MCMC for sampling the posterior distribution in the BETS model for Wuhan-exported cases of COVID-19

These scripts implement random walk Metropolis-Hastings sampling for the prior and posterior distribution in the models described in Section 5 of:

Qingyuan Zhao, Nianqiao Ju, and Sergio Bacallado. "BETS: The dangers of selection bias in early analyses of the coronavirus epidemic", 2020. arXiv:2004.07743.

For script usage see, for example:

$ python SamplePrior.py -h

Requirements

The code was tested on:

• Python 3.5
• Tensorflow 2.1.0
• Tensorflow-probability 0.9.0