Skip to content

Uses Markov chain Monte Carlo to estimate the parameters of an SIR model with COVID-19 data

Notifications You must be signed in to change notification settings

deephaven/covid-19-sir-modelling

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 

Repository files navigation

This program uses the most recent available data from The Covid Tracking Project to estimate the transmission rate, removal rate, and mortality rate of COVID-19 in each state, using a model called the SIR model.

This estimation is done via Markov chain Monte Carlo sampling through a Python package called PyMC3. We utilize Deephaven through the Jupyter environment so that we can take advantage of PyMC3's useful plots, as well as other Python plotting packages including Descartes.

For some of the states, the MCMC process is not successful, and which ones fail will change everytime your data is updated. Some, such as Colorado, seem to fail consistently. There are a variety of reasons MCMC might fail, including poor model fit or numerical instability.

The final plots demonstrate metrics such as overall COVID-19 risk by state, and provide the user the ability to come to their own conclusions about the spread of COVID-19 in the US. Additionally, the probabilistic distributions given by MCMC enable users to answer questions about the spread and mortality rates of COVID-19 in a probabilistic way, which is often more useful than a point estimate.

Releases

No releases published

Packages

No packages published