Experimental Analysis of Deep Hedging Using Artificial Market Simulations for Underlying Asset Simulators
Masanori Hirano
TL;DR
The paper addresses the challenge of selecting robust underlying asset simulators for deep hedging by replacing parametric price processes with fully artificial, multi-agent market simulations. It systematically compares artificial market simulations against Brownian motion and the Heston model using a deep hedging framework under entropic risk and CVaR objectives, across European and Lookback options. The results show that artificial market simulations can achieve nearly identical, and in some cases superior, performance, while also revealing tail-risk dynamics that depend on the chosen risk measure and market conditions. The study highlights the potential of data-driven, agent-based market models as flexible underlying asset simulators for hedging in incomplete markets, and points to the need for more sophisticated simulations and robust training strategies to address limitations observed in tail events and parameter sensitivity.
Abstract
Derivative hedging and pricing are important and continuously studied topics in financial markets. Recently, deep hedging has been proposed as a promising approach that uses deep learning to approximate the optimal hedging strategy and can handle incomplete markets. However, deep hedging usually requires underlying asset simulations, and it is challenging to select the best model for such simulations. This study proposes a new approach using artificial market simulations for underlying asset simulations in deep hedging. Artificial market simulations can replicate the stylized facts of financial markets, and they seem to be a promising approach for deep hedging. We investigate the effectiveness of the proposed approach by comparing its results with those of the traditional approach, which uses mathematical finance models such as Brownian motion and Heston models for underlying asset simulations. The results show that the proposed approach can achieve almost the same level of performance as the traditional approach without mathematical finance models. Finally, we also reveal that the proposed approach has some limitations in terms of performance under certain conditions.