A Deep Multi-Factor Framework for Cross-Sectional Stock Return Ranking via Market-State Attention Pooling
“The model achieves higher information coefficient values, annualized excess returns, and information ratios, indicating improvements in both predictive accuracy and portfolio performance.”
AssetEquities
StrategyFactor investingportfolio optimization
TechniqueFeedforward network
GeographyChina
Also stated: Excess return, Fundamentals
Missing from abstract: benchmark, out-of-sample test, costs, multiple testing, universe
Institutional-grade audit of this strategy.
£5 — Request the audit- Audit of the paper claims accounting for costs, luck, crises, originality, geography, etc.
- Reproduction code integratable into Bloomberg, IBKR, etc.
- Bias-free datasets for backtesting included
- Delivered in 2 working days
The paper's neighbourhoodFactor investing by year
FamilyFactor investing240Momentum and trend following289Machine learning return prediction153News sentiment and text signals14818 more · 854
AssetEquities196
The family is counted across the whole universe. Asset, method and authors are counted inside the family.
| Similar papers | citation | same method | same asset | same authors | |
|---|---|---|---|---|---|
| Multifactor Timing with Deep Learning by Cotturo et al. | 2026 | ||||
| Multilayer Perceptron Neural Network Models in Asset Pricing: An Empirical Study on Large-Cap US Stocks by Lai | 2026 | ||||
| Factor Investing Lecture 9: Machine Learning in Factor Investing (Presentation Slides) by Shi | 2025 | ||||
| Deep Learning for Short Term Equity Trend Forecasting: A Behavior Driven Multi Factor Approach by Luan | 2025 | ||||
| Multi-Factor Timing with Deep Learning by Cotturo et al. | 2024 |