Forecasting China's crude oil futures volatility: How to dig out the information of other energy futures volatilities?

Abstract

This paper explores the predictability of China's crude oil futures volatility by considering other energy futures volatilities. Empirical results show that other energy futures volatilities can provide useful information for forecasting crude oil futures volatility both in- and out-of-sample. To further dig out predictive information from other energy futures volatilities, we employ forecast combinations and shrinkage methods. Corresponding results suggest that both forecast combinations and shrinkage methods make full use of information from other energy futures volatilities and generate more accurate forecasts of crude oil futures volatility. Furthermore, shrinkage methods have better forecasting performance than forecast combinations. Finally, the superior performance of shrinkage methods stems from their ability to accurately select other energy futures volatilities with strong predictive power.