Abstract

SALCEDO, Franco  and  SALCEDO, Galo. Predictive models of COVID-19 infections for the province of Loja - Ecuador. Novasinergia [online]. 2021, vol.4, n.2, pp.62-77.  Epub Dec 01, 2021. ISSN 2631-2654.  https://doi.org/10.37135/ns.01.08.04.

Abstract: The province of Loja, like the rest of the world, has been affected by COVID-19, which put the capacity of health systems and their leaders to the test. Given this scenario, obtaining predictions of the cases of contagion is an important factor in making decisions. To predict the level of infections, the numerical method of divided differences, a simple logistic model, an improved logistic model that includes the proportion of confinement, and the least squares method were used. Data provided by the Ministry of Public Health of Ecuador were used in a period of 399 days from the appearance of the first cases and their processing was carried out with the GNU Octave Software, version: 5.1.0. Logistic models are unsatisfactory due to insufficient information about some factors, such as the rate of infections, recoveries, mobility of infected individuals and their interaction with non-infected individuals, and the proportion of the level of confinement in each political jurisdiction. The method least squares offers better results for predictions, since it does not use rates or proportions and minimizes the quadratic error, that is, it finds the curve that passes between the data points.

Keywords : COVID-19; least squares; logistics model; pandemic; predictive model.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )