PREVISIONE DEL TASSO DI CAMBIO DOLLARO/EURO MEDIANTE LA TAYLOR RULE

The analysis is on the Dollar/Euro exchange rate forecasting based on an augmented Taylor Rule, the predictive content is measured by out-of-sample measures. Rolling regression is employed to forecast exchange rate from January 1999 until July 2018, for the regression 24 and 36 observation are used, starting from January 1997 and keeping the number of observations constant, one period ahead forecast is implemented. The empirical result of the forecast is univocal: from one side the prediction error made by the estimated model is statistically lower than the one displayed by the random walk model (the benchmark model), from the other side the direction of the exchange rate predicted by the employed model is correctly forecasted a number of time that is statistically significant for the test implemented.