kevdur/covid

Coronavirus reproduction numbers in South Africa # COVID-19 Reproduction Numbers in South Africa _See the write-up for full details._ The goal of this project was to derive and implement a relatively simple Bayesian model that can be used to estimate effective reproduction numbers for an epidemic, with a particular focus on the COVID-19 epidemic in South African provinces. The work was inspired by that of Systrom and Bettencourt and Ribeiro, but tends towards the more analytic aspects of the Bayesian approach where possible. In addition to the write-up linked to above, a very brief overview of the model and its solution are given in the summary note. ## Notebooks The source code should be relatively self-explanatory. An example of how to use it to create plots is given in the first notebook. ## Command-line tasks Run `make` to see a list of possible command-line operations. The two of real interest are `make update`, which installs the required Python libraries, and `make data`, which fetches updated versions of the data sets.