Reinforcement Learning for Market Making in a Multi-agent Dealer Market

TL;DR

This paper develops a multi-agent simulation to train and analyze reinforcement learning-based market makers, demonstrating their ability to learn competitor strategies, manage inventory, and adapt to market trends.

Contribution

It introduces a multi-agent dealer market simulator and shows how RL agents can learn strategic pricing, inventory management, and risk aversion in market making.

Findings

RL agents learn competitor pricing policies

Agents manage inventory by asymmetric pricing

Reward functions can induce risk-averse behavior

Abstract

Market makers play an important role in providing liquidity to markets by continuously quoting prices at which they are willing to buy and sell, and managing inventory risk. In this paper, we build a multi-agent simulation of a dealer market and demonstrate that it can be used to understand the behavior of a reinforcement learning (RL) based market maker agent. We use the simulator to train an RL-based market maker agent with different competitive scenarios, reward formulations and market price trends (drifts). We show that the reinforcement learning agent is able to learn about its competitor's pricing policy; it also learns to manage inventory by smartly selecting asymmetric prices on the buy and sell sides (skewing), and maintaining a positive (or negative) inventory depending on whether the market price drift is positive (or negative). Finally, we propose and test reward formulations for creating risk averse RL-based market maker agents.