Aggregation Trees

Abstract

Uncovering the heterogeneous effects of particular policies or "treatments" is a key concern for researchers and policymakers. A common approach is to report average treatment effects across different subgroups based on observable covariates. However, there is likely to be considerable uncertainty about the appropriate grouping. This paper proposes a nonparametric approach to discovering heterogeneous subgroups in a selection-on-observables framework. The approach constructs a sequence of groupings, one for each level of granularity. Groupings are nested and feature an optimality property. An "honesty" condition allows us to construct valid confidence intervals for the average treatment effect of each group. The utility of the proposed methodology is
illustrated through an empirical exercise that revisits the impact of maternal smoking on birth weight.

Keywords: Causality, conditional average treatment effects, recursive partitioning, subgroup discovery, subgroup analysis

JEL Classification: C29, C45, C55