Data-driven dynamic SEIR model

Abstract

Throughout the year of 2020, no number has dominated the public media more persistently than the reproduction number. This powerful but simple concept is widely used by the public media, scientists, and political decision makers to explain and justify political strategies to control the COVID-19 pandemic. Here we explore the effectiveness of political interventions using the reproduction number of COVID-19 across Europe. We combine what we have learnt throughout this book, SEIR compartment modeling, dynamic contact rates, computational modeling, and Bayesian analysis to create a data-driven dynamic SEIR model. To illustrate the basic elements of data-driven dynamic SEIR modeling, we discuss the concepts of priors, likelihood, and posteriors and apply Markov Chain Monte Carlo methods to compute them. We draw early case data of COVID-19 and the daily air traffic, driving, walking, and transit mobility for all 27 countries of the European Union, infer the dynamic reproduction number for each country, and correlate mobility and reproduction. The learning objectives of this chapter on data-driven modeling are to