Seminário
UFSCar/USP ? 23/09/2011 - 14h00
LOCAL: Sala de seminários
DEs-UFSCar TITLE: Approximate inferences for skew-normal independent nonlinear
mixed-effects models with application to AIDS studies PALESTRANTE: Prof. Dr. Victor
Hugo Lachos Davila ? IMECC-UNICAMP ABSTRACT: Nonlinear mixed-effects models have received a great deal of
attention in the statistical literature in recent years because of the
flexibility in handling longitudinal studies, including human immunodeficiency
virus viral dynamics, pharmacokinetic analyses, and studies of growth and decay.
A standard assumption in nonlinear mixed-effects models for continuous responses
is the normal distribution for the random effects and the within-subject errors,
making it sensitive to outliers. We present a novel class of asymmetric
nonlinear mixed-effects models that provides for an efficient estimation of the
parameters in the analysis of longitudinal data. We assume that, marginally, the
random effects follow a multivariate skew?normal/independent distribution and
that the random errors follow a symmetric normal/independent distribution
providing an appealing robust alternative to the usual normal distribution in
nonlinear mixed-effects models. We propose an approximate likelihood analysis
for maximum likelihood estimation based on the EM-type algorithm that produce
accurate maximum likelihood estimates and significantly reduces the numerical
difficulty associated with the exact maximum likelihood estimation. Techniques
for prediction of future responses under this class of distributions are also
briefly investigated. Simulation studies indicate that our proposed methods work
well for small, medium and large variability of the random effects. The newly
developed procedures are illustrated with a HIV case study that was initially
analyzed using normal nonlinear mixed-effects models. Keywords and phrases: Approximate likelihood; EM?algorithm; HIV dynamics; Nonlinear mixed
effects models; Linearization; Skew?normal/independent
distributions. |