PPO-HRAP Achieves 27.62% Return in Risk-Controlled Trading via Regime-Aware Policies
October 2, 2026
PPO-HRAP integrates Proximal Policy Optimization with an interpretable regime prior to balance returns against drawdowns. In SPY test windows from 2020-2022, the method achieved a 0.6447 Sharpe ratio and significantly reduced maximum drawdown compared to standard profit-only policies.
HOW THIS AFFECTS YOU
●
builderYou can use regime-aware priors to stabilize RL agents in highly volatile financial environments.
●
researcherThe hybrid approach offers a structured way to combine model-free RL with interpretable domain knowledge.