Which uncertainty is powerful to forecast crude oil market volatility? New evidence

Citations Over TimeTop 1% of 2020 papers

Abstract

Abstract This paper focuses on distinguishing the predictive power of five newly developed uncertainty indices, that is, global and US economic policy uncertainty (GEPU and US EPU), global geopolitical risk (GPR), US monetary policy uncertainty (MPU) and equity market volatility (EMV), on crude oil market volatility. We construct the extended GARCH‐MIDAS models to assess the forecasting ability of these indices and then compare them with four commonly used oil volatility drivers: oil supply, oil demand, oil speculation and interest rate. Firstly, the in‐sample analysis suggests that all uncertainty indices and traditional determinants have significant impacts on oil volatility. Secondly, the out‐of‐sample evaluations suggest that US MPU, EMV and EPU indices are more powerful than other predictors to improve the forecasting accuracy of crude oil volatility, and US EMV exhibits the best prediction ability. Then, we find that EPU and MPU indices are more effective in forecasting high volatility in crude oil market, while traditional determinants together with US EMV index are more helpful in predicting low oil volatility. Finally, the forecasts of bad oil volatility can be improved by all predictors except for GEPU index, while the forecasts of good oil volatility can only be enhanced by US MPU, EMV and EPU.

Related Papers

• → A comparison of implied and realized volatility in the Nordic power forward market(2015)32 cited
• → Forecasting realized volatility: The role of implied volatility, leverage effect, overnight returns, and volatility of realized volatility(2021)7 cited
• Model-free implied volatility and its information content:Evidence from Hang Seng Index options(2009)
• → Volatility Forecasting and Volatility Risk Premium(2015)1 cited
• Forecasting Volatility in Options Trading - Nexus between Historical Volatility and Implied Volatility(2016)