[MaPhySto logo]
MaPhySto
Centre for Mathematical Physics and Stochastics
Department of Mathematical Sciences, University of Aarhus

Funded by The Danish National Research Foundation

MPS-RR 2002-1
February 2002




Geometric ergodicity of Metropolis-Hastings algorithms for conditional simulation in generalised linear mixed models

by:

Jesper Møller

O.F. Christensen, R.P. Waagepetersen

Abstract

Conditional simulation is useful in connection with inference and prediction for a generalised linear mixed model. We consider random walk Metropolis and Langevin-Hastings algorithms for simulating the random effects given the observed data, when the joint distribution of the unobserved random effects is multivariate Gaussian. In particular we study the desirable property of geometric ergodicity, which ensures the validity of central limit theorems for Monte Carlo estimates.

Availability: [ gzipped ps-file ] [ pdf-file ]

[ Help on down-loading/viewing/printing ]


This paper has now been published in Methodology and Computing in Applied Probability 3, 309 - 327 (2001)