|
Activity Number:
|
480
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 7, 2008 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Statistical Computing
|
| Abstract - #300606 |
|
Title:
|
A Robust Nonlinear Mixed-Effects Model Using the SAEM Algorithm
|
|
Author(s):
|
Cristian Meza*+ and Rolando de la Cruz-Mesia and Felipe Osorio
|
|
Companies:
|
Universidad de Valparaíso and Pontifícia Universidad Católica de Chile and Universidad de Valparaíso
|
|
Address:
|
Departamento de Estadística, Valparaíso, 2340000, Chile
|
|
Keywords:
|
Nonlinear mixed models ; SAEM algorithm ; Robust model ; Repeated measures ; Longitudinal data ; Outliers
|
|
Abstract:
|
The nonlinear mixed-effects models are very useful to analyze repeated measures data. Usually, it is assumed normal distributions for the random effects and the residuals, but such assumptions make inferences vulnerable to the presence of outliers. We introduce an extension of normal nonlinear mixed-effects model considering a subclass of elliptical contoured distributions for both random effects and residual errors. This elliptical subclass, the scale mixtures of normal (SMN) distributions, includes heavy tailed multivariate distributions, such as Student-t and contaminated normal among others. We propose an exact estimation procedure to obtain the maximum likelihood estimates of the fixed effects and variance components using a stochastic approximation of the EM algorithm. We compare the performance of the normal and the SMN models and it is shown our model allows to identify outliers.
|
