Can Financial Markets Anticipate Geopolitical Risk in China? Evidence from Artificial Intelligence-Based Forecasting Models

This study investigates whether China’s financial markets contain forward-looking information that can be used to predict geopolitical risks. It addresses a gap in the existing literature, which predominantly treats geopolitical risk as an exogenous shock affecting financial markets, by examining whether financial market movements can instead provide anticipatory signals of future geopolitical developments. Using high-frequency variables from China’s equity, bond, money, and foreign exchange markets, four quantum-enhanced Support Vector Regression models—QPSO-SVR, QACO-SVR, QFOA-SVR, and QABC-SVR—are developed to predict the Geopolitical Risk Index (GPR), Geopolitical Acts Index (GPA), and Geopolitical Threats Index (GPT). The empirical results reveal significant time-varying volatility across all three geopolitical risk measures and identify a distinct transmission sequence among them. More importantly, financial market indicators demonstrate significant predictive power for geopolitical risk, suggesting that asset prices incorporate information that precedes geopolitical events. Among the competing forecasting models, QPSO-SVR provides the strongest predictive performance for aggregate geopolitical risk, while QACO-SVR performs particularly well in forecasting realized geopolitical acts and geopolitical threats. The results further demonstrate that quantum swarm-intelligence optimization can substantially improve the efficiency and accuracy of SVR hyperparameter selection, thereby reducing forecasting errors. Overall, the findings challenge the conventional view of geopolitical risk as purely exogenous and demonstrate that financial markets can serve as valuable sources of forward-looking geopolitical information. The study also highlights the potential of quantum-enhanced machine-learning techniques for extracting complex and economically meaningful geopolitical signals from high-frequency financial market data.