Paper: arXiv 2303.15216

Authors: David Wu, Sebastian Jaimungal

Abstract

The objectives of option hedging/trading extend beyond mere protection against downside risks, with a desire to seek gains also driving agent’s strategies. In this study, we showcase the potential of robust risk-aware reinforcement learning (RL) in mitigating the risks associated with path-dependent financial derivatives. We accomplish this by leveraging a policy gradient approach that optimises robust risk-aware performance criteria. We specifically apply this methodology to the hedging of barrier options, and highlight how the optimal hedging strategy undergoes distortions as the agent moves from being risk-averse to risk-seeking. As well as how the agent robustifies their strategy. We further investigate the performance of the hedge when the data generating process (DGP) varies from the training DGP, and demonstrate that the robust strategies outperform the non-robust ones.

Complexity vs Empirical Score

  • Math Complexity: 7.0/10
  • Empirical Rigor: 6.5/10
  • Quadrant: Holy Grail — high math complexity, high empirical rigor

Why this score: The paper presents a solid theoretical framework combining robust policy gradients with risk-aware performance criteria for option hedging. While the mathematical foundation for robust RL is strong, the empirical evaluation relies primarily on simulated market shifts rather than extensive real-world historical data backtesting.

Research Flowchart

  flowchart TD
    A["Research Goal: Robust & Risk-Aware<br>Hedging for Barrier Options"] --> B["Inputs: Training Data Generating Process<br>& Target Derivatives Parameters"]
    B --> C["Methodology: Robust Risk-Aware RL<br>using Policy Gradient"]
    C --> D["Computational Process: Optimise Policy<br>across Risk Criteria & Model Shifts"]
    D --> E["Findings: Strategy Shifts from<br>Risk-Averse to Risk-Seeking"]
    D --> F["Findings: Robust Strategies Outperform<br>Non-Robust Under DGP Shifts"]