In this paper, a framework of analyzing risk spillovers between stock sectors is proposed based on the high-frequency data sampled for 5 min. We calculate a weighted directed network through the TVP-VAR model to analyze the time-varying risk spillovers between stock sectors with “different sources and different order moments”. The results show jumps in the realized volatility of five stock sectors during risk events. The static total volatility spillover index of China's stock sector exceeds 70%, which indicates that the risk of Chinese stock sector is high. In addition, volatility spillovers between sectors are asymmetric and time-varying, with “bad” volatility spillovers dominating. After 2019, the sensitivity difference of the overall stock sector to negative news and positive news shocks has a decreasing trend. The sector spillovers of the volatility, skewness and kurtosis are strengthened in the recession period and weakened in the boom period of stock market. Both the industry and utility sectors—particularly the industry sector—are the main net contributors of risk spillovers of volatility, skewness and kurtosis, while the real estate sector acts as the net receiver of risk spillovers. In the future, it is necessary to be alert to risk spillovers from the integrated sector.
Read full abstract