Evaluation of the potential incidence of COVID-19 and effectiveness of contention measures in Spain: a data-driven approach

Alberto Aleta; Yamir Moreno

Stochastic SEIR-metapopulation models are routinely used to study the temporal and spatial transmission of diseases like the COVID-19. Here, we made use of such class of models and implement a data-driven version that allows to obtain realistic estimates for the spatial incidence of the disease as well as its temporal dynamics. More specifically, in terms of time, we feed the model with the available data as of February 28th, 2020. Spatially, we consider that each province (there are 52 in Spain, see appendix B) is represented by a subpopulation. Furthermore, metapopulation models are composed by two types of dynamics: the disease dynamics governed by the chosen compartmental model, SEIR in our case, and the mobility of the individuals across the subpopulations that make up the whole metapopulation system. The latter ingredient, the mobility, connects the subpopulations and allows the disease to spread from one subpopulation to another. In what follows, we describe these two components of our model.