Deep Hedging with Market Impact

TL;DR

This paper introduces a Deep Reinforcement Learning model for dynamic option hedging that explicitly accounts for market impact and liquidity constraints, leading to more cost-effective and realistic hedging strategies.

Contribution

It presents a novel DRL-based hedging framework that incorporates market impact, impact persistence, and realistic trading features, improving over traditional methods.

Findings

DRL model outperforms delta hedging in low liquidity scenarios

The model learns to dampen or delay rebalancing to reduce costs

Incorporates effects of previous hedging errors and asset drift

Abstract

Dynamic hedging is the practice of periodically transacting financial instruments to offset the risk caused by an investment or a liability. Dynamic hedging optimization can be framed as a sequential decision problem; thus, Reinforcement Learning (RL) models were recently proposed to tackle this task. However, existing RL works for hedging do not consider market impact caused by the finite liquidity of traded instruments. Integrating such feature can be crucial to achieve optimal performance when hedging options on stocks with limited liquidity. In this paper, we propose a novel general market impact dynamic hedging model based on Deep Reinforcement Learning (DRL) that considers several realistic features such as convex market impacts, and impact persistence through time. The optimal policy obtained from the DRL model is analysed using several option hedging simulations and compared to commonly used procedures such as delta hedging. Results show our DRL model behaves better in contexts of low liquidity by, among others: 1) learning the extent to which portfolio rebalancing actions should be dampened or delayed to avoid high costs, 2) factoring in the impact of features not considered by conventional approaches, such as previous hedging errors through the portfolio value, and the underlying asset's drift (i.e. the magnitude of its expected return).