Machine Learning Methods for Estimating Heterogeneous Causal Effects

In this paper we propose methods for estimating heterogeneity in causal effects in experimental and observational studies, and for conducting inference about the magnitude of the differences in treatment effects across subsets of the population. In applications, our method provides a data-driven approach to determine which subpopulations have large or small treatment effects and to test hypotheses about the differences in these effects. In most of the literature on supervised machine learning (e.g. regression trees, random forests, LASSO, etc.), the goal is to build a model of the relationship between a unit’s attributes and an observed outcome. Cross-validation plays a prominent role: the methods compare predictions to actual outcomes in test samples, in order to select the level of complexity of the model that provides the best out-of-sample predictive power. Our method is closely related, but it differs in that it is tailored for predicting causal effects of a treatment rather than a unit’s outcome. The challenge is that the “ground truth” for a causal effect is not observed for any individual unit: we observe the unit with the treatment, or without the treatment, but not both at the same time, and so new approaches are required to use cross-validation to determine whether a causal effect has been accurately predicted. We propose novel cross-validation criteria for causal effects and demonstrate through simulations the conditions under which they perform better than directly applying standard prediction methods.