Predictive Modeling of COVID- 19 Confirmed Cases Using Regressive Objective Regression Methodology

Abstract

The use of predictive models for the evolution of the pandemic is of great help in decision-making by the authorities. The fundamental objective of this work was to obtain through the Regressive Objective Regression, predictions of confirmed cases of COVID-19 in the Marta Abreu Teaching Polyclinic of the city of Santa Clara. In short-term modeling the model was significant at 19.7% with an error of 0.12. Variables dichotommics, saw tooth and saw tooth inverted and risk returned in 1.3, and 12 cases the trend is negative and not significant. We can conclude that a perfect result was obtained in the long term with the ROR methodology. The short-term ROR model depends on the cases of COVID-19 in the previous case, 3 cases back and 12 cases back without significant trend. The long-term model is perfect and depends on the cases of COVID-19 in 12 cases ago, with a negative trend.