Francisco Morillas (Departament d’Economia Aplicada, Facultat d’Economia, Universitat de València) and José Valero (Centro de Investigación Operativa, Universidad Miguel Hernandez de Elche)
Abstract: In this paper, we apply resampling techniques to a modified compartmental SEIR model which takes into account the existence of undetected infected people in an epidemic. In particular, we implement numerical simulations for the evolution of the first wave of the COVID-19 pandemic in Spain in 2020. We show, by using suitable measures of goodness, that the point estimates obtained by the bootstrap samples improve the ones of the original data. For example, the relative error of detected currently infected people is equal to 0.061 for the initial estimates, while it is reduced to 0.0538 for the mean over all bootstrap estimated series.