*Z}ZzQJN zhw6HDO-%#D0FiZgZ$lS!`hgGff?So>e>?nSv)rjB5A*>|+V>-t=y1MY;J)-?{OcpF z?d(!&y8B zo<+<(qS>tdGwTzIJcr{h^1U7CX=XbrB2~ofcim@Nb7{-U4?NovV9jE9= z1&QYa-q%C-j@;>=FM7c(C*oy2`9~mf=-IAvCj|}V+@;cqUITxXVNn4alT7V7Ust8I z44>>3ZZYb6Z@U-<^kOYCeFHSg@(P{ZbBh_X=lb)Tpcg~xi2v59Q1|JKxO`&(>#vD>rw+N9BzJ7d*a z6ocZ&R+jE}Rj^p}{5_0{k=IC$>EPh>WEiwoV~JK42}(*z8Y5a}<^spX83-BTPIKg> zIB#X(XeTI^TxIxF0iU=%+*8u-XFA)2E^8)e`dYu~9W8$=DBtG&(^J*P;#QFosy9R- z5Sf#t{ZR<7-k82&Zkw(0f0BN!jqR>Q%ujXZ=F_!iv1#?qwXxLl>M`lt?YM+Q2hH7O zFBPpMKd`rIi?G6}P{~~-8`}{4MbNQw)m1!o{z9g98*^w@W=r;}&{8onyy{J9Jp8BC zMZCAu&yHPOy&E*IeGE~j6rb$E2~}|$$1&wRj_0DYr4SeASI)aF zdygguwbQcaAJpF3kuho;-z-1IP-op%dq>V#rIv-2wWv7eT*0HIA?;N=NCh3R*jlTN zjtBBhfPLWzsGrbr9bD2nxwK_tHGrs zRSx&-8I#nfRd&4GaHzJvWwid$JWPTxt8l$?y`$Q6oF((^tgC_XV|ruPPjRsGJmvKh zn{3WWpZzqc$beW|@YdKSYSm`zfp0T=B;O_EavtJ7lp<79_np6z?L&0ItmVQ0ayiWf z2`$U5zAVU=A&nCr89s33a;dCb$rX!A;o5wveCeyIRdK0G{GnE9R;b4DIFegsCz@T9 zu-K9HH|sC!cc1#1#N~s_9P*C~+H&el7p4Y?{!K{dor2DoomMqfY=>0R3iPp78*kT{ z?04}m{3V-^$SS7k>G>@wsG}P>mmsnOiDfSJ8wZ=3;1C}4lP!uYG~`REiQSPq9P8!c$Xc1>D!7ZU8homcvu2XoUw=R(t7Mq1i+?ZZJliOOmj{zDDO zEmKf_vgF63kX7o|aTG#nKMys>Xy<%nprBbtg&4YVTkW@sD@ob`2L#s=O=J;aL1OGs zgtVI(Zc5S=Nl|gt_pll!-Xr37ZSKZZOhrL=-lRVF<}%t1YlDIqlFmJC?CtMIqvRRP zHh@CvK>?H3agOL2g%2je3CL!r+}D&n8<90RO?!)jsMYcDAP17yJaV0SV6>{OI*n}l z?U}+cC|^UV6q?`EFJ+#VLaNRT#Zxqe5u;E5KI@=r`q|vVdXOoeE0;Q6~yV%%2$kpGmd1ILE%Qi)TUEINKu>R0> z@5py#;X$JpnhCdT-KsLu{rmSvuog@?|?bo=XsXFf{7T(LxMy zIV=2H#M*R3tiLock}*WwS*yskQZ&_wH9g(uu7YA|jZ&-osI?LNC!}Av| z7>_l*xv&*JhZ=%>SO5OQw~N7RZqEJWHqtvsAckvd^0)-WOn|r%G4%QUDk+gbc{bK& zchC+}UC?EV;-1b6=g<4PLg22h9g>lX!rLH}j(0n-mc7MCL@>DP;OW}WpU+94Jx{M$ zI_dqhXV02a44%4sd6l7Hdp*Ei-(_w2$qfJD-_VX%Hqm3WfH5>CJ3+DYlDZvr2HQZj zYnRFcm+;e8?^9Cle>Ik@iF#HzxpS>_92>1<-Ly16i7^P)r~yNa4*J+o(-Z)aVFwje zYFe5XL~lJjvTXjottPbpFs>qfo>J=7x62m9(o~*h--wY5%g`<40aistm4WIp(Yl)R zAkioT%B9dQZ?3CLyCwPd=WWV0wKcvM$Y4|M+ZTW^PvYW5Zw#`4Fkcs&x_S$W!cc~r zLo?^hRe{|XeyyL5EsDbAX4Oa;ppsL987xIb1Jib{pF4m5rlz5JYr?1~AE34xl)kQls3tF+SM-ktm4B^KME?*{T3U+2 z{*7!(Fb{BRWCLz)Fd}zQRIHxuaOP&ov!88GGY-IP7l>5Z6lpso^=hqd%PARCQ}pc{#KLc6 z?joS5^uwo5klD%oYcuCA_Yi9`)zxYeb(wYq$!m=Sp{njqcxdwuszzRI=%371J)`_sb0 z^NWj5e;@Lgj^rZ|dmbR8@0e11?w-)L%fHX|hk<5zG= zk{h$$y_1lz>8JccI!l#g$a)n%+?be{DA|LRDP68Gf3F68;#z;JoTYjiFHCFyt2O zoSdbcy>F#&lgJzj^1UVFAa^>QJ?r8QRmr<(^pxo)wCrjI(A*0*GYopj$VeX4NU0%G ztqV2e((9*!POJqbWMp;Mw%CBh0gXoJRF^=JuC@`vkr1Q4B3xct3hikYn!S5dA&UNZ zpQrxW!1EQ`NJb<=pV`K(pkS4d;g!0LuID(kNjITa9o(6}xCYD1Ad`-P6Y8L%qH^EY z^Qbq;YAMJ&x~LeYsc~~BdvamN8TgH^Z8UqT??K{6ypIbxzbp3anu{Q-8{9b_?!1yE z6*+(uYFsbI4AdL)q0(&q=_KC;lf#4`A$wnUuoI(iBzhL6hfwUc&S&B?IB}w@=g8*G zKB0FB)CSiX7q@tD4)tNrOSf+Q#rd~8-%-oS-UcdX0^0zKp?>Ol{y?dG3X7t(vge?dU|***O4Q-xTn?$ zC;6_(xD?#`T5UxZiH8mGB_ zI4Xj*lSJN8xLCGLG=hPIDQXWDb|+2D?xU8z>_1^48$fiNk;zkTV6ygeb?0V?i8vQrCitWK#ybg~o| z@I50y7{o#25)y{7Y2Z~!N=g#Eg?sn@VgSlL>!G{v*Krv6;qeS#KR@|E9!cpj43Sxc zyfb#+SIK#GvBPhQY4>iT*8txE@`Imc=jI;ccAkHVW_fpLBmY4{moA}z;x7Of$p^L` zC12u#dhQvBJ-A~e#0+>{1Ze#9N%3+X?i-7afk6i7^29Z?^tc=Y3Vk!43^(xe&DR(} zCCP4Pv45=*j*J#79wKz_yac#w$TLLNEYFrbf_y_`;|*%U&_gsfUP}l{ypR)7lK=<{ zfbML@;N362)HWo8)9Rny@vsAu+c?a%g~g&Xjkj?ZjXLE=V=+hQ=_MQvu%+13JLW2` zjX}P>)^Y?N7h^9BPwt0ffe$cB@>!X$a86Q9iB3(Odcgh6!-M*z)S@w#jo8G9^KBbe zFmx>3aYl6`akVOcY|M3ab#*sP^NTrhe1xT|r{`Kt8l)Ra&S8!ey=*hw2}t$Vw8Nl; z@m;6%N2-^GZI&UQf6|%&K+rbjHw95Rv@90CziiIa(!RGlA2$Hj6n7p?2M5ILEYlj; zxhtN!C-aE3%0WrszSUSA=0cr1dh|Q8h|h{%zs$?a%#qcPeitJful(Qpx@T`(!i*oN z6hv(~^PTT`n;`(DDTb8ceIfC$ik^(Pwx2%--aKJy#(D*4I|no7ZGHgrr~#<(M{swKTCakbV=)7tJ1as&x)^=6TUjCw{G1+SHudi zG6tO}+_;em5->9}b8qe9SOF25L-HnnqJ)Fcc`0HYh30>l<4hBz4m^J_ug(>}=N-Sw zrLFDp+`mV6puKIKgSO{f=Tb7j`sOWLJb%C{0~2DuGBG;K1d0F9wS z7cm{$H z!$Ty?Vx9voByKC5ww8q5?5HD@eV}~*t+*Hx?jAqj#;!mb9(En8L_{-}FjNRTR;J1m zbOy9Y!=66*Qw;xk7;4}RutA7uLl-(GXzO6XbQK=`(*4v;K!erA-@of4hC@>e^nuI- zaDnq5RWUX;dp>!xpUTHyf89TNC77?Hy*bYOcbEC-wCi1Sw-ktV^GC8y|+z7p%ZI#7vw@s z4GoEJhJ2XY7R=yBhM-~IQ!F$j1oOZodtfMc#L<-PA~_yY5KlW0_x5&TKuENdv$Hd} znp0C#5N7cVm6ef^ICAXRMMp2g=j&^!j%*1B7WYg2LaT%H;TlwFDzNu#fR?(2{6JFB-}K<}^Pb@+Cb?6o5~s)(@vU-We?{rPrEN;Q-SzHXP~ z{VK9%SFyMoKJl$|6Au;MYhqSz3ZLRv)*EQ5^ z^akA4O7ryyb-JOF+J^cRgBK{p;P_p}pWjb~5kXiwbTTNW2W>R!>OU59C4~f9YGqx? z9q!y2z@hGdcF0gevdZaR1nQV5g>|a>{yqAT_gUWWB=D@@j8KbN*y0_$Af_M01+_~a z0SOZd^tcbFfxkAET)4j;A1){flQZwxGYN*Zo`C_!D&pNSNd9941|k!{Po(j(U^!qq zadKdyq$DL#I%{&^j|^03e`r8*$+b^P$X2Umxk1pz+z5j7UEs9LZY2aABaAyDhDyA8 zHHn{(Kgx)!5i#7iHW8BOWBH4wQLr}Ne7o~+S=)s<%Z zpwq3Lk_Oj)cOO<(_6cH-idxT1Fo>;Xzh#BZt*w*uV&YtZI6Sw$v90;Pk}An6{~&CM zPHgjL6jrpaKyyiN1cG!>k@dYB)=*MJ814?1hJ7m-lAMXQGxt?ho3=7L{KJCT53r1i zD$GHG>+b6nBEk;GS$QHr0nF#MozSf@9jEQ>>5)Tr3I3PCtMAp7GgU@)qT9-t2p%6` zs90R^5sb^F03O`Hcu~MiWQMtfEZJ0sT?)pugm7<<5(b$Hi+WC zK}KRn0+W0w4|xU66Jp&JaN(Q~C(8PBvT<{(tOWY`5d%2XPGb&Ms?p+$L&N~A_^tuN z%OsrSVuyswQ%_I1?Nu9hd?v*$wRjRjk{^jYf$`5J4)&?$oFr-ZY2;oY*@Q1Jvs2HK zV)oc$bWfhJrkQe`4*fF?6uNFM?;S4u>XWMWcLv}%%eH3S19vOW?p=UF5C&LXkZ5gd z>xaD6$EX#lXZue%%$ZxUzO|j4eAT0yb?-uR`hT&dJEE6eXbQjtPihJzgUEB^qE^(j|W;j>TEWlWt6O{~h~S`%O;pw4k@|r4Aic?#H5?!Mxp!>HYim?V~Mv!#e(7 zhL0>(7MKG{RBdf}Is`0DNtryzLBX}11ke?hB=n4oI zt-zSyvd^PWj^e-kTN)C6DCy2(E)4Q1$5&HfnQhX6c1k?0VWlxN`uWqR#y9;8zvt$D zYgVhSiHiNPzOsj-c?yZ%8-fBcf#3^XO-e%InrIHTA9IM%i{O6g>eVWakKr#Pk)Mu-_f5B=f&^36Qn}hGF3`uIH?|Y;j#rSGP1wOkKjn z#N_eLgYUB?`0Uqr6{mfyvom6t3e1W$+--AU4XMbFis3k>sIYU06as71H8o}ZPJ5e0 z0ySqtEeB5oBXCC=U?l7z?)DyOx;tHUym3=TGbV5us#wew6&)YX@{^`@~!ecdzp z*2=w6LAnXY=qTj3KxPN1p}o@7+`RLEz%W1=kYgGKQ@B2r;Ekzrp9rVG^(O9#JTgnc z#EdYE;o)KMx_9rMT^=UPV2}Sr@0NPmhUuREV-yq!>py?~&z43#-a@#+9TXI&?s9MC z+I{CC85xNS(0YQ@nB&qM7k+-%YibfaF=~jQo&)V3c!e~0_*Mc~gpjfJLU<{%)sZ^a(jYp4C zZ2e~L!5QG?okUQm^c+xx$8O3HILJ#%=@;?$t|}c=qgpH1k>RsZjH8|DXL&i@OSCPa zcQ<|J&(7Md>mN1-jWFmuV%rL+d1m0C{_o$1W@dasxomC!dw$mSEO!^iOru_ph8>If zIrnjfwH#0FS$P*sQlNQ>nZ&N2ZxTOTkL!)z94&-SnYM)I)2Hw+I>e7Y?p{mv)}EJl zaUT9yjClqUJ?OTosHhOMmNtB|Zj#kvDaL`o(ZTPC=`0Z1!$(`e6%5SdOg)Qk7$PN! zz1&f2a(Ev-y)6{tAR9-_y#&}`!XzWitivc-Xqai%yq$u2VJ*B)(+9<>(sk3621I9# za0MgcJ-N`hlibi?A&s+lU3424(ieDdGNU*8lw#(`jw}6HzjIvthsQC}s*f0*K17q~ zuT812bseA4;ojoDAQNiOX860c^;^}|iHReEDj`VmV15ZvN1=KaNqvis^z8qVc}x9E z${Q=Z%8seTpt<`9rCGX0N6(-k$dhYY5JO^re|-|qLCv*weY;;F0>!sf5^fw~!N5=% z7s0gJ@K>+)Y+$m|Aps7fFMI9ySW31nc^&XiBG!kDAw2tHXAC^#y)FsrssU50Fi^&T zK&;ihaqh2PWmIGMpLVV5a4~Ccpp7G4NJK;=rEK3QS#%}o&eRTXT%iIj{+vX7)|a%E>)<*dNCV(NS_++$ni7ammEllFeJbDnB^|eGvb|@O+F@&T!49bshN3-Fo0NlCES z^SyiRI(M!Ez4uZ2n~3l(MZYXNQ5M$q8cxv*63Z1{MFx!y#%o_G0p#0Seyo9>9k0)O ze=ow-I3lcC^6S4a?WmDdC&&L6YS!^(Viyu}kB`?iKz72$=Ek}u)okcKM1-CT6@}d8=PkYc$C{daLgYxBv#KcWwu1S$u$4e3DF+6I5mKHSLgQa`xvVP@laE&CraIC&N%h_ zc!{0F+X`rx8TPGb2Hq>8eg|zw);!1}EW&$^;)6!*BnC}k>M&A|pk)wQ6%5)HOjS(X zZ?bvKsa}r@JkNaKK>Vf&7Z(@Q|DufS_Pw+p&FaTUCgLx;L!_}E0p$+Py1FWpT$QJFZ zmoJy&l06u0jjJ*|GvBi(QMt^uQfJ-3vJ>r&z*2k+W$>d}D3TGmAV{{LlZtAc6N!TQ zU7Tsm=Ok#MAs7*dw2^W)l57`{J7l;a7Q`Vg{(nmnbw?a6B&Npx^kMB?wLK@y!C{Qe zPP5d?w!Bj}bLhh3;kF!=qM9n2lO&|)hYyboX|&s)ha%@xi=UG&;!#qO1F*lPa4W~; z$U~z*y0>e|_~Te2%$mZKOM4N=Y@}t z;`iz71Xm*ZP@Yk?O45FKDWvpr8QR)u`=v%Gb9X4&BQ)Ie8mFJSHRS6vqd7 zxBj6#2c5RApv>-!Q;D$C96f(gMD2kXN(urq9M++`b*Y@64?)oA7W5pr2lnlKPIYZE*fI@(#|7sz)93*GBNIaJ~Z3yTh>>jH$;J-{5`CqTsJW5k`$$EKhgeP%F)6&xN7K72REv1JEGn;u#cry%FjEZEM!(7ZSiSpbiJ-r zSUpJFDfz9U!mRZ5sgirj%Jukr7~_Pw+D^rkDev6f-J#AcK3U3(s8A!zq8GV%B96iA z^VVJ-q`!t%_ee_u8_Ubd)fxpaF+E&Qvm7_SuL%k zzJ3O<0&$aOx=COaE{!JF(@(Ej|2H<+fG`;l2pDS_FD%oNj|7Q9zx4Gzd=s&@pRaE& z#`A#~;^FO`f!}uj&!drD-PdWXb8kEJ3p|V8Ad)cFzI^tKigp*dgQKK`Y3m2ewOZuOJHW*REdcSXcD_T_ zOc(}60^SfnH>v9!*?J8KSX_SN7$L_6cYGoA_Vn4b|EWO1vFs8ZcY*S;duh1%DvE^s zR>Zb|2y%iYDJf~=)}7ZhS=4w)cKqy{|GFeD5(W`b(I-qj#hgpegSgJ^ICdCzjZ@GcGK^D+7B`^5<1kd zuO7dxW25V2hUyiyR#uyGCdq;aSG~KFMV>Z!oq8Pjkz4g~0`<@8^X$KlyJs1_< z!tR-3;~uOm8FFo{D~}%aj3&+5L`9z_p)GSpa#2(36DKHl`Zpz{7#t+U7D78wf+k}k zNoJ6Xj*^k!oq5VUZ8B!V!zdyN+=0=YAcF`7$_X%&mp>+%md%vo2!-1)e;sHZW7Q4s z%p<}G$7pWUoz(LaFbg`P!%00(G3t>&MZbQN`z6P8BITMHSet0))%mvjg)Np(- zrFa2zfJJeHd|)MGRU zVRNEZUbc3D6EdK%g-_q1oL*F1+>)d(bTEum8a}+i3Ah3r49rRoIC@LwziQi1|`R%Z7Fpm^6LoCB$9K=@xK_(&de@>mcT7B4C)TfQG;A(5c zms}~Kyu!1#cI=1FsU!)lZ_;j8**~mVnVEYzlsR*V$@TuOkgL}fw6n3v%*vAX{QXyN zuglMOicv?7{w8pi(5b^4h&cp4ka@uKryLNx+k6CTF7w8}x7_C);1qIMm)P|@v|X+a z6CKTDg+J*j#(_i98Nt9Rv3L;@l8a*Lvg6e}qJY+Jar5Vr%PrBhgx%Up;q23N=Me2A zsb`#k`aAf8s1ClE!bnG5YhpqHd@8H0jZH0Jn$+g6(j<%~BrJHiKQqf~|EVtIEWEwE z9FfBK)B++7QVn;7P3zv*p%xGF?~y!&UaYeIsnS{IYKIQd$gaUhJQnY`kKNRtQznfc zjMc%8BN80H{H=VDE8-kL#5~wAy@~4P76K2uyyE5A$o*57$BS8f<>f^tba$1(dT}Tx zGy>>h*t7AXFSqu}&tu^s;QJs1=aO+#I=q$_F}K*VF}H}oK}e!@j7Q?IBMX=N&%Uim zdaCD%=1V}?RH_qiD)tN(Av&BfJ6Dn1$iLs8@is^Qb%3j*?|$L{{1b9jxZQC@08q*aX( zh2X^bb89=HZf-1I^5e&Abnoh7Vu}~MJY1yy2dwRatE?5y?L|w-Ke;M!P z$A`o4D}a072L%^{HJ~oUi`xUhfMm1}PH|6n_x%)8_;zqkUhV~F9xSY{tmJgL(nMt( zAC0CH~L{-qm&j4d6AHh_1F1 zB$nGZ|LIg)S(*4-6(yT(Q&km(&xIkm_^U*^6=g$d)Cpie#>%qcA+1rD_P>T9*Tbc| z_HH_`1|ZDUwsdS1hxVT|ilp*Ee1Qnv0cw(#`GkdS@mz@c!2X znSQ^FztKk24ZWE{gMttz5pqIyc6Kt&A8uU$c12KBrk@hkFUy@*8+mb8aNYKPQV|0O zw7gIjdRI4)#Y1Glr5zm|EhBUJ@^cU9M$MvDPYmBc4KPjmjSI;@g}nYuZ}SwKI6z=C zByk8XIobuWwx#;_F$o831Y)cKEak6M(VHBiYlPpzJG93OWi70%M&Jx^WK0xyybzIf zF(4g~Jv{mUG97o|d_%f3+r~3r-5?iTzs`pa%l-j|tq2!)mc=keRi5L=e-z#*Op^)l ze^(Q8P?Pu_vc0DdZ&<-yeCF<+8KndzA&}YhZgO_2<>yaKDp`D@*>gSNb!0qY_FT?? zqp@qfWka=gEbjC0Y+S_!&dF+<0s{L642<$5ZkV(kB#DSIu~>rx$<&!rm)tQ&)}Z`2 zXX*Oh9^#?t3Mw-#}C>cM3G8`+~X}4}Ug`12i92FEQ!E(f@ z<`{s#(@xR%3f=N3~T5NSkx;TyR&A2qi=n%qY5@-@j5ISe1@L^XVHY>azA zSzBalD;4h=x(<2exH0#q?!?TD|22$X$=SK7{~YO*hTu0=A>{P>^?qqoH`9jBnlr7R z@}JL%YoF-6y5fJET-1bBq%gSbIf-da`SNAMkGoBu^iZ0}niU2FL@Ez=fh7*K6fj#0 zUJklAkI(oQ6!`e~zKG~5`p;I;;kLrY0OZe)i6ajRZLWnH5sPF~SXOb9R zq;kO0zid<{MkG_F87c$l($4^|&-L-bUUi}a$#MBmi(Vk~JyW;j3UfQ6Xq`f7y z-M_N?`8TPxSS`e>ZKDIM_*hnd7~wzLH-3)$;JpUw4Q_XkfrL~5XGIjU0csx{+_ny% z3Y&VCsg}w>@HYkj5QtvLdbG7{Aq0b7K;rSS@nlM;rD6Hc?48=*i4Ckq2?WWmK4U7u z3=n}19N3n5NS|bA4*)hi*>ON$dJNtOSjP zsV*rR#>yQH6^l>cTnE{1UplFBdn3TlY4SpG%YO>YErSeT9{Dg))oRR({^YLuA&y0I z@BY|lR7Ky{LGAIqQsfRMCJ9GSto=2W*=GxgQ4U#FKCZ zE#MNI1_75aO6Su7t_>?-7dOmIO;2;7=>3kUh5d7#e03e2hj_ju(?)pw>ob}}LfUR1 ztB1%IglM8;KTC4HI=o~2&pX+qRV0Is==Zt|NAtre*p2|)S z*0B=Wc}-$qLD;P<*-HDdU;lLOV*R=+d^ZlnZhCsv@$(Od)~^Ds;8z3x6Pk9QN~x-|3*55{R)+2;fd$nDAAwM zhA16C0;obWQfh}ZaAlCVvlAfl(aRx*@ef^0);-Es{S9Z-kxjy(&4M+>k^$7;V1S=D zntVuxXcTVOoXIpjPs7M)5FUlF5uiGbn+i7`2wvl-m@}nQ5OJLr-UV?T(1jI(s^JG{ zQ^9^cz`qfW2+(iKR5t&hp=q^}%fr*7-yOU*LCf5pP2rEUjEoxWJjz*l@VvFPwZ`H` zpmgISpSB)lWMtI%K>0A(!s_}@A5V{}rWAIi?^SCJDK290EvTWR`nw2sN#v7` z2Z_T(Q0{0snq$xZBY5Z#eld`j^u*S}@ONGpqtoGwNrtF);>MI!H1AT?GGPL+Ne- zhmsZi*(9uo+jX`hq>~kM&Sfhn1b(2vxNm419B-j<8>UZ}~L^OoLRD zCLU>|eK_Bp|0_;<=F4A-sy4SuT@Detils5Gu}{Yn^WSZ@L@e#5O>1FbyM>f+=uEok zQgO<#GUD4e328(Tixti*_ffV&@rx675-8ONB)J1jOkp7*!+&`G3P#ENER(-?FWat; zgsfb|s5=vhVC`4uP;Fd_^QvWAz#}p;5*J_v`5l!wMb;xnjv#P>i$WE~*AkZNoa+|h zOogS;r%z@4eEyF#MdO|g?M6tQ`n#eNEfWU3!Knv~RwS?0`+kEKgv7qU*4ih*-KW-; zyWB;7H~}kggzd1Cjvcs(yUHf9wbS ze~$mGO9mwsmG9m?dtS+gBWDfgAu1p+<0aT`SQG6P9v-eK%K(k~j~OM-=#&L|;J})e z$!3eGk~<_7tbw4+Zmb`>a>DTe6wwfIBq7?`u{gK5ND%ShcTluvfcYzCb~b77L3t0~ z65Du;cokCArl#c+zw}6u)IcHPSwK(_aWAAUUhHq;g*^nY-c0|i9Wlm9rzIQdgCY+= zCc6NSWZIaBG7122Bl05!tIJ~uU#3vcF#q*!2hcnTu#TF3-?#+7)NVJF7(Z{n_&`G9 z8{!r!@9Wz@{%r)5hQC71@;$OC1PQ(-S|0vbQ)4ClYQaxkqEs*lX77s^8t3?_U{tznB0qO)6-lNw3ur?^X5`9B3?X5UyU}bk>E}ySguHE z#=RK&mL_*}^Hz@2YNt-UzfrYr&>Z(ujHB-@ol!1C=&z~FSQ7(5J-^p2nggS330H*7 z>(8G*(dZ_IND{>D?QcYvlnjrrWd#d0%P+ZPM^+-rCa|k~McL2+rsF5`RdM1&cLzL8#%h~CFLw}ue z@UaDm5w#r{%+tW0Kxwdi?8lGCVu=u4WH&`mfoMO>TOq3SLGfX*&RUFFs9ABzMH@jb z4qO@+vI_Xh3c+MQu8kWv% z*ykCRo=>-1Q{KFI^p#9NeO084obUf4?l0r2T%)&9bb^8ygaV?Z(v6fzqoTA(H%Lo& zry_`ibayvMcPIh^0@B^m-3@z8*Z=+P_ru=jJ?GPzU)EZzIiGpY=eh3@*L96CDl7lz zx$?qum7)#JZvaLS z7vJ69#wf1qy^Z`Ih>O+WC{KO84~z8@zwFlEnsrYWFG%}FRjr3 z8Tk28G4t=ec+uYxNs*EFNl6>rFPuB0Ak_j3-C3RO@87N$S_YdMG#|nL;6ViKBQF$p zFs$(WBjjNZfVqb{!u{1jhO>Uq3xM4PCPlapu-08*`Nz{0A4%sau9WutuOIgz1d#;B z1E?4A+_xz&%#-oX^~oGVLEVrNXo*~8`X)&SzekyV{@|HsWk1MP!OR8 z*a;Zt`$R+xNX7=hAFj1U(3*grUJN`oR3ve!lLt#%)TNYhh!H*!ii!E3NYAtbvJx0- zxsVJ(x#}a+sk-U~Bn=55&aOo>I+Gf;pkM@3UcC#aM?>1}f8|thkjwu+qC@fA0PH+U zVq@ogA54i!l$j$(W%Lt+ikhc_7I{OzQ9&q5ODh^WYAOJQYn%LOX=&_iY?M&=LHGdU zSJr50xkn?tLSNH|)d>|Q)#Y=iK%%AP668ml?EhbZq(_4H?r}JmQP!Uo^6^vAvB?$| zI&bk`?0nXhK*kh^IM)9&sx&f@E#v3k-;B+`H!&t#S(xadxw7IskNH{gUky)%?l%86 z2)jeVV`ImF4g;Vm5e)X2G|UA75Ml@dWpHU?P~^iIBZ*dof1KrYgVXS-1P# zP*h`M!-Zm=5^pveaz)?hyigiogMiHrg)1TvC^+BJu^eK3fR#bfHl!v%6zMn}o&666 zMWlE~-Q>@sjI2CqvcbZO%EM49PMuGbn56mpwx7fQTcaM3C;!jb6l+{pXh#&wRSSU@ z2ttRcy-}~qT&sr)6@)y=$<4P{Op0<;IZ1fZu?1+?e;^mwv>0&1Q4FvtD7l@6X~w`h zK(hHtP*4!aTv)2$$=@&tQ20RV32>d602tW=OOuohi%p0vx=ee$ zveyF_X}DCC7BpL$`*$M*VB0}>%QELv>pqRtV!-mj<&1-!|MTbVH80dr^gC}ABhj(s z2XjwW$#9ZcbzuQMcY4F5JNo;(5B@Ea1kVH?lx2T+yBM93f^8-vp7UZ}cq z1GEB!H%LmTp`qdFnZOLk>mbFop^gqVN9VNHPYeD%fLs3GsrDf*I|nKxSE&ezI|Apj zUL!3M|kzay{4+Fal&;s|9xULhv z1pp*R+n@p~6rvT(JO6Z)o;F|&feJ9>9)bSsoBVKM%x|6yH=H2{J%#4V=ji-zmxA8x zCJqh`Vo1q&U|_vUB`{_Xii=s>*qA`Tny~|HGUzrE(v1gmALbBz^X`nYJ1I%?M=r%v zO`Edm20Hp7$XEM7r3Pr~A;@7gD@mc@H?nT}{ZRfW=nAMXiJv`zeE`Mbux(sGi~@AR zHB{8gt*gr(>S@F~p^dRzO${1p9%S=&1mYK$S5XKNgolvlB4sjym&fzS`qZn-3kHQ; zfZbL80X%M?!AMW1l;~+Abph7UDioPv128Eu&%0Hp>vWIQ(l=M3+#rzDMN&qFUgpPt z0hW-!gP0Fde-{=Y>d1v%hR>ub-v{grl6VIH33)A`U^Kk-Q0;30K@NawqcknYCX(C^q>wc+1!b8i{)r*Eaqe7ZU?&& zk@v!q!s*5_D1|^!D>^0oG-vo{9wmX<#vFE>L7OMnmuUM6Z#Q3|hh#d|YPwi@!HA26w=KQ+U{&v4na z`ftJ8f%Sp z+8SW>-*$?R`~M|o#Okdo7Px|O0EC?4drQ3kDRtifi8l>q3y@tf!qg|n#}oN+K|wvx z>zoq!c}Pg7Ex&l`+ASw@+vu?w^^*SwA~$`3`~!|&DKUrH7{^f}z@?SM!~(7(O+rm( zbKV5-z(Hp6AK64>K-tvR(jvwdBMJFNN4+5+n3EF|Ky=u&!rM3IKit40ju%Plj2f&g z8Y*N8|F@~0J^cR^Dpr+BM#GBsCv3z;E&r2VZt?&&1jmNy21pFr>z9696>t~IWqk!w zllND#ZmfVtD6{%o;ZI)v$m4E#^#4UrI8oqpN5(i;wJbO<&!A`M!?mx>$fgQSlPxnY zGa!H00=0EP(oY3l9;f?+|p#Q9EYtl?Mk7g<+kJP ze5B}Hqv_Js%ZXe^RV^qq0EQ*vG!9P(91ie9BpD3>d?+ONID~{iVhQT~!s)D?2E}p) z;7}vqbY|?sVsbuQqgvUh#3#(Png^sFlc>x(BaPu7hDHGBG8M=w^C4#dG%Jz?%s&+s z4Go~Q(Om{&P-1vB=g$Hq9wuQgK8z2EkgW`BDHNxoV@IaXwH$QwhH|_Q?&~8~BZT#N z)_&eWu;hRX1~v$YVoiAHK=&_0(Z$-D8t5TlHi9)kwMc>HSb1K-Ei`lKn>S&Hg=R{H zWo4D`L@3GUMSdj(2px+N=`NczS_(Xupkv74e^AQP9Q3MrVPW8du=&%o5<84+ZRGUQ zN!{diTbHYs@b1l3ox-OO6HK=_u3DA~bcu+Ys*nu20sHsxgRV6Vqp^Z$O!`vvg=l{8O8;6s~vtO$m+L8XX@ zj1SlJpAwi`A-`;s=elE)Z=TXYD^VziakzMRl~H8~1wL`U+4%mTlDFc2HwLO4p6Dz} zk_hA@V@@se$@BC4y*Q5u2SZ0^PlntvLbD)(s9cA1hKxlFN}ebAgSuJj zlFv5#wLFZ*IA91@K^*#>KfZVW_qUslJb|!QgBK!F2SWmau6VjIu`rsozhbg9J{5U-?b^2spily&Pj`yWr=8S*oOap4K{r+9@##X{)`RLz45r> z#(17OH-&;!Olla7LS><1kzydBYPmS%uU}{a)R7HCnl^xBoQSYwLu5zYE@&AojisRVklkdRwgL%dtd2g982JV~8|Nl9P(A=!D<`0`^3HKi{`$ug8jCQp-(6L4kS{%i5?R zxJSzR)7-`GK{AhJFKiCSZ{KB+2W-L|W4ozfXJ0-;{Aq(hZ|drh)MO1Erved|@riYT zdX7Yt-qR(LnM#Xck1GZi7O6y2Gb;xWLK{!2rirXwU2e5j-hs;?Cagw&VpRn?IH=$pYJ!_a^&Z3!8WP#Vp7je za+h7{;~KHQtUr=Z_Z7HQuC?b*{g(|^fVf~&zP+7LKs^*v%%GlwB*mOmB~2MfEA9l3 zsvi#`f5BA3M4XY`$tn`Ei5jhA|^im6kcnvRZ%e>-n{>2e^*xAPt!w#*I{Y~gz5^p z15#Ygok5?j2NKhXQHQ~SW2sNiRS4AjXJ`cP&V1l)522{Bd>;BH{?8ZOO4rk(MtZ7O zQ&S!$>a~gxcd+U2|7LeSRN*w%qihc%%(zXkzulaxEVH+`axh$Oy}+b}+>%1f$orA5 z{qkZzZ0Gau&hTo1dI@@^HxcwJ%Q8|)g)99EDIXJ^UMKM8J@-1-Je+yn6+@w<)D@jF z^NvX+)uM4QmeaVu^<4s73vr1&C)$Kh%XZLA2L6OloUYJ70(hTg=nHNvkAhrQu8pDF ziGD-j6UT3T{ewkG#>0{6v(1oNqhk+I1pS(vY#l^~n{WK3))Bw-`Yw*vCLQc{(^FIT z$4qkHKG)Pd;c;lW^`ChoH8XqoL`WfTb>_L-ams}_^5MpVjTFKP6O)QF2HPcAh<1B- z)8io5=T1nvFGd8dwi%V%xV|?4B#jo9W=b2Kci*Y|_R}vh`Y;#Tz z2V>a1S&8ATZ-0TlCg8!dw<;<_m`fJaE+(g3Q*VuP-PTSn9J&%1fPMq_xWPNr70;hH zTDozW0Jp8l2sCpOB`l2ti(T|8vdy2DdiNHh6lyEJk_`tuJY-ra9|i-Nl^B|pC2%1R zn5;t;5g3e%(_#xU_|A221oq63dUa~$RA~Cn{K8g}h@p~_iEYx8?_U|l8&6(CQs8LS zl>pm7@2LsWDjX7S8gQD9JOXpa<68ChGduy$ut!bTsyP2y^1-|;C^J;Lg2sbwUu$f| zntdkAdQJ~uPw4AC$XQxdSi|)G=iarOK^sECd&`ciwu}SwAvp-cmRA)pNmLzQ!hLIU zs^YFeARI!M-y=;c2@h}|=ZOHQL^2O%CFN=favIxSWVsw>+nI>%9w>xS8xKi##jRbQ zEv&>8`Af%TQi$N)&y3P6(bW5K{a`&O<>g?Cm*4#^ix4zcx0~&|1#(@nstxKU z!>h$@w{(%Y@iS~xK2NnLCXuqTlyu~Ckv&}K`wjRloc zeaX+qDy26!?a7cLk8Urx`S;7r^4!kCBlNfTyAv$_EU)Cmuv4u}aGQ;$GO0n(d5`Mv zYU~jjG%0NHhiHF|xWT(U{g){Jwr8f4Tu)-IY|Fdyll{fG8b=utUR9F^9HM(mYj8UD zLr#w>xBkG_SkeQ3SDEZ&!)ak&5AOQgb)Q~;@EEe{6Esshf-vovbG&g;K@|U%us{$yB-XGD0qqf|8 zNO+o5VLrG$3*e^PrPC~?GgGOGmRklc4KX_p>??NUzM^nDjHxhJ$dPs4>mqWX)fhYN zO@5AkRN2$Nl*B+Uo*}MM`Mq&yf5qBicztxNahmz177at^=_Q{SRM1F7-O~ND=DPcZ z$3bzt;_%c6_Le8y_3p6H)VmavM!SC}4r4s#F%|N&r#*)2oSf7xFI_Ro3fsdDdPPp> zZktUf77_YjHyUATYdfFl>w|A-s$)^GOma<4@=zQtso4@p5ODJ{ytU56J$f431Q%H! z_2x$|{@k63!G7nLr+0m7Ho_V;1u!5Jo?S)=$v<6Nku1DPRS+-IB#-un&mcs z-pkgW6HegCI6bgXZH{K8mwAEo>brc*3PSDe+?DJGU7_vi4%34>^M@q+JoBw9>nah< zOk#KM)>x*z6#*rjTTsAfwLtZlmGe{N&{OUy67?_#Q2X;slkdjfqdy=;8~eTmi>l# zwU7DaZgz@=7MP)nGl-M>4G(TZ=3IZ2sIJaolWDhiA{JI=8WYLbi!GR~FHw8PnWU4% zPqTg?JH+7k+XS9ic9D9QJo}9w5BSxJG*wH4K_|2(0v>h7^7=6mAeU{1Q2hdu5Rm^( z=Pfm3Dqh;mnKMm{6p6u~%RyGK#}{(O9R#IE!8KR&44#m}&4 z2&r^1XI~x8krRP#e?bBO_*aaExUVU?U6j|=ak#sZ@@qhhpCdxrd%SZx+?ek{bGg!i ze{o()FaNgMx>8-z-qC#I&zR0={Z@j*Kx537<0)el8#H**+)rsnS=H5nN` z12yn2p|&R_5Di4o=ZIiCIVzuBUDmb+DJcS3Dc9l`3WfY!T-CJ6k$h)yqEh$O`g^^KYkFid}#4!5YTM)0bSaU z2S=)89+3V}8cjS~>%Q&H$SPCsvbK^nSY3MUH5?>wFGSKj6u)8f@BjI7Ro?>LgVKP0 zgPqGDml;r0ghZdA_74zXXz>JnGXec`*0& zUq45N0tS$>${3VF^wygO?$O8yy6di7ja$HtiOKmbnUSH<*fa9CJrok)tfY#Hy6unl zo4*S@JrQOT*)?_rS`b)(%rrBhGh8NuK2~trnMDY;8k}4UH$VKa_Vj=|Fc$vt)e6(PCwB2PqRD1d-IA^EyP4mH+jM0f$MPILOt}*`M*sRaXflR8e{( z!Q%q`bMgZ2oSfuHK192&Jl*bgWsQThyrd+Q-!m5>E10)&kZJfIaQ* z_Spsb#q+_huC89G@ur!Y;yzMYZyz2$u1-wk=CZtVcB~5uoD%(m)3txGHH3iPw+FL= z2svw>f41M)-|OMA6k6&r8D@sJU0o{m@P)N;xioK^YQrHKHRLpnPIPMhTA9nZ)t;vg zuIVp_D(R)ifkORd%~)CWrF1aA>v{3vhTJG1QR0|H#|r z9W21{IHsuu5N5Ram)jSpmc6yK)DB_>d=Y}-hwSnxACbGINi6zzB*6_{{?gG2P|5*f zDhh>BM65c(04~^gbnrtxD)iX`Bo{RQAxM@L_?G*!)z!v-lD`9N42?>@f#61D*gEjH zq()sf+@!wz^nKfnk(5u@_4Tq<*r{kF0KtVEjp*sGK?MN;LJnG*wtUTk$1n4VJ_4nj z`pRx6m5tqQIoSOa(blLtU*G-Ib^FcO++x|fB9rUb5$%*(T4-oS3*z41Dnw_8gYn$lQ5~1JBN7Yu zZ547-|Gd38o94zH)onBGqm1dYwvzdBkDE(aRyI61868u;A|GtZyX5k#^AiAD3`!>^ z3%H&!%F4m-g1bIivO2A$sX26bu*PYfUQ+URB_~(GX2>KxlU_ixK#hS?n1pvR{T#p) z?T^0=oSbs3^BVt>s{7sTPdGmd);Sz zdv<8=anNLkMJk`bBQ714u35k9ZjsuRkx9=e*%p+i!CRm#Gm@reyOGOoSZ1y+c+}?fWU2o)I=PqZ3jC-6y$n!bOp^Mk)FZd-(}{*4QK#m z+(Wpx_7rpbYiN9az@6$^Ex|aa)$qQ)zl)LV7tA23Y`n;m^wH9Yct$d^a$|kWPhY-3sCT#lxrw{G^G-{A%!c44gnr^;AyH9@ zuCk=0Lqs@mhR2^q(530LaOenkimeQK9D4GzNnoONmCaCWdYUr3!3AAX{)NLg8=Ag2WGKa+sU_9! ztFC;0CUMkw64`!A!@wlvagosx0R}`UCOK`te2P8R@{znyVfY_7)+X<7R}sBRLG;J6U$S|lZu;gudD#x1d^+bag#1W zeZA$Zq=Dq;Q--5|j(3I}9*TQk)T_B3L>!Ewh-~OY$LBxy8YtB0ZiM{Rg`$bKd|1=P z45XN^{P3WeEK-ZPD=!;j5knW&k$aG{@ z?saG=kk#4z5{aP*+O*cN{b+0B5u|N59V*rx5<4>d-Gf01mWAJK7l(1cGAR}Q z>X!==h{b)Tt5ar|*Ab=En@Y$ndwDr@IGA@Zq*mjsu(|2H{8-`}pOY-lh0>0oM}usg zZB6CXCV`+N=C?wvqWy~rRsShg{q#7E$+71}sLheF4Hdp6JxS9tH_R4`Sgg3Bgs};k}x_GU(rrt0}?3dUtU5A%EE;kk#a6r6WKQxhsPDGpssnp7p9bjtqysmDB0xnhTVkPk zd7ITVOFb`jQgw9XQ$M|UdC)?@%{4qyzP5{c0B|WrhH&IxJ_qgxAe40qZsyFMX5pNjZ<1j`H9gxHZoU3^Rcy#RZ9aNGoHChN?+~)^tgFt zBIvFzmkBAbfWQzDM7G!Y{67M=8RLm!^Q{=JIFKsu`XzrcvdQeHawB*yxR9N-peVxJn*EZ+Ffsbgi z{`be~xGSfanNN4x;+3A;F9)}+LfUE2(g6ucUq zvOHO18O+bQN5x^Kb2#}CpFi@G8&AHiWv0@%M%A)kCoXtIdDm_Lo)(N#f1uc{Q`QxS ze>Wwc5wPwLH>}O`DQVV5SSMYI{0N5kdX^@Z`D{hgL^=l*P*9A=IF@_{Aa1o3Qpgp~ zx{+iyQaF3T%ssDq*>HT8XK%JPi7P0muMf@#|6b}nD(k2DAplo$n&ig~7+m2%lvv}c z0u_8mJx=Z&bNfab^|{H(sAgYLXWqQ_bT9@*4}c5MXzj3p@5*C=tmN7`s9QlYuA3JM z`i29j+I{eB2imLfKy@jU>cAie?SC7c@%h%EQo&w%b)0}kQDY&!!v45LH9wK*m()m+ z*Ci#pBVN!`1YlfLRNKG54z5Jp5?lpB8te;zZiU0FBAb%JZSu$(&JHk1ul!fHwKXa& zjn6;~PmhgF!uP(7f3Q$fVPgz-ZdbJSB>J?@iM97^W?>`tOwtp*UwGGByi*woe6JM@nH3D?W*Dm#4DuA)_0*JOP1G7E6ezB} zA2gnH`DmwDoJ_un;>(8rIDRQ8PRsW(rMBRQp0fzS!lOi&o_5ZOzb*#kl9I^~hX;H5 z;cIK}Oxb!{OzVtSG}nfCOq((1DQh1Xw=m++T>4&^4E7Ub-Oe9e|<{8@=fpN z+`P8&Eote&`=-LLe@v4pDjpxt2i)IMt9+u1E-U*s^%FV5<$@3q8j80V;t~9?A8qU%ye~kW>9jKL{4_2St;iC>P!fzrDWxkU5~X}4|H|! z-bK;Udd982Jrf}EQ}Bwm(8B1;eI~X=*DpbB9#<*uZX=_!dp+#*ilhQR)-xsJu*h`+$x4mTTL89`hDpB;o-x{HUvhB@t1$f){Fp6f z9oYE8j_%Lih9H&xuY=sIrIDX8IsFU!t=CK4&S*WM1M&m;U$!k7)qVM#DC=B46LMNE zVMVgEIZUsJ#s?#?vH4tly5f54YZp0KG?Pt7kLEy!aK0DxRpHdOp@@`!d)#8V4{f}1 zjaKh#S5i=K+U&{~kE^ctjp1cx{x3goPdX?!@HJZBz`^(a{c^Lve|9U#U+bo67PMzI-DvO|K*p=kM?n3kW9 zm^XjqIPjlFM=*X#6DfA)d}=OC;c=2V`Sb~^)0Q*`KS|ry*)F2!7`oW5jIZo39#Nxj zZmz!co16P)j2{~MPJ2f)aIDq=16w4JkXGRqlbX0iiHpN^i)t>^+fpH8*)XKLdys9) z)tYT{@@2Enw{p<5>Roqy{5YBo&0_k;?>ptj^;|c^l^?7Qd8bkMVE6cbhoIv3EP2@X zW(o?=0RQGRk?*_C-zz1gn!``c*0D%X5&0@!IdYGq7<@j7ay@yBQl$C#>3vO2_;v=R zhRVxyj&xCKuQ0jRgRs!;Q_lGKx4C$i$Jflw{VuMdzuQp7NU82d^9;o7PVCCI?B%BM zS4I2&nQZR&L$ehWVqTy7OzB;foU5^)S98COG?D>~HZqijx=_^kC&f#rR*`9vTz+lCpj1Q_BAFBikiqMvI1&{7w$^m{q8n1o~6C55s;Nj4{8qD(} zrR{qwu&}guVDicP`y|QXAqK{;V_rYMAGr!oIK2}lU0%8e1mHZpNknu&N^8yCog|`4 zC?55?+S+fEJM=U5*FDVm1gkUq z*1vIRds9>$4$fzvUX9De?Mx6~bQAiWp2OUo!58GW%9zVZuNOItGFFM@m zBATf8C*;KAR}cR%k-+Qsn6I^Ud;8b>s5hcQa1I7O)bpPnb~)B0@OX=b1;^gP#dUPl zyY6RY1zV)@h-jRx2wlq3>Pk!EERtzUA(@s2rf9X58gB1=@Vd6x(}VGTsi!SpCAd>t z?-O`NW4}s^okh&4y63KZ^h&Q+`i>8V19~wjUcVbHIr71yE5c6J6MM!p*Mw44%W&wi zW_1P%>D#rB zgu~nJv)uPYd@3mTR@u`k!DF)gXX5_MiH)`{1;T1MxFPG;LaMtPimzChbzE(qKf#ag zHwc|8ceD0PX#&q|4Yl+PBZv7ES`4+PX1RZ`u#ZnZA=qT(!c-WD1e_m7T7?b5*5+3WE$%nu2v zax6}^A$UPa?dON@U0&|EpN(Tp@Pf~Ye031lpTI;aJou50F1sd+mY*-i-Mf!9>u`#+ z@~*!BT9`X`Dt0J(W-qB;L)o1~Ao7(~)8)|j+mpKwb7J{UR`Py5zTI#58_#CZS1Ru3 z>Y(57?t;#oaIsC+GyC=Wwdoov=-`TilIr5H(~PaNuH0g>{jRq-?RZB7Rz}XrNpbPr z>&LVf%(tttYcWjv!mNQ;RAJ^_S$Vuj#A=@peS1Z(`lWxgyb7nMA|~+;3fkV2tD{%M ztEfn|((TO&bG+oExbVIEV?h1^(QRB@Z4(B-cwE!dr4?e>QjHa?mfpNR?M`67_ue%o z)=wgetxuvye4~g$gspElJB9b6!R_IS$>2TD2b|(-G>^1It^PEX)n`S8pP0`(ua}^3o|n_b91PNy9B-oB%n}=rU^}JA~;m>g4^|=J0YnW?d{?B zKkhrGrW)_esoQyxv0ayl{F{rIb_Wai&M2mrF(t?sL%kaH_U%vJE6WdxvMw)*4KM{x zuAd#kCOc%RtFyckP%L_Vb*VxcD_&ptY+8dzc$<9BTf_#V$q_*KCDjv76v-a^x-o6(%N-QBAk zW!#kWcI)$)=cAWORwqIdzxrTW?T}MT{n%eQiQx4dQLd_;U+6sYiR@g+mb-Cv=^w++ z>h3DA$gI*x>UOF(aXqJiixdCx|{EGp(4v6a0JAZj(D8 z?%vJTB+1Faz-s;5j>hBq@*cqrZ0tJ64>fjHEB6SwT37m=xLjWfhEdZQ@SU$Rys9-t zIXg8mow~!${>jH6z?xRHPd?V4KzI4ii(mDYKU-!OBn}d8-hA>D>8p&`BBtuTSD1-P z3Ok>=*yIR{fK;26_4(f7!!wJCrhFB@YK{6jJhJ!I#L`J~?YILu*Jx<&5Jj?9FJ?+b zb7Aw^>zxEfTa2JeL!UK!XjS!kH0E6zw9fIkJTcTJIOjM0KtumyZf?q?`OnqCssia> z=5nir<8mh(((u_;1#MGCNvT)L(VIWVD{cyfGZ$=ZWTpFZnh3uQj^>o;MNcdBc;26h ziUYkfMWJ;fLO6n+hfo*Vc3r$K&@7nmP^-WYddB@_xTsW;UEdG!si^2H7RC$TN{d&v zWpg&ESuRJPs;h(Bf5x!K;xcL#4!iUeA3Mra(m&D&YJwWZyz+kmbM+m3A29#)Ly4cR-`vUVP-vIe@*4UVv zF_Q7TO3Bs!R~OY{J%w))Zt*w_qK=H-ESNOR!JtoM+P*WxiGhs`L#lG2v4+vmHE!_; z`auvX00iIYa#Z-OGw&-3-`$0yF(C)y(r@kW{sGxz33yq{1$K7qe^fByW;U0v<;ubxa!DB|Z>xl=GBr$RWxk>qdVN=9TreL0jOFy}K1X}v`cmwq z?U_?aYWEvvma2&e3YF6X+@T><95lXEd2T=94!0&x)UJeEB-b2ek(Cyp+$DtpdZ6Zn z{<_FIQ6vEiY6B>g#Ix@$(?&=OzlsoGS~V^i)G2}nK$dp+p2sJK^?hW#kw<~W9hr6s zg8h}|gOhKQ5}C)!aj~#|Olh)~3RPS4yuK^J^gA+g?)x>f_I=l_8`!xHLdVb(cuU;Z z7vt3c-n~%$*tVco?V+eRG}5b85ZjY;$ji9l;8V@b+oeZAv%=tmx_cd@x4wUO)?m63 zo}#-L_suv0A?$sNoQ|#$=~mns^YgDi!Xg_||8`D_nHZb;#9=X(*v~dS{$8b^p=+ttTMk7nGmxZ z30!vN&a&N3J3WF(ga5FeYd_A94}Du+_=zV?D!JH0zTvzm)`eh!qFeaaAB~lFda+5y zQ6pyO)07yB)8%MTe1Thf8D$rJR)&Xg z;xFUwx^vQo6x+YPK0Bz;)5FJNqA{tl`gX*XN_SV;|Iz|1lX=LmkJhOmevGQ9C1{Y! zcI(k$28oN^9n8Hc#Hgi-10Rr~ltuBu#PxhXQT3=b=+71IotYfpcgU^MH(FXv_`yOi zd>ma&f13|Lh=5-_TCz6pH91A?*QgOY3sI+;!vZR9SfZu*04WZ41^B+@5H>NQ?_ti- z)=H8Ya<-3x`D4e&w7J;ehr{|^!Waw>4-2wY6%cb-No7=WtxrK0!KC6`gK-BBhFIHg zfL};eVqvw*-CTW~z%{$B^6EiOeU$K5jjH4awgo(_6`^eT9gV#+ELyj2+(7Ydyp$CT zFPJ6b&NRJB;*!32YkwtbW4a)IP`3L(Bbn zktVGi`t0nt;@)^_Dzppi?f}bHnea$F}Tn8cfn(lcVN>ArR1Q;GgtiX(`bOza zC_4SMx&4)i`;Tt$%ST}LI`s*qF8#dB-B)K5lI}^mO+a_dU3bOfp87Q=ky=&J#O}%R z@j(+V>@L}*k@$M?a2%R9f9a2YiE10uBebIpyHqq_d6BdVwMCki25TIMT%Je%&7 zz0lUYrNKuZ6Zz2|37knz?~m2SE(w|D&rz>W74Yt)i=pD!t-ilaG2H-#cD~7v?&48q zO5x(ttYRD=5EDMyC&gqA%usUcibW z5}%)j`0xQ0he^#KMyQALk@b4xRhiUCX1)QB;}b1wf||^1Vh+)^i}qpE>^JL52>UI? z-@cCnua=6oP+Jj+ei-Q6!r)`dR9&z%tCRk$9c~1)o)gzWcbc*XYm^0{5%W-!7dxJ=Qb-Vr^Ll)Kax5pOX4g$zn5VK2JO5-I0Y2R^iTZ(m& zPt~Ilp6dEo#(XJ#V9AhxS2E=u#>Vfddfe{rEw?w4QZ3D&$C&fn(4J>7sa>}%KzaPB z%0;Qnh(h4}Elkh`wXf|8d4*~3Y2hcF7a(-;9HQzE$NjO?rqjkl@JoKE9H_ga>NPn; z!kYRb{z4se$o%4VcKwm5>CaNblSCFeNF@_>XZ#<1g@U#z{s(&ea4-qdI zab`u%bShwTvQ#oM*mLt{f_r<(&}V7BWKUH$~DXu#D6Cx?MU0$KtxaPKY!+`9L~ZLEYct9aR{MsPvYn^Dhm;>p4g1{%0o68R)Y#1 zdq1HHWc-q4U-R1B5R=v1r)So}>r+e>7EpI;Byql(u)X~!FYmBW1Jzj)gT>B4$qh>M znA}YBEp72UN6IDCN(SU#QB)tIzyRtWKZWZd`EYvcc@wv!SAo)y4O`)lK-uFfJ{M|8 zml+iC-e=~Llm847G>*fgxWenXe>|b!gB;_5kG<5BCrwZ!C?}_^vOs?%5oP%o0};>n zC5)!j?R2gMAj|vrxmlXp!;bf)5snvx*U*<;A_w&6R~1Yrd@UNgoHh}fb>JHiPz7sY z@imdZ*qVx26ENy%0e+?GYmP@)^Q|aNL4rTNKX-7}(%N73K3HRQ<~Uq$XN!p@%l2#yZ=lJ}O4is%WjEI| zrl+Q!8B_u%lthNFgl9VNm6i1@c3uUh<6p0@x4Nil;KY!Tv2+koC&nTc8n3ACAQ7_V z0JQCt^YT)y=oPEQZ8}1hhGOu(@ z^S5qLVB_~N?<0K|Ir-^T_f!Lpdu$$|AnDZ9FRNPVwV`Sz&jFOj+}t!rfP%;KM&%&{ z=*4?4uTDA7FG{Ua<&D^k@dQq3E2TEToYRIeKp{3P|x+HF|t6R<~n^HnYl za_Q-A&bM6?3M2%ilj8Nva_`5^`Vk$70^!9I&K&4|ajO*Z8S60z$LFQkm=&UZgM}42 zN^1J2i2_}RD?y~Q8{_Jj+Ewl(YeOhPEq-GLEpilS!EMk^ym5Oai@S%UNy}sQ&tqjZ z-H|7C&KbjW&{5gJ0u0Pu3d>Vl^EH-q)RA%BRr&h9PCezi2M24q-%tD$h&QImOor!n^?^G1f_p%3{MBqK;5Na` zdH|Kg%%YE2*!IOyl|54>yU(h{p(&pes-Pe-&vj{OKQ8ZlyWr}}V^`J{I9$bwiu%olhFY>IQ_3kg=XEK};Cgv{GP_oU z(lkMAvjlMvm;q*ux4H7}3wazm*F(d04u<{8PtmaexT~_d=1;(o>5}gH1Ca=QxqO$e zI2{}gddJY#E2nRc-zQphNFNyrK zbD@oBwB5*H3SPZeTCJ@Qo_+gv8zq3y`>&qs0Y0b(LrPwt-68u5K$PhNX3gJhV^^~` z;-7u+!^gLosjW`2(v6N;Vg533I+^y_d828rWqS|o=yE@lyssSO`TZ+NH2V!W zs|pODMLFv*?4eg}gn?{=^0iM-4_~ROR_Ig7fWwVKm+w+nidLT4#xDgv0aJaGo#)>&CZL`0p|-2JRFhc_SpwXTNQ zupx)$<-#W~nWL`F99cAoOL*ZtVf%Pl$<439dXW6(%VNw%#xqIZ089!Bl7v0GNMVfh z3>44e;!TMKp~W@0;_b)r2?9hu9v*E3QtVb9C?*I#~^j8@?8D_tO8HuF8(d|tNWh&wtr`yu3ZgMM!oFG^6d^? z_Q;8ssl5cCQ$8p&eP*iO^|uC;@`)8o?|i?1oVo(t#m_OHqxB3AbHOwxXvZ})G~}&u zpWc44C#ex;M!=K+duZSMt@A}oH^4&ByklW$sS;+i{$Re1F9IVsL`6ksj#%(-t(-Xk z-l65X^XmJOxwf_pGzR(DQ;AcI*6e)wHEWPuX`=l$NF&@k7I;cw|PD32Z*{b*<^YJ@4!L z3urH-Cx90l0$45NN$~Z(gSh7S-yi?~`PL40n1^U!V6eEj=&gZ6|MmZB@67*lO#425 z*_R};W@ju3nMz0tMrkjll$1!SD{E0`AKQdGg^HSpTt&;6w4jABSz0WWOkGlDvBWV|b#{OKY_a|?R&ENf z?Z>Dgaj8X1boF#?ZEYUZ zPPMG?NChK|Y#>-re`v*vhSlT+1_FtI48n=y9UdNHfc2b`XEzQ!8)G$9Z~5{oZY8%L zJZR#Gf!%R9S_cmt5F4(D(>(-`w9c#XqMHvGF<84^oTvvYp)hu=FOW3QcJLV=J~S&@ z4Luzhh$;~AkYnN7U(d}&y=xVmrk=Crk5FJ!SF=08;5j)tL8X1s*jRaX>At%mV>g+^ z96w$PJECb66De0hHLWO965(7J9hrOZPi+_M-27K1U=*KEbN`?r z(A`D=)kRO+Qga^4KxfaMs6K%r@1Ym*S zG;Gv)0;>xOeRg(sQd8cp_h%kD9`qB{%jY~EX7Lb$>($gFOYR6mX}Jkn7tISna-dyd zSdxR%6${CxDDeJvowQ3cwyuGY>azk{zwp#s?Jqd0vM9N?oZ=4EG&#=`Ub{bgvKX4u6AZmesfhuOeR)9C42C7 zle${n`#*ULD#`se;Olfwo*YySSqT2;j4db!eB{@|!sAhn969pqkL&j)Ce93EQ8_AX z-#$CbzKiQ&2L3fYd-dWkc5-5p;erEihvvNqdXd2C?iskid3W}%_Y=mCmp#i%6@mw+ zjA2hguAd+;FR%th@Nl{~fxLf6lIzTgL13rrzO6QwoP}999OqUnXa^ zmGgDK4uQ7J@_uY>UHPqL%);&KRxyWd;g9zCQB`%|fB_bobf3314#b`xHl(ezNM&tc z-erFGU$z||xMlamvk;i+i{IRh z9(M?6p23SP6Gfa1|D62TbzmKm1;%5Acv^p+(0{W2p{%Sd3~=x2$jHMJx<83yel!Vb z+Z8qSW;;=-| zqr>9tTiV;NZ~tp=OYG#ylM87Wpl%ThvUh=HU()!SeK~J7oCH9MH$x%mpNmaE} zD$QmhNXYG2^U7)8oMyg7Y3LJ9+#`o@Wm`=(!m}t|1AGK4{-LEskcwSDD(r9~)0>{D zX=$Dw9wn!iM29Kd1U@(LlC9Nd9-5QNsQcXGQ&^Y;_6p@BMW9`1iwNpv6@8}mdW>g4 zVUZ6Ye%FY4NJ*s!TcW0Gf3x~@@p4ASY{wf~LBLrh2?_NZ4<)))-tf<98}Ou}NAGBz z0MWaeytyIsAyVGvsut{U;LsAUA;J_6M3jGhk7=iwdFpCR{bpclY@7^nW_RG=Qj}oW z%v(Vn6Rt|P^@-biEHZNXgZIDH?Yp4W2I8!z(#R2%90jEf0J>Ka&{k?sJy`^=?hE|O z!4tF=Dk=ugntWoU+4ql~{M%<&&0lY+ZD_dQmjJrWMEbZ(l#6?GcX!7<2?`9C?psOt zRr*CG;8elK89xKbZ~4yV;{z+@mr3Dh@8DgaSt;A5o+WSG<52zg0bz0&)^;k==S`p9 zy?25)6J#6SJx55Yg9G8Uo*lS6yr?HE>Xf8yhg$40vyAmyww$&egA`Y0zV05qjxXwS z<~%4ZO~ZykaH%g?ASWxU?OA)DM5wH+3)~%O3TvVs{+k@@|BU}2hw%|m>mkyDn zXINz9Q{rw|*bou;AaTu&>n)EV!G?>mpvU|VfzXi0v1~ z&y^>yvM|GX`1EAah1dks4NVgkFLJ0y-FFE+46K2__ z16-5ayou9a^a^V+x3{lC;lTaROLO$zLt%!z878$Gi!ZAoEg;~*xemPO)k|f;0!QZT zR+~Dw?iCx@l{HDmF~8aTWU%LLr1&O%|0ItTzE!mPrk=qeA?1O6w#C_7K7Ez9r*%5F zO;+b;wy91li~WSQ2)C$z^PPZWrw%DCU7rvO~s1noTZZ zsZ)8)V*?={w(L}Ek^h&bt00N*f0aokfPjYM_BHLeX{wI-_wTy_iYom`5!8}XtT$Rn z1?Cp&xe;LrENHiomzK`BS5wms`kJ(~Xmy`3)%u3H ze^d!$!4ADFpDk$G{S~?Yuw0bM^!7CYEwzTLHY^A)s;~l~_rhp-+hV`@$It8^QsdY9 zfGiSHn+QVg8(bBa$cx;YQ$0U4aRkQ2%e1kFo2zTXt5>Jw^b)4j@d_}0ZY&MwaJImF zZo*EWM;CtRUxmlv%3Ct6>_ZAvJb)x%omHu<;qXJz)~$=YQ+Xf${;W!fimNPZ58{7 z%iue4uqXfY>-w6N(fngE@L>Wb5FtiWQ-{k=2k!R17dr9e)Bam3QqO%H*()x${D!}; z*t?EY+<9;uacBf5r_b_F2E{6FTWx38+0hXnAO9TJ&h>yXWdbgI$Di2o@Xd|)!}?E_ zDAaG+ym|1rg)2Zt`ae8g;7t83Y9^h*WjmXtf;JSNhmE$!##U)Bu0h#aeK2x<0LVLI zN}-~|aL0EO+mtTYtx)s*xLLeU?&f#P(qQggW3=*LIPY~hXm@=S$t$UhMEE_dd+z=g z3zbz6pCNDefZcr;I;xc%3{+VKRbA->H5Ze~!;c=$hLUYG7FQ9Obl6G|-NxOB@U(#0 zKkro@r*M?$U}WqLVACjl!fP?NfGlO};oy;##rr^1F}0Kt0Q!|`M>9?YaORcyVLVWoGwqD{4?R*Pi`&PuSgC1Z<+|TbXXsH$+Yg($z$z4FzJj9juWLH#q#pmZ`~lTk*n?F){W(_DPs>x2kGf5M8Utc@RA9 z`i|yQT?+b&HwT*ZO@hzd*?o!xKKgYQb+4s6_AF*0=#5zJvb;aQOqmq?r4~JG+@SQ&(;sEEiF{(IEi$Sx|g!^ zP#__b$Emz*yPS1Tc+!3Xx*m-8t5%$k*aTY2aRQG%&b#lZeXw*T7$iP3cEpQ1m_xq$R=WBJTO1`nvT$O3dCdVdq9 zuvbDiiEDg5-mTAy+|301b0*W1ai(-GgM$tF3Fi@X`)(iz;GPg0oc(rhuB$+imixn# zQDodbNvg~eUEQ{BD_noOI3l9JxzCEx1?=p-`}UD^&|$Jz7xh|*!sG_j*TZXnn>1yL z(!uY&PBA3m3PRa!rqgmHnH>(D7QfTh)>hon#W}(xwLj?B+|WBR zl13Isi>)^cZmP{*hBx7I7QT5#h=Dj*pZe6oEsymzU;j82G*B77|>H49OF6 z1fpvC<;w#G4YJanws7G!I)ph zyLsFQZ`q#09CEIP+s|kkg-c@v(H80L+55wa>JVf(rhLui!I;eS;^JbVxi)U$_9g9) zpZ>NciYHT@ja<=m$iqbsb2$h z&wi)r<#WUFSYn^G{fCYkMV!|VH(-n9r49NrT0uf?Qyq(xy3r?7RsJ?qvAPlh@~lUe z*U#>>nTBWx$-s&UMg|6-5f73qJ9_kJse!Mdk5Be+udxaWoM2s`u5P_YQAg(nr0vqx zf;AM?6)wxl;=Af{az|HJQaOq{DrN3$P0K>GcWnCiSjQOy!0t(-R8&<3fs^n8D>+ky zvD|aNLAaJ)*^(TY`J`g(`0@Vij?;Rl8MyiRX$4^!f}OXA<>Y+9o+SrAvEOv{^<8zo z(crY%j-ohDJn`Q;A}8vXf9xATks#qvp%q*(e{xzM$o9 z=QVDBW+wSvGVJ}nlcr9s#oIiLu{l$<1JA$pyXUV%?lW+^@M)$>Qk#f=wj|@+B7Zv* zlVZf#M7w=`(ZEw_T`dS83Ip2<4UBO!8tD2XP&P`0Gov6ain&=(V3;(R@nu+NM{**o zjW5k;hzmt*%MT$lkn#?b}nOOSXzIvQB47D5DvkYbpzkkS-2K5K(P?T^NNB=*H^r4Jx;B zdoF1uS*$7>LV=Wr2NxPd9Qq&f)ASPvn$aE}AC#Aumz9-uHyz+T?V)pPO(#?fuR`a4 z1>w!7iSGMznvMDCxbTIh+EE@v>|x!P>wup}>@wVr?OF8^k7YQv@KBBC?8%dt z((l$*YYa1)K5TW0!G$Z2eU!Zz1=2&q0fq(#pTZd<$I{aLcgMYw5{s;hYg`I{Mo3GW zOv3NV^3TR<<1ZFG^EK1ez0;d^$dtIMMM#S3PxPk&I_NA43}`4GLlrG2#}{`=cZUgu zYQJWF2jiFGjqeKAKU)=n{^~`WmE$qgB=LxcBZ|=c`wodssfFiqxQ;= zp6wM6rEI6~wrJGpBw~o+SnPL2Rs%)T86?0z&f1jga;8RjC4>LHJD#Sl|3mW1pXQz* zbMe~9W%W!5aT!Au6r!xhh&_FYC!`r}S>b7aITXc{X+Du}NJjOsnDz&=v{EG>V!*>n zuZ-ODS~E@-P1M@`qP{-i!z(&9J-9z*pgo-?wPb=UUYdr8a_Ud=+_=8C`W2&hPyE~3 zfh)$UyU)V{ic50tu6=`pIX<5*M9POW;w#A=0xa7=} zFKeLIc7oRKu(=0{Zu2|ly%S2z;Wxi@b{buk61vOE zhAm+yGdaJTPLsbcPL?O28m(J%4C~k}+`rHbpH=wz!L3_MH}n)|EmWt6kEY4YAfl(6 zy~}1{Omi?J=GsVZB$i4%h)n|JK8Bg1on31ReLfn4(%&O(xU5Jhj8~A7o}Tk)^b%Wp ziwLh#p=f%9T3+q$skZB#y;vtuonLsuNR@Q3&P*JYEv#v@y79|$W|ukGFcY1IfaqQ< z|A6miWIuDwm6VKwRrm^a&IO-W%&D0CGscZ;GtW$rGW#>?b&e7p1*`$(%NUNGs1Mjd zzE%=!d$={@h{AL_x5SZu3Sn z4%eHCh4fqnxwc<~mIu$(Q&EzH93ugKmTP-IlHv+pGX01i7p;?@+5khBA748Dig3`d>gDL~y91$_2IkNR0@2xdcENliD#f+6 zwM?tFF$9vsE|B8FaWjEI>n5G~iX5mDyPLWAG*VUvo^j_+giL??mGNvtgR8r{50*`= zFtS-oON0om`h(c;3SJVgP{}}`vNDUdhlm3;Gtew&eoe;<=08UBw5Fz}prDaJ=K3(^ zSllOe0ZON7LGO!6G=xSvx5h&B{r&vN%4qO*fBvumfn3}d8J?Ek=Vscrbo2VHp!`l* zbg@C>hab)p-&yTFzP^D>zG&YLl9O8;*p-{nKj!k~4`evj_w-3$ozP$5CBDrh}uaP6gMrPKXOJR z2U!_Iecsov9T!7G<;aI^%1_hg5)mF%Xd|krr58`ToSFHJ&Z_#293HchKI!a0;y9YC zNs&4WgmNI*;k%DWM&^)p?55!;%-@jCJB*c_gGQ%up^n?l3Svy=qT=C%Q=sBU}Ncu*e z^5cR9pU_;YxXh8vu1+4*sJ)L+C+JNd=W+Hn4g|JJ?hyF(X_ oyZ!&6o`3Q=um4uJ=Ig|tuQz=v9 - Click for explanation + Click to show code ```{r, eval = FALSE} condom <- read.csv("toradata.csv", stringsAsFactors = TRUE) @@ -64,7 +65,7 @@ Run a CFA for these three latent variables. - Is it reasonable to proceed with our evaluation of the TORA theory?
      - Click for explanation + Click to show code ```{r} library(lavaan) @@ -80,6 +81,9 @@ fit <- cfa(mod_cfa, data = condom) summary(fit, fit.measures = TRUE) ``` +
      + Click for explanation + - Yes, the model fits the data well, and the measurement parameters (e.g., factor loadings, residual variances) look reasonable. So, the data seem to support this measurement structure. @@ -90,7 +94,8 @@ measurement structure. - The TORA doesn't actually say anything about the associations between these three factors, but it makes sense that they would be positively associated. So, we should find this result comforting. - + +
      --- @@ -108,7 +113,7 @@ Estimate the basic TORA model as an SEM. - How much variance in *intention* and *condom use* are explained by the model?
      - Click for explanation + Click to show code ```{r} mod <- ' @@ -130,6 +135,9 @@ summary(fit, fit.measures = TRUE, rsquare = TRUE) tmp <- inspect(fit, "r2") ``` +
      + Click for explanation + - Yes, the model still fits the data very well. - Yes, the estimates all align with the TORA. Specifically, *attitudes* and *norms* both significantly predict *intention*, and *intention* significantly @@ -138,6 +146,7 @@ predicts *condom use*. *intention* and `r (100 * tmp[["behavior"]]) %>% round(2)`% of the variance in *condom use*. +
      --- @@ -155,7 +164,7 @@ Update your model to represent the extended TORA model that includes - How much variance in *intention* and *condom use* are explained by the model?
      - Click for explanation + Click to show code ```{r} mod_tora <- ' @@ -176,6 +185,9 @@ summary(fit_tora, fit.measures = TRUE, rsquare = TRUE) tmp <- inspect(fit_tora, "r2") ``` +
      + Click for explanation + - Yes, the model still fits the data very well. - Yes, the estimates all align with the updated TORA. Specifically, *attitudes* and *norms* both significantly predict *intention*, while *intention* and @@ -184,6 +196,7 @@ and *norms* both significantly predict *intention*, while *intention* and *intention* and `r (100 * tmp[["behavior"]]) %>% round(2)`% of the variance in *condom use*. +
      --- @@ -291,11 +304,14 @@ Use some statistical means of evaluating the most plausible way to include *Hint:* There is more than one way to approach this problem. +--- + +**Approach 1: Testing Effects** +
      - Click for explanation + Click to show code -We can tackle this problem in a few different ways. One possibility entails -testing the indirect, direct, and total effects. +One way to tackle this problem is to test the indirect, direct, and total effects. ```{r boot2, cache = TRUE} ## Allow for partial mediation: @@ -317,12 +333,23 @@ fit1 <- sem(mod1, data = condom, se = "bootstrap", bootstrap = 1000) summary(fit1, ci = TRUE) ``` +
      + Click for explanation + From the above results, we can see that the direct and total effects are both significant, but the indirect effect is not. Hence, it probably makes the most sense to include *control* via a direct (non-mediated) effect on *behavior*. +
      +
      + --- +**Approach 2.1: Nested Model Comparison** + +
      + Click to show code + We can also approach this problem from a model-comparison perspective. We can fit models that encode each pattern of constraints and check which one best represents the data. @@ -361,6 +388,9 @@ anova(fit1, fit2) anova(fit1, fit3) ``` +
      + Click for explanation + The above $\Delta \chi^2$ tests tell us that the full mediation model fits significantly worse than the partial mediation model. Hence, forcing full mediation by fixing the direct effect to zero is an unreasonable restraint. The @@ -371,21 +401,19 @@ association represents the data just as well as a model that allows for both indirect and direct effects. Hence, we should prefer the more parsimonious total effects model. +
      +
      + --- -While the two tests above lead us to prefer the non-mediated model, we cannot -directly say that the the complete mediation model fits significantly worse than -the non-mediated model. We have not directly compared those two models, and we -cannot do so with the $\Delta \chi^2$. We cannot do such a test because these two -models are not nested: we must both add and remove a path to get from one model -specification to the other. Also, both models have the same degrees of freedom, -so we cannot define a sampling distribution against which we would compare the -$\Delta \chi^2$, anyway. +**Approach 2.2: Non-Nested Model Comparison** -We are not completely without options, though. We can use *information criteria* -to compare non-nested models. The two most popular information criteria are the -*Akaike's Information Criterion* (AIC) and the *Bayesian Information Criterion* -(BIC). +
      + Click to show code + +We can also use *information criteria* to compare our models. The two most +popular information criteria are the *Akaike's Information Criterion* (AIC) and +the *Bayesian Information Criterion* (BIC). ```{r} ## Which model is the most parsimonious representation of the data? @@ -393,13 +421,36 @@ AIC(fit1, fit2, fit3) BIC(fit1, fit2, fit3) ``` +
      + Click for explanation + +While the effect tests and the nested model comparisons both lead us to prefer +the non-mediated model, we cannot directly say that the complete mediation +model fits significantly worse than the non-mediated model. We have not directly +compared those two models, and we cannot do so with the $\Delta \chi^2$. We +cannot do such a test because these two models are not nested: we must both add +and remove a path to get from one model specification to the other. Also, both +models have the same degrees of freedom, so we cannot define a sampling +distribution against which we would compare the $\Delta \chi^2$, anyway. + +We can use information criteria to get around this problem, though. Information +criteria can be used to compare both nested and non-nested models. These criteria +are designed to rank models by balancing their fit to the data and their complexity. When comparing models based on information criteria, a lower value indicates a better model in the sense of a better balance of fit and parsimony. The above results show that both the AIC and the BIC agree that the no-mediation model is the best. +
      +
      + --- +**Conclusion** + +
      + Click for explanation + So, in the end, regardless of how we approach the question, all of our results suggest modeling *perceived behavioral control* as a direct, non-mediated predictor of *condom* use.