Volatility forecasting of Chinese energy market: Which uncertainty have better performance?

Abstract

This paper examines the effects of seven uncertainty indices on Chinese energy price volatility and analyses the influencing factors using the extended GARCH-MIDAS model. Our in-sample analysis shows that energy market volatility is negatively impacted by global and Chinese economic policy uncertainty (GEPU, CNEPU), the geopolitical risk act index (GPRA), and the climate policy uncertainty index (CPU). Out-of-sample forecasting results demonstrate that the CPU is a major cause of energy volatility in China, and our extended model exhibits higher forecast accuracy. Additionally, we discover that the CPU has a greater capacity for energy volatility during periods of low volatility, while high energy volatility is often associated with GPR. Finally, the outbreak of the Russia-Ukraine conflict resulted in a decline in the predictive capacity of the CPU while causing a boost in the EPU's predictive power.