Extreme risk events such as the financial crisis, trade frictions, and public health security have posed huge challenges to the stable development of global financial markets. With the proliferation and spread of the COVID-19 epidemic, international commodity prices fluctuate sharply, and the Chinese financial market is facing great risk shocks. However, previous literatures on risk contagion in financial markets ignore the perspective of extreme risk events. Few scholars measure the risk contagion effect between the international commodity market and the Chinese financial market. At the same time, they cannot trace the source of external risks.

This paper uses the DCC-GARCH model and the Granger-Geweke causality test to measure the risk contagion effect between the international commodity market and the Chinese financial market, and traces the external risk sources. It is found that: First, during the period of extreme risk events, the co-movement effect between the international commodity market and the Chinese financial market fluctuates sharply. During the period of the COVID-19 epidemic, the linkage between the international commodity market and the Chinese financial market has increased significantly. Second, under the shock of extreme risk events, the risk contagion effect between the international commodity market and the Chinese financial market has increased. During the period of the COVID-19 epidemic, the interactive shock between the international commodity market and the Chinese financial market is asymmetrical. The impact of Chinese financial market on the international commodity market is very limited. Third, tracing the source of external risks in the Chinese financial market shows that energy, precious metals, and industrial metals have a higher impact on the Chinese financial market than other commodities, and the Chinese stock market and the exchange rate market bear stronger external shocks. This paper provides theoretical support and policy reference for China to prevent international financial risk contagion.

The academic value of this paper could be concluded in three aspects: First, during the period of extreme risk events, we measure the general probability distribution of the dynamic condition correlation coefficient between the international commodity market and the Chinese financial market, and identify the general volatility trend of the linkage between the international commodity market and the Chinese financial market. Second, we build the dynamic causal network to measure the general and time-varying characteristics of risk contagion between the international commodity market and the Chinese financial market during the period of extreme risk events. Third, we decompose the external shock of the Chinese financial market, and trace the external risks of the Chinese financial market accurately.