search

SABS-R3-Epidemiology / branchpro

Using branching processes to estimate the time-dependent reproduction number of a disease with imported cases
BSD 3-Clause "New" or "Revised" License
4 stars 1 forks source link

Compare negative binomial to poisson noise #224

Closed I-Bouros closed 1 year ago

I-Bouros commented 1 year ago

In this pull request we have implemented equivalent classes for the branching processes in the original project, with the exception that these new classes allow for negative binomial-distributed noise.

We also include a python notebook in which we compare the effects of different values of the overdispersion parameter associated with the negative binomial noise in the data on the inferred $R_t$ trajectory (inferred under the assumption that the data is poisson, and not negative-binomial).

Closes #222

codecov[bot] commented 1 year ago

Codecov Report

Merging #224 (2f10b91) into main (0ce1498) will not change coverage. The diff coverage is 100.00%.

@@            Coverage Diff            @@
##              main      #224   +/-   ##
=========================================
  Coverage   100.00%   100.00%           
=========================================
  Files           11        11           
  Lines          762       828   +66     
=========================================
+ Hits           762       828   +66
Impacted Files Coverage Δ
branchpro/new_models.py 100.00% <100.00%> (ø)
branchpro/__init__.py

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

I-Bouros commented 1 year ago

Plus do you know why the doc tests are failing

Probably I haven't done the docs