A0509
Title: Enhanced marginal sensitivity model and bounds
Authors: Yi Zhang - Rutgers University (United States) [presenting]
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. To tighten MSM sharp bounds, the enhanced MSM (eMSM) is proposed by incorporating another sensitivity constraint, which quantifies the influence of unmeasured confounders on outcomes. Sharp population bounds of expected potential outcomes are derived under eMSM, which are always narrower than the MSM sharp bounds in a simple and interpretable way. Desirable specifications of sensitivity parameters are further discussed, related to the outcome sensitivity constraint, and both doubly robust point estimation and confidence intervals are obtained for the eMSM population bounds. The effectiveness of eMSM is also demonstrated numerically through two real-data applications. The development represents, for the first time, a satisfactory extension of MSM to exploit both treatment and outcome sensitivity constraints on unmeasured confounding.
