🤖 AI Summary
This study investigates the heterogeneous roles of high-frequency trading (HFT) strategies in market information processing. To address the lack of interpretable, quantifiable strategy classification, we propose the first behaviorally grounded, measurable HFT strategy taxonomy distinguishing liquidity-providing from liquidity-demanding strategies. Leveraging financial time-series data, we develop a supervised learning framework integrating causally motivated feature engineering and strategy attribution analysis to model the nonlinear interactions between HFT activity and market dynamics. Empirical results show that liquidity-providing HFT exhibits heightened sensitivity to information events, significantly curbing delayed arbitrage and accelerating price discovery; in contrast, liquidity-demanding strategies display lagged responses and pronounced procyclicality. Our work provides novel microstructural evidence and a refined conceptual lens for understanding HFT’s role in market information production mechanisms.
📝 Abstract
We design and train machine learning models to capture the nonlinear interactions between financial market dynamics and high-frequency trading (HFT) activity. In doing so, we introduce new metrics to identify liquidity-demanding and -supplying HFT strategies. Both types of HFT strategies increase activity in response to information events and decrease it when trading speed is restricted, with liquidity-supplying strategies demonstrating greater responsiveness. Liquidity-demanding HFT is positively linked with latency arbitrage opportunities, whereas liquidity-supplying HFT is negatively related, aligning with theoretical expectations. Our metrics have implications for understanding the information production process in financial markets.