The Promises and Pitfalls of Machine Learning for Predicting Stock Returns
42 Pages Posted: 27 Mar 2020
Date Written: March 1, 2020
Recent research suggests that machine learning models dominate traditional linear models in predicting cross-sectional stock returns. Indeed, we confirm this finding when predicting one-month forward looking returns based on a set of common equity factors, including predictors such as short-term reversal. Despite this statistical advantage of machine learning model predictions, we demonstrate economic gains to be more limited and critically dependent on the ability to take risk and implement trades efficiently. Unlike traditional models, machine-learning models have struggled less over the last decade in discerning valuable predictions from cross-sectional equity characteristics.
Keywords: Machine Learning, Data Science, Interpretable Machine Learning, Return Prediction, Cross-Section of Stock Returns, Gradient Boosting, Factor Investing
JEL Classification: G11, G12, G14, G15, G17
Suggested Citation: Suggested Citation