Get Started
Home
Topics
Search
Library
Research questionHow can we reliably select a heterogeneous treatment effect estimator without observing treatment effects?Heterogeneous treatment effects vary across individuals or covariate-defined groups, but the relevant counterfactual outcome is not observed for each person. This makes it difficult to identify the best estimator and distinguish genuine superiority from selection error.
Economics
Evaluation & Benchmarks
Machine Learning
Research Paper
Statistical Machine Learning
Latest papersRecent research connected to this question, newest first.Reliable Selection of Heterogeneous Treatment Effect EstimatorsThe evidence concerns ground-truth-free selection of HTE estimators using cross-fitted statistical inference, with asymptotic familywise error control under regularity conditions. Empirical support is reported on the ACIC 2016, IHDP, and Twins benchmarks; finite-sample performance beyond these settings is not established by the input.research paper · Sep 3, 2026
Related questions
How can we recover a target site's full treated-outcome distribution from controls alone under cross-site heterogeneity?How can sequential treatment allocation learn unknown outcome variances while preserving efficient ATE inference?How can extreme quantile treatment effects be estimated under heavy tails while respecting location invariance?How should treatment-effect heterogeneity and sample size guide sample splitting for individual effects in causal forests?