Towards multi‐agent reinforcement learning‐driven over‐the‐counter market simulations
From MaRDI portal
(Redirected from Publication:6196291)
Abstract: We study a game between liquidity provider and liquidity taker agents interacting in an over-the-counter market, for which the typical example is foreign exchange. We show how a suitable design of parameterized families of reward functions coupled with associated shared policy learning constitutes an efficient solution to this problem. Precisely, we show that our deep-reinforcement-learning-driven agents learn emergent behaviors relative to a wide spectrum of incentives encompassing profit-and-loss, optimal execution and market share, by playing against each other. In particular, we find that liquidity providers naturally learn to balance hedging and skewing as a function of their incentives, where the latter refers to setting their buy and sell prices asymmetrically as a function of their inventory. We further introduce a novel RL-based calibration algorithm which we found performed well at imposing constraints on the game equilibrium, both on toy and real market data.
Recommendations
- AI-driven liquidity provision in OTC financial markets
- Dynamics of market making algorithms in dealer markets: Learning and tacit collusion
- Reinforcement learning equilibrium in limit order markets
- Market Making and Incentives Design in the Presence of a Dark Pool: A Stackelberg Actor–Critic Approach
- Cooperation between independent market makers
Cites work
- A learning market-maker in the Glosten–Milgrom model
- A stochastic partial differential equation model for limit order book dynamics
- Agent-based model calibration using machine learning surrogates
- Algorithmic and high-frequency trading
- Algorithmic market making in dealer markets with hedging and market impact
- Convergence rate of linear two-time-scale stochastic approximation.
- Convergent multiple-timescales reinforcement learning algorithms in normal form games
- Convex optimization: algorithms and complexity
- Dealing with the inventory risk: a solution to the market making problem
- Equilibria in symmetric games: theory and applications
- Execution in an aggregator
- Flows and decompositions of games: harmonic and potential games
- High-frequency trading in a limit order book
- scientific article; zbMATH DE number 3951715 (Why is no real title available?)
- scientific article; zbMATH DE number 1405930 (Why is no real title available?)
- scientific article; zbMATH DE number 7064064 (Why is no real title available?)
- scientific article; zbMATH DE number 3069635 (Why is no real title available?)
- Learning a functional control for high-frequency finance
- Liquidity in competitive dealer markets
- On the convergence rate in Lyapunov's theorem
- On the theory of policy gradient methods: optimality, approximation, and distribution shift
- Optimal market making
- Potential games
- Pure strategy equilibria in symmetric two-player zero-sum games
- Size matters for OTC market makers: General results and dimensionality reduction techniques
- Stochastic approximation with two time scales
- Welfare effects of market making in continuous double auctions
This page was built for publication: Towards multi‐agent reinforcement learning‐driven over‐the‐counter market simulations
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6196291)
