Bayesian Causal Effect Estimation for Categorical Data using Staged Tree Models

TL;DR

This paper introduces a Bayesian method for causal inference with multivariate categorical data using staged tree models, capturing complex dependencies and providing uncertainty quantification.

Contribution

It develops a fully Bayesian framework with novel priors and inference algorithms for staged trees, enabling flexible and interpretable causal analysis.

Findings

Effective in modeling asymmetric dependencies

Provides credible treatment effect estimates

Demonstrated on medical case studies

Abstract

We propose a fully Bayesian approach for causal inference with multivariate categorical data based on staged tree models, a class of probabilistic graphical models capable of representing asymmetric and context-specific dependencies. To account for uncertainty in both structure and parameters, we introduce a flexible family of prior distributions over staged trees. These include product partition models to encourage parsimony, a novel distance-based prior to promote interpretable dependence patterns, and an extension that incorporates continuous covariates into the learning process. Posterior inference is achieved via a tailored Markov Chain Monte Carlo algorithm with split-and-merge moves, yielding posterior samples of staged trees from which average treatment effects and uncertainty measures are derived. Posterior summaries and uncertainty measures are obtained via techniques from the Bayesian nonparametrics literature. Two case studies on electronic fetal monitoring and cesarean delivery and on anthracycline therapy and cardiac dysfunction in breast cancer illustrate the methods.