Stock Return Forecasting Using SHAP-Based Feature Selection and Risk-Controlled Portfolio Construction

DOI

10.2991/978-94-6463-823-3_76How to use a DOI?

Keywords

Stock return forecasting; SHAP; Feature selection; LightGBM; Risk control

Abstract

This paper presents a stock return forecasting framework that integrates machine learning, explainable featureselection, and portfolio construction controlled by risk. A LightGBM model is trained to predict monthly stock returns based on a comprehensive set of financial indicators. SHAP (SHapley Additive exPlanations) values are used to rank feature importance, enabling the selection of the top five predictors with the strongest ex-planatory power. These selected features are used to retrain the model, and the resulting predictions are used to form an equal-weighted portfolio of the top-ranked stocks. The portfolio is rebalanced monthly and supplemented with a maximum drawdown control rule to mitigate downside risk. Historical backtesting demonstrates that this strategy out- performs a market benchmark in terms of both annualized return and risk-adjusted per- formance. By combining predictive accuracy with interpretability and explicit risk management, the proposed method addresses key challenges in applying machine learning to financial forecasting. The results suggest that explainable artificial intelligence can bridge the gap between model transparency and performance, offering practical value for asset management.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Xuan Zhang
PY  - 2025
DA  - 2025/08/31
TI  - Stock Return Forecasting Using SHAP-Based Feature Selection and Risk-Controlled Portfolio Construction
BT  - Proceedings of the 2025 3rd International Conference on Image, Algorithms, and Artificial Intelligence (ICIAAI 2025)
PB  - Atlantis Press
SP  - 770
EP  - 778
SN  - 2352-538X
UR  - https://doi.org/10.2991/978-94-6463-823-3_76
DO  - 10.2991/978-94-6463-823-3_76
ID  - Zhang2025
ER  -