Alpha Go Everywhere: Machine Learning and International Stock Returns
55 Pages Posted: 5 Dec 2019 Last revised: 22 Mar 2023
Date Written: March 22, 2022
Abstract
We apply machine learning techniques to predict international stock returns using firm characteristics. Market-specific features are important, as neural network models (NNs) achieve stronger results when they are trained in each market separately than in a global model trained with U.S. data. NNs outperform linear models in predicting stock return rankings and forming profitable portfolios. In contrast, regression trees underperform linear models when the number of observations is low. We also show that adding foreign variables constructed from U.S. firm characteristics further enhances the return predictability of market-specific NNs, consistent with the notion that the markets are partially integrated.
Keywords: International Asset Pricing, Cross-section of Stock Returns, Market Integration, Neural Networks, Regression Trees
JEL Classification: C52, G10, G12, G15
Suggested Citation: Suggested Citation