Discovery of a 13-Sharpe OOS Factor: Drift Regimes Unlock Hidden Cross-Sectional Predictability

2025-11-16

AI Summary

This paper addresses the well-documented out-of-sample failure of cross-sectional factors under high Sharpe ratios. We propose a conditional factor construction method based on individual stock “drift states”—defined as periods where the proportion of positive-return days over a 63-day rolling window exceeds 60%—activating only a composite signal of value and short-term reversal strategies during such drift regimes. Our key contribution is the first identification and exploitation of the drift regime to uncover latent cross-sectional predictability, thereby dynamically coupling factor logic with market state. The factor employs fully frozen parameters and rigorous out-of-sample validation: it delivers an annualized return of 158.6%, volatility of 12.0%, maximum drawdown of −11.9%, and a Sharpe ratio exceeding 13. It exhibits near-zero exposure to conventional risk factors, passes 1,000 randomization tests at *p* < 0.001, and supports an estimated strategy capacity of $100 million to $500 million.

Technology Category

• Reasoning under Uncertainty: Stochastic Optimization
• Data Mining & Knowledge Management: Anomaly/Outlier Detection
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: The sharing economy
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

We document a high-performing cross-sectional equity factor that achieves out-of-sample Sharpe ratios above 13 through regime-conditional signal activation. The strategy combines value and short-term reversal signals only during stock-specific drift regimes, defined as periods when individual stocks show more than 60 percent positive days in trailing 63-day windows. Under these conditions, the factor delivers annualized returns of 158.6 percent with 12.0 percent volatility and a maximum drawdown of minus 11.9 percent. Using rigorous walk-forward validation across 20 years of S&P 500 data (2004 to 2024), we show performance roughly 13 times stronger than market benchmarks on a risk-adjusted basis, produced entirely out-of-sample with frozen parameters. The factor passes extensive robustness tests, including 1,000 randomization trials with p-values below 0.001, and maintains Sharpe ratios above 7 even under 30 percent parameter perturbations. Exposure to standard risk factors is negligible, with total R-squared values below 3 percent. We provide mechanistic evidence that drift regimes reshape market microstructure by amplifying behavioral biases, altering liquidity patterns, and creating conditions where cross-sectional price discovery becomes systematically exploitable. Conservative capacity estimates indicate deployable capital of 100 to 500 million dollars before noticeable performance degradation.

Problem

• Identifies regime-specific conditions for cross-sectional equity predictability
• Develops factor combining value and reversal signals during drift regimes
• Demonstrates exploitable market microstructure changes during specific price patterns

Innovation

• Regime-conditional signal activation for factor timing
• Combining value and reversal during drift regimes
• Walk-forward validation with frozen parameters