Title: Modeling and estimation for COVID-19 under-reported counts

Abstract:

The officially reported COVID-19 daily counts of infected, recovered, and perished people are grossly underestimated. It is widely known that only a portion of infected individuals receives professional coronavirus testing. Even those who have been tested, confirmed, and then recovered, not all recoveries are reported. Furthermore, the proportion of unobserved and under-observed counts varies by territory and changes in time, because of different and changing diagnostics and reporting standards.

We propose and develop a stochastic model that includes untested individuals and unobserved COVID-19 recoveries and casualties. The model that we call SICROUD (Susceptible – Infected – Confirmed – Recovered – Observed – Unobserved – Died) generalizes the classical SIR epidemic model, extending it with additional compartments. Its main parameters are the infection rate, the testing rate, the recovery rate, the mortality rate, and the reporting rate, which may vary continuously in time. The proposed Bayesian algorithm uses observed counts to estimate the model parameters and unobserved counts dynamically, updating the estimates with new data on daily basis.

The algorithm is applied to evaluate the trends, produce forecasts, and compare epidemic situation in different world countries and US states.