BGM-IV: an AI-powered Bayesian generative modeling approach for instrumental variable analysis

TL;DR

BGM-IV introduces a Bayesian latent space approach for nonlinear instrumental variable regression, effectively handling high-dimensional covariates and complex causal structures.

Contribution

It reframes nonlinear IV regression as posterior inference in a causally structured latent space, improving performance in high-dimensional settings.

Findings

Performs best in high-dimensional covariate regimes.

Remains competitive in low-dimensional regimes.

Provides a principled latent generative modeling framework.

Abstract

Instrumental-variable (IV) regression enables causal estimation under endogeneity, but modern IV problems often involve nonlinear structural effects and high-dimensional covariates. Existing nonlinear IV methods directly learn the causal relation in observed feature space or rely on learned representations within two-stage or moment-based procedures, which can struggle when the causal information is embedded in a high-dimensional representation. We propose BGM-IV, a latent Bayesian generative modeling approach that reframes nonlinear IV regression as posterior inference in a causally structured latent space. BGM-IV infers latent components that separately capture shared confounding structure, outcome-specific variation, treatment-specific variation, and covariate-only nuisance information. To account for endogeneity, BGM-IV replaces the confounded outcome likelihood with an IV-integrated pseudo-likelihood that averages over instrument-induced treatment values within the latent model. Across various benchmark datasets, BGM-IV remains competitive in the classical low-dimensional regime and performs best in high-dimensional covariate regimes. Together, these results show that structured latent generative modeling provides a principled and effective strategy to nonlinear IV estimation with rich covariates. The code of BGM-IV is available at https://github.com/liuq-lab/BGM-IV.