ABSTRACT
The third step in Bayesian analysis is to evaluate the model fit to the data and determine
the sensitivity of the posterior distribution to the assumptions. This is typically an interac-
tive and iterative process in which different approaches are taken until there is reasonable
evidence that the conclusions are both statistically reliable and stable under modest changes
in the assumptions. So the emphasis in this chapter is on model adequacy: the suitability of
a single model under consideration. The subsequent chapter is really about model testing:
determining why we should prefer one model specification over another. Both of these are
critical considerations, and it is important to keep them distinct.