Alpha Go Everywhere: Machine Learning and International Stock Returns
57 Pages Posted: 5 Dec 2019 Last revised: 17 Mar 2020
Date Written: March 15, 2020
We apply machine learning techniques and use stock characteristics to predict the cross-section of stock returns in 33 international markets. We conduct a stringent out-of-sample test to allay concerns about overfitting: the models are trained with past U.S. data and used to predict international stock returns. With fewer variables (based on past returns, size, volume, and accounting information) as inputs, we arrive at a conclusion similar to that in previous studies predicting U.S. stock returns with hundreds of stock characteristics and macroeconomic variables; complex methods outperform linear models in terms of both predicting returns and generating profits. We achieve even stronger results when the models are trained separately for each market, allowing for country-specific return-characteristic relationships. In most markets, we obtain out-of-sample R^2 and Sharpe ratios that are comparable to those in previous studies. Neural network models, which can handle complicated interactions among the predictors, produce the best results.
Keywords: Machine Learning, Cross-section of Stock Returns, International Markets, Neural Networks
JEL Classification: G12, G15
Suggested Citation: Suggested Citation