Temporal Prediction Model of the Evolution of Confirmed Cases of the New Coronavirus (SARS-CoV-2) in Brazil
Keywords:
Data analysis, Forecasting, Mathematical model, Nonlinear equationsAbstract
The present article presents a model of temporal prediction of the evolution of the number of confirmed cases of the new coronavirus (SARS-CoV-2) in Brazil. This model is based on the analysis of the data source of the Our World in Data platform. The data analysis technique used was Forecasting (nonlinear regression) implemented in the Python programming language. The analysis ranged from 03/26/2020 (100th confirmed case of COVID-19 in Brazil) until 05/15/2020. This model was validated, using the cross-validation technique, from a comparison of the prediction scenarios and the official numbers released by the Brazilian Health Ministry. The results showed that the proposed model presented a good accuracy for short-term forecasts (7 days).