A numerical simulation of the COVID-19 epidemic in Argentina using the SEIR model

Santos, Juan E.; Carcione, Jose' M.; Savioli, Gabriela B.; Gauzellino, Patricia M.; Ravecca, Alejandro; Moras, Alfredo

A pandemic caused by a new coronavirus has spread worldwide, causing an epidemic in Argentina. We implement an SEIR model to analyze the evolution of the disease in Buenos Aires and neighbouring cities (RMBA) with 15 million inhabitants. The parameters of the model are calibrated by using as data the number of casualties officially reported. Since infinite solutions honour the data, we show a set of cases by considering different situations. The first set of parameters yields initially a reproduction ratio R0 = 3.33 decreasing to 0.95 in April 8, after the lockdown, but increasing to 1.55 after April 27, most probably due to an increase of the contagion in highly populated slums. The infection fatality rate (IFR) is 1.88 % and the predicted number of casualties is 173000 deaths with 9 million people infected at the end of the epidemic. However, keeping $R_0$ = 0.95 after April 27, would cause only 1881 casualties and 92955 infected individuals. Other cases, assuming the present trend, predict smaller incubation periods (4-5 days) and yield between 20000 and 70000 deaths and IFRs between 0.5 % and 1.1 %. We also consider doubling the number of casualties, with a death toll of 44000 individuals and 5.1 million infected individuals. Other choices of parameters also provide a good fit of the data, indicating the uncertainty of the results, which may differ from reported values. The analysis allows us to study how isolation and social distancing measures affect the time evolution of the epidemic.