Inference for multivariate normal hierarchical models

Journal of the Royal Statistical Society: Series B (Statistical Methodology) · 2000 · Volume 62 · Issue 2

Abstract

This paper provides a new method and algorithm for making inferences about the parameters of a two-level multivariate normal hierarchical model. One has observedJ p-dimensional vector outcomes, distributed at level 1 as multivariate normal with unknown mean vectors and with known covariance matrices. At level 2, the unknown mean vectors also have normal distributions, with common unknown covariance matrixAand with means depending on known covariates and on unknown regression coefficients. The algorithm samples independently from the marginal posterior distribution ofAby using rejection procedures. Functions such as posterior means and covariances of the level 1 mean vectors and of the level 2 regression coefficient are estimated by averaging over posterior values calculated conditionally on each value ofAdrawn. This estimation accounts for the uncertainty inA, unlike standard restricted maximum likelihood empirical Bayes procedures. It is based on independent draws from the exact posterior distributions, unlike Gibbs sampling. The procedure is demonstrated for profiling hospitals based on patients' responses concerningp=2 types of problems (non-surgical and surgical). The frequency operating characteristics of the rule corresponding to a particular vague multivariate prior distribution are shown via simulation to achieve their nominal values in that setting.

Keywords

Constrained Wishart distributions, Importance weighting, Interval estimates, Medical profiling, Multivariate empirical Bayes procedures, Rejection sampling, Restricted maximum

Enable JavaScript for citation tools and additional paper interactions.