Piecewise estimation of R0 by a simple SEIR model. Application to COVID-19 in French regions and departments until June 30, 2020

The estimation of R0 , the so-called "basic reproductive ratio", of the COVID-19 pandemic is of particular importance to help decision-makers take the necessary safeguard measures to protect the population. In this work, we examine a method based on the successive estimation of R0 over 3 non-overlapping periods (before lockdown, during lockdown and after). The approach is based on a variant of the, simple but flexible , SEIR compartmental model that allows to exploit the number of recovered individuals that are reported in the daily database published by national health agencies. The results of the approach is analysed w.r.t. data from France, at two levels of geographical subdivisions, i.e. the 13 regions and 96 departments that make up the metropolitan territory.