Enhanced Marginal Sensitivity Model and Bounds

2025-04-11

AI Summary

To address bias in causal effect estimation due to unmeasured confounding in observational studies, this paper proposes the enhanced marginal sensitivity model (eMSM), the first framework jointly bounding the influence of unmeasured confounders on both treatment assignment and potential outcomes. Compared with the conventional marginal sensitivity model (MSM), eMSM derives sharper (tighter) bounds for causal parameters via extremal optimization and establishes a doubly robust point estimator along with its asymptotically valid confidence intervals. The method integrates sensitivity analysis, double robustness, and asymptotic inference techniques. Experiments on two real-world datasets demonstrate that eMSM substantially narrows causal effect bounds, enhancing both inferential robustness and precision. This work provides a novel, more interpretable, and statistically rigorous framework for sensitivity analysis under unmeasured confounding.

Technology Category

• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Security and Privacy: Large-scale security measurements
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for incomplete, noisy, or partially observed Web-related graphs

Abstract

Sensitivity analysis is important to assess the impact of unmeasured confounding in causal inference from observational studies. The marginal sensitivity model (MSM) provides a useful approach in quantifying the influence of unmeasured confounders on treatment assignment and leading to tractable sharp bounds of common causal parameters. In this paper, to tighten MSM sharp bounds, we propose the enhanced MSM (eMSM) by incorporating another sensitivity constraint that quantifies the influence of unmeasured confounders on outcomes. We derive sharp population bounds of expected potential outcomes under eMSM, which are always narrower than the MSM sharp bounds in a simple and interpretable way. We further discuss desirable specifications of sensitivity parameters related to the outcome sensitivity constraint, and obtain both doubly robust point estimation and confidence intervals for the eMSM population bounds. The effectiveness of eMSM is also demonstrated numerically through two real-data applications. Our development represents, for the first time, a satisfactory extension of MSM to exploit both treatment and outcome sensitivity constraints on unmeasured confounding.

Problem

• Tightens bounds for unmeasured confounding in causal inference
• Incorporates sensitivity constraints on treatment and outcomes
• Provides narrower, interpretable bounds for potential outcomes

Innovation

• Enhanced MSM with outcome sensitivity constraint
• Narrower sharp bounds than original MSM
• Doubly robust estimation for eMSM bounds