Minimax Optimal Nonparametric Estimation of Heterogeneous Treatment Effects

Abstract

A central goal of causal inference is to detect and estimate the treatment effects of a given treatment or intervention on an outcome variable of interest, where a member known as the heterogeneous treatment effect (HTE) is of growing popularity in recent practical applications such as the personalized medicine. In this paper, we model the HTE as a smooth nonparametric difference between two less smooth baseline functions, and determine the tight statistical limits of the nonparametric HTE estimation as a function of the covariate geometry. In particular, a two-stage nearest-neighbor-based estimator throwing away observations with poor matching quality is near minimax optimal. We also establish the tight dependence on the density ratio without the usual assumption that the covariate densities are bounded away from zero, where a key step is to employ a novel maximal inequality which could be of independent interest.

arXiv commentTo appear at NeurIPS 2020 as a spotlight presentation

References in corpus (8)

• Recursive Partitioning for Heterogeneous Causal Effects
• Meta-learners for Estimating Heterogeneous Treatment Effects using Machine Learning
• Estimating treatment effect heterogeneity in randomized program evaluation
• Estimating Mutual Information for Discrete-Continuous Mixtures
• Divided Differences
• The Nearest Neighbor Information Estimator is Adaptively Near Minimax Rate-Optimal
• Causal Etiology of the Research of James M. Robins
• Adaptation to lowest density regions with application to support recovery

Cited by in corpus (3)

• Quantifying Ignorance in Individual-Level Causal-Effect Estimates under Hidden Confounding
• A causal fused lasso for interpretable heterogeneous treatment effects estimation
• Doubly Robust Off-Policy Learning on Low-Dimensional Manifolds by Deep Neural Networks