Estudo do isolamento social no período pandêmico em São Paulo

Abstract

The present work aims to perform a descriptive and exploratory analysis of social isolation data during the pandemic period in the state of São Paulo. The descriptive and exploratory analysis aims to assist and optimize the subsequent modeling process, standing out as crucial steps in the data analysis flow. By analyzing the graphs, it was observed that the model presented an excellent fit to the data, capturing the growth trend at the beginning of the pandemic. It was concluded that the time series exhibits a strong seasonality pattern, occurring weekly. The SARIMA model suggested a low-order autoregressive component, a moving average component that includes both a short-term term and a weekly seasonal term, and a random error component. This model proved to be adequate and well-fitted to the data, offering valuable contributions to understanding the behavior of social isolation during the pandemic.