Utilização de regressão logística simples na verificação da qualidade do ar atmosférico de Uberlândia

Abstract

The objective of this paper was to analyze the air quality in Uberlândia, in Minas Gerais, by simple logistic regression models. The PM10 sampler was used to obtain data of the pollution that considers only concentration of suspended particles smaller than 10 μm. The temperature, precipitation, relative humidity, wind speed, day of the week and seasons of year were used as predictions variables in this study. The day of week and seasons of year were used as binary system and as predictions variables too. The results showed that there is a significant relation among the temporal and climatic variables with the air quality of Uberlândia. The simple logistic models show that precipitation, humidity, summer, spring, Saturday, and Sunday significantly contribute to achieve good air quality, while the winter and the increasing flow of vehicles tend to worsen the air quality.