Investable and Interpretable Machine Learning for Equities

The authors propose three principles for evaluating the practical efficacy of machine learning for stock selection, and they compare the performance of various models and investment goals using this framework. The first principle is investability. To this end, the authors focus on portfolios formed from highly liquid US stocks, and they calibrate models to require a reasonable amount of trading. The second principle is interpretability. Investors must understand a model’s output well enough to trust it and extract some general insight from it. To this end, the authors choose a concise set of predictor variables, and they apply a novel method called the model fingerprint to reveal the linear, nonlinear, and interaction effects that drive a model’s predictions. The third principle is that a model’s predictions should be interesting—they should convincingly outperform simpler models. To this end, the authors evaluate out-of-sample performance compared to linear regressions. In addition to these three principles, the authors also consider the important role people play by imparting domain knowledge and preferences to a model. The authors argue that adjusting the prediction goal is one of the most powerful ways to do this. They test random forest, boosted trees, and neural network models for multiple calibrations that they conclude are investable, interpretable, and interesting.