Tamanho da fonte: 

In this paper we use a set high-dimensional and univariated models to forecast the Brazilian inflation. The models are the simple AR, the factor model and LASSO type models. Our results showed that the LASSO has the smallest errors for small horizon forecasts. However, for long horizons the AR is the best model; the factor model also produces some good long horizon forecasts in a few cases. We tested all the models for the two most important Brazilian inflation measures, the IPCA and the IGP-M. The results also showed that there are differences on the selected variables for both measures. Finally, the most important variables selected by the LASSO models are, in general, related to government debt and money. Variables such as unemployment and production were rarely selected by LASSO models.