Abstract:
|
Probabilistic sensitivity analysis using either Monte Carlo (MCSA) or Bayesian sensitivity analysis (BSA) has emerged as a popular technique to examine unmeasured confounding. BSA uses Bayes theorem to formally combine evidence from the prior distribution and the data. In contrast, MCSA samples bias parameters directly from the prior distribution. Intuitively, one would think that BSA and MCSA ought to give similar results. Both methods use similar models and the same (prior) probability distributions for the bias parameters. In this talk, we illustrate the surprising finding that BSA and MCSA can give very different results. We show that certain combinations of data and prior distributions can result in dramatic prior-to-posterior changes in uncertainty about the bias parameters. This occurs because the application of Bayes theorem in a non-identifiable model can sometimes rule out certain patterns of unmeasured confounding that are not compatible with the data. Consequently, the MCSA approach may give 95% intervals that are either too wide or too narrow, and that do not have 95% frequentist coverage probability. We illustrate with a data example from epidemiology.
|