Bayesian parsimonious covariance estimation for hierarchical linear mixed models

Frühwirth-Schnatter, Sylvia and Tüchler, Regina (2004) Bayesian parsimonious covariance estimation for hierarchical linear mixed models. Research Report Series / Department of Statistics and Mathematics, 11. Institut für Statistik und Mathematik, WU Vienna University of Economics and Business, Vienna.


Download (314kB)


We considered a non-centered parameterization of the standard random-effects model, which is based on the Cholesky decomposition of the variance-covariance matrix. The regression type structure of the non-centered parameterization allows to choose a simple, conditionally conjugate normal prior on the Cholesky factor. Based on the non-centered parameterization, we search for a parsimonious variance-covariance matrix by identifying the non-zero elements of the Cholesky factors using Bayesian variable selection methods. With this method we are able to learn from the data for each effect, whether it is random or not, and whether covariances among random effects are zero or not. An application in marketing shows a substantial reduction of the number of free elements of the variance-covariance matrix. (author's abstract)

Item Type: Paper
Keywords: covariance selection / random effects models / Markov Chain Monte Carlo / fractional prior / variable selection
Divisions: Departments > Finance, Accounting and Statistics > Statistics and Mathematics
Depositing User: Repository Administrator
Date Deposited: 18 Jan 2005 12:54
Last Modified: 22 Oct 2019 00:41


View Item View Item


Downloads per month over past year

View more statistics