Graph-Split Bayesian Causal Forest for Spatial Heterogeneous Treatment Effect Estimation

2026-09-28

AI Summary

This study addresses the challenge of estimating heterogeneous treatment effects in spatial observational research, where conventional axis-aligned splitting rules are ill-suited to underlying spatial dependencies. To this end, we propose the Graph-guided Spatial Bayesian Causal Forest (GSBCF) method. GSBCF introduces a spatial-structure-aware graph splitting rule to replace traditional axis-aligned splits and develops an efficient information-proposal sampling algorithm. By integrating GS-BART with Markov chain Monte Carlo techniques, the method achieves full Bayesian inference. GSBCF effectively unifies graph-guided splitting trees within a Bayesian causal forest framework, substantially improving both estimation accuracy and uncertainty quantification for spatially heterogeneous causal inference. Empirical evaluations demonstrate that the proposed approach outperforms existing causal BART methods.

Abstract

In spatial observational studies, treatment assignment and outcomes often exhibit spatial dependence patterns, and treatment effects may vary across space and subpopulations due to both measured and unmeasured spatially structured confounders. Accounting for spatial dependence while estimating heterogeneous treatment effects (HTEs) is a central task in spatial causal inference. Causal Bayesian additive regression tree methods are popular nonparametric methods for modeling and estimating HTEs. Despite their flexibility and uncertainty quantification, the axis-aligned split rules often adopted in these models are not suitable for modeling spatial structures. We propose a spatial structure-aware Bayesian nonparametric method, called Graph-Split Bayesian Causal Forest (GSBCF), that integrates graph-split Bayesian additive regression trees (GS-BART) with the Bayesian causal forest propensity-score regression framework for spatial heterogeneous causal inference. Spatial confounding is accommodated through graph-guided split rules in modeling decision trees of the prognostic and HTE functions. We develop an efficient informed proposal sampling algorithm for posterior computation, enabling full Bayesian inference of the spatial conditional average treatment effect function. Simulations and a real data study demonstrate substantially improved estimation accuracy and uncertainty quantification over existing causal BART methods.

Problem

• Spatial Heterogeneous Treatment Effect
• Spatial Causal Inference
• Bayesian Additive Regression Trees
• Spatial Confounding
• Observational Studies

Innovation

• Graph-Split Bayesian Causal Forest
• Spatial Heterogeneous Treatment Effect
• Bayesian Additive Regression Trees
• Graph-Guided Split Rules
• Informed Proposal Sampling