Intraday Liquidity Patterns and Their Implications for Market Risk Assessment: Evidence from Global Equity Markets

Abstract

This paper examines intraday liquidity patterns across global equity markets and evaluates their implications for market risk assessment. Utilizing high-frequency order book and transaction data from five major exchanges (NYSE, NASDAQ, LSE, TSE, and HKEX) spanning January 2019 through December 2023, we analyze the temporal dynamics of multiple liquidity dimensions including bid-ask spreads, market depth, and order book resilience. The empirical analysis employs panel regression models with flexible time-of-day indicators, vector autoregression with impulse response functions, and principal component analysis to characterize liquidity patterns and cross-market dependencies. Results reveal pronounced U-shaped patterns in bid-ask spreads across all markets, with statistically significant time-of-day effects and substantial cross-market heterogeneity in pattern magnitude. We document strong regional commonality in liquidity dynamics, with correlation coefficients ranging from 0.832 between NYSE-NASDAQ to 0.265 between NASDAQ-HKEX. The analysis identifies asymmetric spillover effects, with developed markets exerting stronger influence on emerging markets than vice versa. Integration of intraday liquidity metrics into GARCH-based risk forecasting models yields 12-18% improvements in prediction accuracy, with the largest gains during periods of market stress. These findings provide valuable insights for risk management professionals, enhancing both risk assessment frameworks and execution timing strategies in global financial markets.