B1173
Title: Sensitivity analysis for principal ignorability violation in estimating complier and noncomplier average causal effects
Authors: Trang Nguyen - Johns Hopkins Bloomberg School of Public Health (United States) [presenting]
Abstract: An important strategy for identifying principal causal effects, often used in noncompliance settings, is invoking the principal ignorability (PI) assumption. As PI is untestable, it is important to gauge how sensitive effect estimates are to its violation. The focus is on this task for the common one-sided noncompliance setting where there are two principal strata, compliers and noncompliers. Under PI, compliers and noncompliers share the same outcome-mean-given-covariates function under the control condition. For sensitivity analysis, the function is allowed to differ between compliers and noncompliers in several ways, indexed by an odds ratio, a generalized odds ratio, a mean ratio, or a standardized mean difference sensitivity parameter. Sensitivity analysis techniques (with any sensitivity parameter choice) are tailored to several types of PI-based main analysis methods, including outcome regression, influence function-based and weighting methods. The proposed sensitivity analyses are illustrated using several outcome types from the JOBS II study, and code in the R-package PIsens is provided.
