generated from rlevy/9.19-syllabus
-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathclass_sessions.yml
77 lines (61 loc) · 7.46 KB
/
class_sessions.yml
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
lectures:
Introduction:
topic: Introduction to the class. Brief review of elementary probability. Brief introduction to R.
slides: "Class intro slides ([no builds](assets/slides/2024-04-02-keynote-intro-slides-no-builds.pdf); [with builds](assets/slides/2024-04-02-keynote-intro-slides-with-builds.pdf)); [Intro probability slides](assets/slides/2024-04-02-intro-probability-slides.pdf)"
related readings: "[Probabilistic models in the study of language, chapters 2–5](https://www.mit.edu/~rplevy/pmsl_textbook/text.html)"
Introduction to causal models:
topic: "Elementary statistics: maximum-likelihood estimation of a Bernoulli model. Introduction to causal inference: the potential-outcomes framework."
readings: "[Causal inference in R, Chapters 1–6](https://www.r-causal.org/)"
related readings: "[Hernan & Robins 2024, Chapters 1–3](https://www.hsph.harvard.edu/miguel-hernan/causal-inference-book/)"
slides: "Introduction to causal inference ([no builds](assets/slides/2024-04-04-intro-to-causal-inference-no-builds.pdf); [with builds](assets/slides/2024-04-04-intro-to-causal-inference-with-builds.pdf))"
Causal models continued:
topic: "Causal inference continued: more on the potential-outcomes framework, and causal graphical models."
readings: "Pearl 2009, Chapter 1 (available on Canvas and Piazza)"
slides: "Potential outcomes framework ([no builds](assets/slides/2024-04-09-more-causal-inference-no-builds.pdf); [with builds](assets/slides/2024-04-09-more-causal-inference-with-builds.pdf)); Causal Bayes Nets ([no builds](assets/slides/2024-04-09-causal-Bayes-nets-no-builds.pdf); [with builds](2024-04-09-causal-Bayes-nets-with-builds.pdf))"
Finish causal models plus elementary parameter estimation:
topic: "More on causal graphical models. More on parameter estimation: frequentist and Bayesian methods."
slides: "Causal Bayes nets ([no builds](assets/slides/2024-04-11-causal-Bayes-nets-no-builds.pdf); [with builds](assets/slides/2024-04-11-causal-Bayes-nets-with-builds.pdf)); Parameter estimation ([no builds](assets/slides/2024-04-11-parameter-estimation-no-builds.pdf); [with builds](assets/slides/2024-04-11-parameter-estimation-with-builds.pdf))"
readings: "[Levy in progress](https://www.mit.edu/~rplevy/pmsl_textbook/text.html), Section 4.4, Chapter 5, Sections 6.1–6.5 and 6.7–6.8"
related readings: "[Nicenboim et al., 2024](https://vasishth.github.io/bayescogsci/book/), Chapter 10"
Roger away:
topic: "**Roger away, no class**"
Confidence intervals, hypothesis testing, Monte Carlo:
topic: "Confidence intervals & hypothesis testing (frequentist & Bayesian). Monte Carlo methods."
readings: "[Levy in progress](https://www.mit.edu/~rplevy/pmsl_textbook/text.html), Chapter 5; Kruschke 2015, Chapter 14 (available on Canvas and Piazza)"
related readings: "[Carpenter et al., 2017](http://stat.columbia.edu/~gelman/research/unpublished/stan-paper-revision-feb2015.pdf)"
slides: "[Confidence intervals, hypothesis testing, and Monte Carlo](assets/slides/2024-04-16-confidence-intervals-hypothesis-testing-Monte-Carlo-with-builds.pdf); [Generalized Linear Models](assets/slides/2024-04-16-generalized-linear-models.pdf)"
MCMC and intro to GLMs:
topic: "Markov-chain Monte Carlo. Introduction to probabilistic programming using Stan. Begin quick review of Generalized Linear Models (GLMs)."
readings: "[Simmons et al., 2011](https://journals.sagepub.com/doi/full/10.1177/0956797611417632)"
slides: "[Confidence intervals, hypothesis testing, and Monte Carlo](assets/slides/2024-04-16-confidence-intervals-hypothesis-testing-Monte-Carlo-with-builds.pdf)"
related readings: "[Agresti, 2015](https://www.google.com/books/edition/_/dgIzBgAAQBAJ?hl=en&gbpv=1&pg=PR11&dq=generalized+linear+models+ agresti), Chapter 1; [Gelman et al., 2020](https://users.aalto.fi/~ave/ROS.pdf), Chapter 4 and Parts 2 & 3"
Linear regression:
topic: "Linear regression: parameter estimation, confidence regions, *t* statistics, credit assignment problems, decomposition of variance, nested model comparison with the *F* test."
readings: "Gelman & Hill 2007, chapters 3 & 4 (on Canvas)"
slides: "[Generalized linear models: linear regression](assets/slides/2024-04-23-generalized-linear-models.pdf)"
related readings: "[Nicenboim et al., 2024](https://vasishth.github.io/bayescogsci/book/), Chapter 4"
GLMs, effect size, power analysis, preregistration, repeated measures:
topic: "Model class, contrasts, reparameterization. Logistic regression. Effect size. Power analysis and preregistration. The problem of repeated measures."
readings: "Gelman & Hill 2007, chapter 5 (on Canvas); Kruschke 2015, chapter 13 (on Canvas)"
related readings: "[Gelman & Karlin, 2014](https://journals.sagepub.com/doi/full/10.1177/1745691614551642); [Simmons et al., 2011](https://journals.sagepub.com/doi/full/10.1177/0956797611417632); [Wagenmakers et al., 2012](https://doi.org/10.1177/1745691612463078)"
slides: "[Categorical predictors, interactions, and logistic regression](assets/slides/2024-04-25-categorical-predictors-interactions-logistic-regression.pdf)"
Mixed models I:
topic: "Hierarchical/mixed-effects/multi-level regression modeling I: Fundamental concepts and formalization of mixed-effects models. Maximum likelihood based methods. Random effects structure and Keeping It Maximal. "
readings: "Gelman & Hill 2007, chapters 11 & 12 (on Canvas); [Barr et al., 2013](https://www.sciencedirect.com/science/article/pii/S0749596X12001180)"
slides: "[Hierarchical/mixed-effects/multi-level models](assets/slides/2024-04-30-mixed-effects-models.pdf)"
Mixed models II:
topic: "Hierarchical/mixed-effects/multi-level regression modeling II: Bayesian methods. When and how much to worry about wrong modeling assumptions, and what to do about it."
readings: "Gelman & Hill 2007, chapters 13 & 14 (on Canvas)"
slides: "[R script for getting started with interactive practicum](assets/resources/practical_issues_with_mixed_models/getting_started.R); [Completed practicum R script with comments](assets/resources/practical_issues_with_mixed_models/practicum-complete.R)"
Mixed models III:
topic: "Hierarchical/mixed-effects/multi-level regression modeling III: Bayesian methods. Post-hoc testing using a fitted model. Reporting results from a mixed effects model."
slides: "[R script for getting started with interactive practicum](assets/resources/practical_issues_with_mixed_models_II/practicum-2024-05-07-getting-started.Rmd)"
Mixed models IV:
topic: "Hierarchical/mixed-effects/multi-level regression modeling IV, and end of class"
slides: "[Practicum script 1](assets/resources/practical_issues_with_mixed_models_III/practicum_2024_05_09-getting-started.R); [Practicum script 2](assets/resources/practical_issues_with_mixed_models_III/2024-05-09-by-subjects-analysis-getting-started.R)"
Missing and observational data:
topic: Dealing with missing data. Dealing with observational data. Instrumental variables. Identifying good "natural experiments".
High dimensional models:
topic: "High-dimensional predictive models. Dimensionality reduction. Regularization. Case study: comparing artificial neural network and brain activation patterns through encoding or decoding models."
More multivariate analysis:
topic: "More multivariate analysis techniques. Representational (dis)similarity analysis. Applications to comparing artificial neural network and brain activation patterns."