Sensitivity Analysis for Unobserved Confounding
How to know the unknowable in observational studiesOutlineIntroductionProblem Setup2.1. Causal Graph2.2. Model With and Without Z2.3. Strength of Z as a ConfounderSensitivity Analysis3.1. Goal3.2. Robustness ValuePySensemakrConclusionAcknowledgementsReferences1. IntroductionThe specter of unobserved confounding (aka omitted variable bias) is a notorious problem in observational studies. In most observational studies, unless we can reasonably assume that treatment assignment is as-if random as in a natural experiment, we…