A simulated pseudo-maximum likelihood estimator for nonlinear mixed models

Authors:
Didier Concordet;Olivier G. Nunez
Affiliations:
Ecole Vétérinaire de Toulouse, Unité Associée INRA de Physiopathologie et Toxicologie, Expérimentales, 23 Chemin des Capelles, Toulouse Cedex, France;Laboratoire de Statistique et Probabilités, Université Paul Sabatier, 118 Rte de Narbonne, Toulouse Cedex, France
Venue:
Computational Statistics & Data Analysis
Year:
2002

Citing 0
Cited 3

A parameter expansion version of the SAEM algorithm

Statistics and Computing
A new method for the estimation of variance matrix with prescribed zeros in nonlinear mixed effects models

Statistics and Computing
Maximum likelihood estimation in nonlinear mixed effects models

Computational Statistics & Data Analysis

Quantified Score

Hi-index	0.03

Visualization

Abstract

An estimator for parameters of nonlinear mixed effects models obtained by maximisation of a simulated pseudo-likelihood, is proposed. This simulated criterion is constructed from the likelihood of a Gaussian model whose means and variances are given by Monte-Carlo approximations of means and variances of the true model. The main advantage of the obtained estimator is that its asymptotic properties are known when the number of observations per individual is finite. The performance of this estimator is compared to existing estimators using simulations.