Abstract: In recent years, against the backdrop of intertwined international geopolitical conflicts and global economic and trade frictions, commodity markets have increasingly become a focal point for global risk convergence. We first examine the risk linkage between commodity and financial markets using ARMA-GJR-GARCH and Copula-CoVaR models. Subsequently, we employ the expected shortfall (ES) as a tail risk measure, combined with a dynamic semi-parametric Generalised Autoregressive Score (GAS) model, to more precisely quantify extreme risk shifts across both markets. Additionally, by constructing a multi-layered network structure, we systematically analyse risk spillover effects between different commodity categories and financial markets. Our findings indicate: ① Significant positive risk linkage exists between commodity and financial markets, with upside tail risks in commodities being more readily transmitted to financial markets; ② The persistence of commodity market risks and the range of risk loss variations are both smaller than those in financial markets such as banking; ③ Commodity markets exhibit a pronounced net risk input to banking, securities, and insurance markets. Among these, commodity markets for raw materials such as chemicals, non-ferrous metals, and oils and fats demonstrate a strong net risk output effect on financial markets, whereas agricultural and precious metals commodity markets experience positive net risk contagion from financial markets. Our findings not only illuminate risk transmission between commodities and financial markets but also provide decision-makers with reference points for risk early warning systems and macro-prudential regulation.

Key words: commodity market, financial market, systemic risk, risk spillover, network analysis