Exploiting independence constraints for efficient estimation of bounds on causal effects in the presence of unmeasured confounding

2026-05-12

AI Summary

This work proposes an efficient method for estimating sharp bounds on causal effects when point identification is precluded by unobserved confounding. The approach uniquely integrates the conditional independence constraints implied by the causal graph into a sensitivity analysis framework, leveraging influence function projections and semiparametric estimation theory to substantially improve the statistical efficiency of bound estimation. Empirical evaluations on both simulated data and real-world applications—including the effect of job training on earnings and the impact of ejection fraction on heart failure mortality—demonstrate that the method achieves high efficiency and robustness under non-identifiable settings.

Abstract

Causal graphs may inform covariate adjustment for estimating causal effects and improve estimation efficiency by exploiting the graphical structure. In many applications, however, the target causal parameter may not be point-identified due to the presence of unmeasured confounding. Sensitivity analysis methods address this challenge by characterizing bounds on the causal parameter under varying assumptions about the magnitude or form of unmeasured confounding. We focus on semiparametric efficient estimation of causal effects in non-identifiable settings, assuming a known (or hypothesized) causal graph. We propose an influence function projection approach that exploits the conditional independence constraints implied by the graph to improve the efficiency of semiparametric estimators of upper and lower bounds on the average causal effect under a given sensitivity analysis model. Our approach applies across multiple sensitivity analysis frameworks and causal estimands, thereby connecting knowledge of graphical structure with the sensitivity analysis literature. We illustrate our approach through simulations and real data examples thought to be affected by unmeasured confounding, including the effect of labor training program on post-intervention earnings, and the effect of low ejection fraction on heart failure death.

Problem

• unmeasured confounding
• causal effects
• bounds estimation
• causal graphs
• sensitivity analysis

Innovation

• influence function projection
• conditional independence constraints
• semiparametric efficiency
• unmeasured confounding
• sensitivity analysis