Diagnostic Plots for Improved Parameterization in Bayesian Inference
Susan E. Hills, A. F. M. Smith · Biometrika · 1993
A major problem with the Bayesian analysis of statistical models is that the computation of posterior and predictive summaries typically requires the numerical evaluation of complicated multivariate integrals. The accuracy and efficiency of these computations are usually sensitive to the choice of parameterization for the statistical model. This paper describes a graphical method which can be used both for assessing a proposed parameterization and choosing a new improved parameterization.