Variable importance measures for heterogeneous treatment effects

Abstract

Motivated by applications in precision medicine and treatment effect heterogeneity, recent research has focused on estimating conditional average treatment effects (CATEs) using machine learning (ML). CATE estimates may represent complicated functions that provide little insight into the key drivers of heterogeneity. Therefore, we introduce nonparametric treatment effect variable importance measures (TE-VIMs), based on the mean-squared error (MSE) in predicting the individual treatment effect. More precisely, TE-VIMs represent the increase in MSE when variables are removed from the CATE conditioning set. We derive efficient TE-VIM estimators which can be used with any CATE estimation strategy and are amenable to ML estimation. We propose several strategies to calculate these VIMs (e.g. leave-one out, or keep-one in), using popular meta-learners for the CATE. We study the finite sample performance through a simulation study and illustrate their application using clinical trial data.

Replication code is available from https://github.com/ohines/tevims

References in corpus (7)

• Recursive Partitioning for Heterogeneous Causal Effects
• Smoothing parameter and model selection for general smooth models
• Meta-learners for Estimating Heterogeneous Treatment Effects using Machine Learning
• Demystifying statistical learning based on efficient influence functions
• A general framework for inference on algorithm-agnostic variable importance
• A Flexible Approach for Predictive Biomarker Discovery
• Rank-transformed subsampling: inference for multiple data splitting and exchangeable p-values

Cited by in corpus (2)

• A Flexible Approach for Predictive Biomarker Discovery
• Inference on summaries of a model-agnostic longitudinal variable importance trajectory with application to suicide prevention