The Usefulness of factor models in forecasting the exchange rate: results from the Brazilian case

Abstract

This paper studies the usefulness of factors embedded on the common movements of exchange rates in forecasting the exchange rate Real/Dollar. The results show that considering the entire period of the sample from January 1999 to August 2011, no one model containing the factors is able to beat a random walk model. However, when the period directly following the adoption of the floating exchange rate regime is discarded, there is evidence that several models containing these factors beat the random walk. Lastly, the paper shows that the addition of factors improves the predictive power of the models comprising only macroeconomic variables commonly used in the literature to forecast the exchange rate.