Deep Hedging to Manage Tail Risk
This work addresses risk management for financial portfolios, particularly in crisis scenarios, but is incremental as it builds on existing Deep Hedging methods.
The paper tackles the portfolio tail-risk hedging problem by extending the Deep Hedging paradigm to parameterize convex-risk minimization using deep neural networks, achieving significant one-day 99% CVaR reduction in realistic market simulations.
Extending Buehler et al.'s 2019 Deep Hedging paradigm, we innovatively employ deep neural networks to parameterize convex-risk minimization (CVaR/ES) for the portfolio tail-risk hedging problem. Through comprehensive numerical experiments on crisis-era bootstrap market simulators -- customizable with transaction costs, risk budgets, liquidity constraints, and market impact -- our end-to-end framework not only achieves significant one-day 99% CVaR reduction but also yields practical insights into friction-aware strategy adaptation, demonstrating robustness and operational viability in realistic markets.