Abstract:
In this paper, we employ the generalized autoregressive conditional heteroscedasticity-mixed data sampling (GARCH-MIDAS) framework to forecast the daily volatility of 19 dollar-based exchange rate returns based on monthly metrics of oil price uncertainty (OPU), and relatively broader global and country-specific energy market-related uncertainty indexes (EUI). We find that the global EUIs tend to perform better than the OPU, highlighting the need to look beyond the oil market to capture energy related uncertainties. The country-specific EUIs outperform the benchmark in a statistically significant manner for at least 14 currencies across the short-, medium-, and long-term forecasting horizons.
