Bayesian model comparison and model averaging for small-area estimation

摘要

This paper considers small-area estimation with lung cancer mortality data,and discusses the choice of upper-level model for the variation over areas.Inference about the random effects for the areas may depend strongly on thechoice of this model, but this choice is not a straightforward matter. We givea general methodology for both evaluating the data evidence for differentmodels and averaging over plausible models to give robust area effectdistributions. We reanalyze the data of Tsutakawa [Biometrics 41 (1985) 69--79]on lung cancer mortality rates in Missouri cities, and show the differences inconclusions about the city rates from this methodology.