The role of economic uncertainty in forecasting exchange rate returns and realized volatility: Evidence from quantile predictive regressions

Abstract

In this paper, we investigate whether the news-based measure of economic policy uncertainty (EPU) can be used to forecast exchange rate returns and volatility using a quantile regression approach, which accounts for persistence and endogeneity, using data from 13 different countries. Our main findings suggest that: (i) EPU is useful for forecasting exchange rate returns and volatility, (ii) forecasting ability–quantile order relationships exhibit a U-shape, possibly asymmetric form around the median

and (iii) asymmetries are more pronounced in the case of forecasting volatility.