Simplified identification of the epidemic process model

Abstract

The SIR model currently remains the basic model of the epidemic process. This model and its numerous generalizations (SEIR, SEIRD, etc.), as well as some similar models, belong to the class of compartmental models, that describe the dynamics of the epidemic process as the interaction of “compartments”, i. e., various groups into which, according to their role in transmission infections (susceptible, infected, recovered, etc.) break up the entire population of people among whom the epidemic is spreading. Identifying the parameters of these models is often a challenging task. This task is somewhat simplified when conducting retrospective analyses of epidemics, where full incidence data are available at all stages of the process. We propose a simplified method for identifying a stationary SIR model using a minimal set of observed data that allows for relatively reliable determination. Based on the proposed methodology, the inapplicability of the stationary SIR model to describe the first wave of the COVID-19 pandemic in Germany and Italy is shown.