Primary Submission Category: Machine Learning and Causal Inference
Distilling causal effects: stable subgroup estimation via distillation trees in causal inference
Authors: Ana Kenney, Melody Huang, Tiffany Tang, Tanvi Shinkre,
Presenting Author: Ana Kenney*
We introduce a method, causal distillation trees (CDT), that allows researchers to stably estimate interpretable causal subgroups in their studies. CDT allows
researchers to fit any machine learning model of their choice to estimate the individual-level treatment effect, and then leverages a simple, second-stage tree-based model to then “distill’’ the estimated treatment effect into meaningful subgroups. As a result, we are able to leverage the theoretical guarantees from black-box machine learning models, while preserving the interpretability of a simple decision tree. We theoretically characterize the stability of CDT in estimating substantively meaningful subgroups, and provide helpful diagnostics for researchers to evaluate the quality of the estimated subgroups. We empirically demonstrate our method via extensive simulations and a case study on jobs training program experiments. We show that CDT out-performs state-of-the-art approaches in identifying interpretable subgroups.