Generalizable estimation of conditional average treatment effects using Causal Forest in randomized

2025-06-14

AI Summary

This study addresses the degradation of conditional average treatment effect (CATE) estimation generalizability when extrapolating randomized controlled trial (RCT) results to the source population, due to selection bias and high-dimensional covariates. It provides the first systematic evaluation of four strategies for mitigating selection bias within the causal forest framework. Results show that directly incorporating selection variables yields theoretically unbiased estimates but incurs substantial variance inflation; in contrast, inverse probability weighting (IPW)-based correction achieves lower bias and well-controlled variance across most simulation settings, demonstrating superior robustness. The study proposes a practical IPW–causal forest integration framework that delivers CATE estimates with both theoretical validity and empirical robustness for RCT external validity. This work fills a critical gap by offering the first comprehensive assessment of selection bias correction methods under high-dimensional settings.

Technology Category

Machine Learning: Causal LearningReasoning under Uncertainty: CausalityIntelligent Robots: State Estimation

Application Category

User Modeling, Personalization and Recommendation: Attacks and countermeasures in recommendation systemsSearch and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for rankingWeb Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Generalizing conditional average treatment effects (CATE) estimates in a randomized controlled trial (RCT) to a broader source population can be challenging because of selection bias and high-dimensional covariates. We aim to evaluate CATE estimation approaches using Causal Forest that address selection bias due to trial participation. We propose and compare four CATE estimation approaches using Causal Forest: (1) ignoring selection variables, (2) including selection variables, (3) using inverse probability weighting (IPW) either with (1) or (2). Identifiable condition suggests that including covariates that determine trial selection in CATE-estimating models can yield an unbiased CATE estimate in the source population. However, simulations showed that, in realistic sample sizes in a medical trial, this approach substantially increased variance compared with little gain in bias reduction. IPW-based approaches showed a better performance in most settings by addressing selection bias. Increasing covariates that determine trial participation in Causal Forest estimation can substantially inflate the variance, diminishing benefits of bias reduction. IPW offers a more robust method to adjust for selection bias due to trial participation.

Problem

• Estimating CATE in RCTs with selection bias challenges
• Comparing Causal Forest methods for bias adjustment
• Evaluating IPW performance in reducing selection bias

Innovation

• Uses Causal Forest for CATE estimation
• Compares four selection bias adjustment methods
• IPW performs best in bias correction