ARL-Based Multi-Action Market Making with Hawkes Processes and Variable Volatility

TL;DR

This paper introduces a novel market-making strategy combining Adversarial Reinforcement Learning, Hawkes Processes, and variable volatility to improve adaptability and robustness in dynamic market conditions.

Contribution

It develops a multi-action market-making framework using Hawkes processes and variable volatility, expanding the action space and demonstrating improved adaptability over traditional methods.

Findings

High-volatility trained strategies maintain stable performance.

Strategies quote on both sides at least 92% of the time in high volatility.

Hawkes processes better capture market dynamics than Poisson models.

Abstract

We advance market-making strategies by integrating Adversarial Reinforcement Learning (ARL), Hawkes Processes, and variable volatility levels while also expanding the action space available to market makers (MMs). To enhance the adaptability and robustness of these strategies -- which can quote always, quote only on one side of the market or not quote at all -- we shift from the commonly used Poisson process to the Hawkes process, which better captures real market dynamics and self-exciting behaviors. We then train and evaluate strategies under volatility levels of 2 and 200. Our findings show that the 4-action MM trained in a low-volatility environment effectively adapts to high-volatility conditions, maintaining stable performance and providing two-sided quotes at least 92\% of the time. This indicates that incorporating flexible quoting mechanisms and realistic market simulations significantly enhances the effectiveness of market-making strategies.