Bayesian estimation of the intraclass correlation coefficients in multivariate mixed linear model

Abstract

Bayesian multivariate procedure is introduced to assess the intraclass correlation coefficients under the multivariate mixed linear model. Unified estimators for the multivariate measures are proposed as the conditional posterior means of multivariate inverted matrix Dirichlet distribution. These estimators have explicit expressions and generalize the results of case of variance components. A real data example is also provided.

Keywords:

• conditional Bayesian posterior
• jettreys prior
• wishart prior
• inerica matrix Dirichlet distribution
• multivariate measures ot correlations