I thought it might be of interest to the community to share that PyMC3 was used in our paper, Inferring the effectiveness of government interventions against COVID-19, which was just published in Science. You can check the paper out here.
Disclaimer: I’m one of the authors
A short summary: we build a hierarchical Bayesian model that links the timing of interventions to observed cases and deaths (in the first wave) in countries to infer the effectiveness of different interventions against COVID-19. We find that limiting gatherings to fewer than 10 people and closing educational institutions was highly effective.
We’d also like to thank the PyMC3 community for their help - I regularly posted questions here over the past few months.