Volatility forecasting of exchange rate by quantile regression

Exchange rates are known to have irregular return patterns; not only their return volatilities but the distribution functions themselves vary with time. Quantile regression allows one to predict the volatility of time series without assuming an explicit form for the underlying distribution. This study presents an approach to exchange rate volatility forecasting by quantile regression utilizing a uniformly spaced series of estimated quantiles. Based on empirical evidence of nine exchange rate series, using 19 years of daily data, the adopted approach generally produces more reliable volatility forecasts than other key methods.