On Optimality Properties of the Power Prior

摘要: The power prior is a useful general class of priors that can be used for arbitrary classes regression models, including generalized linear mixed models semiparametric survival with censored data, frailty multivariate and nonlinear models. specification the coefficients focuses on observable quantities in elicitation based historical D0, scalar quantity, a0, quantifying heterogeneity between current D, data D0. distribution then constructed by raising likelihood function to where 0 ≤ a0 1. precision parameter viewed as measure compatibility data. In this article we give formal justification show it an optimal informative sense minimizes convex sum Kullback-Leibler (K...