Improved market making under complex order flow and price impact risks
Robust Market Making with Hawkes Order Flow and Price Impact via Adversarial Reinforcement Learning
Machine Learning
Summary
Market makers buy and sell assets to help markets run smoothly but face unpredictable conditions. This paper improves a method that uses artificial intelligence to make market-making strategies more robust by modeling more realistic patterns of order arrivals and price changes. The new approach includes patterns where trades can trigger more trades and price effects that persist after trades. The authors add technology that remembers past events to better understand shifting market conditions. Their experiments show that this method reduces the risk of large losses under complex and changing market behaviors.
What this means in practice
- •For quantitative trading teams: Deploy more resilient market-making algorithms that handle clustered trades and dynamic price effects in limit order book markets.
- •For financial risk managers: Assess and mitigate worst-case loss scenarios in automated trading by simulating adversarial market behaviors with self-exciting order flows.
Authors
Hao Yang, Zhenguo Xu
Abstract
Market-making strategies in real limit order book markets face substantial model uncertainty and regime-shift risk. Existing adversarial reinforcement learning approaches improve robustness by formulating the Avellaneda--Stoikov market-making problem as a zero-sum game between a market maker and an environmental adversary. However, these approaches typically rely on Poisson order arrivals and neglect trade-induced price impact, limiting their ability to capture important high-frequency market microstructure effects such as clustered order flow, self-excitation, and post-trade price feedback. We extend adversarial reinforcement learning for market making to a more complex environment with Hawkes self-exciting order arrivals and trade-induced price impact. To mitigate the increased non-stationarity introduced by the expanded regime space, we incorporate an LSTM module that explicitly models the temporal structure of recent observations. We further characterize the equilibrium properties of the proposed framework through both game-theoretic analysis and numerical experiments, and introduce a robustness evaluation protocol focused on improvements in the left tail of the return distribution. Experimental results across a range of market regimes show that the proposed method achieves improved left-tail performance in most complex microstructure environments. In particular, the gains are pronounced in regimes with strong Hawkes excitation and low-to-moderate price impact. Bootstrap tests provide no evidence that these improvements are obtained through a stronger terminal directional inventory bias. These results suggest that combining adversarial training with temporal state representation can improve the robustness of reinforcement-learning-based market-making strategies under order-flow self-excitation, price impact, and regime uncertainty.