Analysis of Case Fatality Rate Related to Coronavirus Disease 2019



In this article, we provide a basic idea to classify the Case Fatality Rate (CFR) of Coronavirus Disease 2019 (COVID-19) in terms of region. We propose two different Bayesian mixture models. The first model assumes the low CFR group following truncated normal distribution and the high CFR group following normal distribution, whereas the second model assumes Gamma distribution and Beta distribution respectively. We implement the mixture model by the application of EM algorithm. The simulation results shows that the truncated normal-normal distribution mixture model has a greater predictive power and the results is robust to all level of CFR; and the Gamma-Beta mixture model can only accurately classify the extreme cases of CFR level.

Final Report