Skip to content

Latest commit

 

History

History
42 lines (35 loc) · 1.57 KB

README.md

File metadata and controls

42 lines (35 loc) · 1.57 KB

FHM_replication

Reproducing FHMs "skattning av det momentana reproduktionstalet"

About

Estimating the reproduction number of an active pandemic can be tough. However, with simple available tools one can produce estimates that potentially aren't that far from the truth.

On this problem, as for active pandemics, formating the data and tune the models is usually the problem. In this short code sample, I aim to show how one can simply reproduce number similar to what the Public Health Agency Of Sweden has publiced here https://www.folkhalsomyndigheten.se/smittskydd-beredskap/utbrott/aktuella-utbrott/covid-19/statistik-och-analyser/analys-och-prognoser/

CODE

In short, this is the code needed to replicate their results.

config <- make_config(list(mean_si = 4.8,
                           std_si = 2.3))
method <- "parametric_si"
res <- estimate_R(df$incidence,
                  method=method,
                  config = config)

Which will give you FHM replicate for the region of Uppsala.

See the full code under /R/ and the data is found in /data/ the figures produced can be found under /fig/.

Suggestions

The noisy incidence creates poor estimates, what we can do to address this we can use (all are for Uppsala, one can change region in the code)

  • rolling mean on the data Rolling mean
  • uncertainty in the gamma distribution. Uncertainty in parameters
  • both of the above rolling, and uncertainty in parameters