Network-aware IV Regression for Causal Node Discovery and Estimation

TL;DR

This paper introduces a novel two-stage regression method that combines instrumental variables and graph-based regularization to improve causal effect estimation in network-structured high-dimensional data, handling invalid instruments and providing theoretical guarantees.

Contribution

It develops a new framework integrating IV regression with graph regularization for causal discovery, addressing latent confounding and invalid instruments in structured data.

Findings

Improved causal variable selection accuracy over existing methods.

Theoretical non-asymptotic guarantees for estimation and selection.

Successful application to brain imaging and genetic data identifying causal regions.

Abstract

Estimating causal effects from high-dimensional, structured exposures is a fundamental challenge in modern applications ranging from neuroscience and finance to environmental science. While the literature has addressed high-dimensional instrumental variable (IV) regression, and separately leveraged graph structure in penalized regression, the integration of both, especially for causal support recovery in the presence of latent confounding, remains unexplored. In this work, we propose a novel two-stage regression framework that incorporates instrumental variables and graph-based regularization to uncover sparse causal effects among network-structured exposures. Our method accommodates both valid and partially invalid instruments, and encourages structural similarity among connected predictors through a graph-fused penalty. We establish non-asymptotic guarantees for estimation accuracy and causal variable selection, and demonstrate that our approach yields improved performance over existing methods that ignore network dependencies or invalid IVs. Applied to ADNI brain imaging and genetic data, our method identifies interpretable causal ROIs associated with cognitive outcomes, underscoring the utility of graph-assisted IV regression in neuroscience and beyond.