🤖 AI Summary
This study addresses the challenges of estimating heterogeneous causal effects and discovering subgroups under imperfect compliance by proposing a two-step, model-agnostic framework. Methodologically, it introduces a non-Bayesian tree structure compatible with arbitrary machine learning algorithms, integrating DRRF-IV, GRF-IV, and forest learner techniques to achieve precise estimation of the compiler average causal effect (CCACE). Experimental results demonstrate that the proposed method attains accuracy comparable to BCF-IV while significantly improving computational efficiency. Furthermore, when applied to clinical ICU data, the framework successfully identifies specific beneficiary subgroups, providing a reliable basis for individualized medical decision-making.
📝 Abstract
We study the impact of conditional complier average causal effect (CCACE) estimation methods on the performance of subgroup discovery and heterogeneous causal effect estimation under imperfect compliance. Building on the Bayesian Causal Forest with Instrumental Variable (BCF-IV) (Bargagli-Stoffi et al. (2022)) method, we introduce a two-step, model-agnostic approach that allows any suitable machine learning method to be used for the CCACE estimation in the first step. Specifically, we implement two non-Bayesian tree-based methods, both using forest-based learners: DRRF-IV, a debiased transformed-outcome regression-forest approach, and a GRF-based IV adaptation of the generalized random forest framework (Athey et al., 2019). Through a simulation study, we assess the precision, bias, and the ability to correctly identify the underlying subgroup structure of the proposed methods relative to BCF-IV. The results show that the non-Bayesian methods perform competitively across the considered simulation settings, with performance improving for larger sample sizes and moderately large treatment effects, while reducing computational runtime. We apply our new methods by revisiting an empirical study that examines the effect of prompt admission to intensive care units (ICU) on 28-day mortality across 48 UK National Health Service hospitals. While previous work finds no significant overall treatment effect, we investigate whether subgroups of patients may benefit more from prompt ICU admission.