AlphaLedger improves stock trading models with cost-aware factor learning
AlphaLedger: Portfolio-Grounded On-Policy Distillation for Regime-Aware Alpha Mining
Computational Engineering, Finance, and Science
Summary
Stock trading strategies rely on predictive factors that can become less valuable due to costs or overlaps. The authors present AlphaLedger, a method that trains a local model to generate trading signals by learning from a larger expert while considering portfolio performance and trading costs. This model adapts to changing market conditions and stock-specific data without needing continual access to the expert. Tests on Chinese stocks show AlphaLedger outperforms other methods with higher returns and better risk management.
What this means in practice
- •For quantitative trading teams: Build local trading models that adapt to changing market conditions while accounting for factor costs and portfolio impact based on historical expert guidance.
- •For portfolio risk analysts: Use AlphaLedger's regime-residual gating to better adjust factor weights according to recent market regimes, improving risk control during varying market conditions.
Authors
Xiang Li, Zikai Wei, Yiyan Qi, Wanyun Zhou, Xiang Liu, Penglei Sun, Zhenheng Tang, Yongqi Zhang, Jian Guo, Xiaowen Chu
Abstract
Hosted LLMs enable automated alpha mining but incur recurring API costs and expose research context to external providers. Training a local generator requires supervision aligned with portfolio utility: a predictive factor may add little after redundancy and trading costs. This incremental value is assessed only after the factor program is complete, whereas generation proceeds sequentially. Deployment must also adapt to evolving stock-specific evidence and changes in factor effectiveness across market regimes. We propose AlphaLedger to address these challenges. Economic-context on-policy distillation (EC-OPD) conditions a frozen teacher on completed factors and their cost-adjusted marginal portfolio contributions, then distills its legal-action distributions along the student's own trajectories. The deployed generator requires neither teacher calls nor outcome inputs. EviPool constructs a dynamic stock universe from lagged multi-source evidence, while a regime-residual gate (RRG) reweights factors using matured, state-matched returns with shrinkage and bounded deviations from equal weighting. On Chinese A-shares, AlphaLedger achieves a 55.4\% annualized net return, a Sharpe ratio of 2.06, and a maximum drawdown of 10.8\%. In a common CSI500 universe, its annualized net return exceeds the strongest baseline by 43.6\% (31.6\% vs.\ 22.0\%), with a 13.5\% higher Sharpe ratio.