Conditional Generators for Limit Order Book Environments: Explainability, Challenges, and Robustness

TL;DR

This paper explores the use of conditional generative adversarial networks to simulate limit order book environments, analyzing their explainability, robustness, and challenges for improved trading agent development.

Contribution

It investigates the dependence of CGANs on input features, identifies weaknesses through adversarial attacks, and proposes improvements for realism and robustness in order book simulation.

Findings

Identified vulnerabilities of CGANs via adversarial attacks

Enhanced CGAN realism and robustness through insights gained

Provided a roadmap for future research in order book modeling

Abstract

Limit order books are a fundamental and widespread market mechanism. This paper investigates the use of conditional generative models for order book simulation. For developing a trading agent, this approach has drawn recent attention as an alternative to traditional backtesting due to its ability to react to the presence of the trading agent. Using a state-of-the-art CGAN (from Coletta et al. (2022)), we explore its dependence upon input features, which highlights both strengths and weaknesses. To do this, we use "adversarial attacks" on the model's features and its mechanism. We then show how these insights can be used to improve the CGAN, both in terms of its realism and robustness. We finish by laying out a roadmap for future work.