Time series models have been used for climatological variables such as temperature and rainfall. In this paper, time series of monthly mean temperatures for the municipality of Uberlândia, Minas Gerais, were analyzed describing their components and making forecasts using the SARIMA models. The analysis showed trend and seasonality components. The SARIMA model was adjusted by AIC (Akaike Information Criterion), BIC (Bayesian Information Criterion) and MSE (Mean Square Error) criterion. The results showed that the SARIMA (3,1,0)(0,1,1) was a good model for forecasting.
regression; SARIMA model; seasonality; trend
