Description:
<jats:p xml:lang="fr"><abstract><p>In this paper, we consider a compartmental model to fit the real data of confirmed active cases with COVID-19 in Portugal, from March 2, 2020 until September 10, 2021 in the Primary Care Cluster in Aveiro region, ACES BV, reported to the Public Health Unit. The model includes a deterministic component based on ordinary differential equations and a stochastic component based on bootstrap methods in regression. The main goal of this work is to take into account the variability underlying the data set and analyse the estimation accuracy of the model using a residual bootstrapped approach in order to compute confidence intervals for the prediction of COVID-19 confirmed active cases. All numerical simulations are performed in R environment (<inline-formula id="math-09-02-136-M1"><inline-graphic xmlns:xlink="http://www.w3.org/1999/xlink" xlink:href="math-09-02-136-M1.jpg"/></inline-formula> version. 4.0.5). The proposed algorithm can be used, after a suitable adaptation, in other communicable diseases and outbreaks.</p></abstract></jats:p>