Distilling Causal Effects: Stable Subgroup Estimation Via Distillation Trees in Causal Inference

• In Event: Causal Inference

Brief Overview

We propose a method, causal distillation trees, which constructs sparse, interpretable subgroups for researchers to understand which moderators are most driving the underlying treatment effect heterogeneity in their study.

Authors

• Melody Huang, Yale University

• Tiffany Tang, University of Notre Dame

• Ana Maria Kenney, University of California-Irvine