Forecasting China's crude oil futures volatility: New evidence from the MIDAS-RV model and COVID-19 pandemic

In this study, we focus on the role of jumps and leverage in predicting the realized volatility (RV) of China's crude oil futures. We employ a standard mixed data sampling (MIDAS) modeling framework. First, the in-sample results indicate that the jump and leverage effects are useful in predicting the RV of Chinese crude oil futures. Second, the out-of-sample results suggest that jump has very significant predictive power at the one-day-ahead horizon while the leverage effect contains more useful information for long-term predictions. Moreover, our results are supported by a number of robustness checks. Finally, we find new evidence that the prediction model that considers the leverage effect has the best predictive power during the COVID-19 pandemic.

Automated summary

This study focuses on the role of jumps and leverage in predicting the realized volatility of China's crude oil futures using a standard mixed data sampling (MIDAS) modeling framework. The in-sample results show the usefulness of jump and leverage effects in predicting the RV, with jumps having significant predictive power at the one-day-ahead horizon and leverage containing more useful information for long-term predictions. Additionally, the prediction model considering the leverage effect exhibits the best predictive power during the COVID-19 pandemic, as supported by robustness checks.