Bayesian joint modelling of the mean and covariance structures for normal longitudinal data

We consider the joint modelling of the mean and covariance structures for the general antedependence model, estimating their parameters and the innovation variances in a longitudinal data context. We propose a new and computationally efficient classic estimation method based on the Fisher scoring al...

Descripción completa

Detalles Bibliográficos
Autores: Cepeda-Cuervo, Edilberto, Núñez-Antón, Vicente
Tipo de recurso: artículo
Fecha de publicación:2007
País:España
Institución:Universitat Politècnica de Catalunya (UPC)
Repositorio:UPCommons. Portal del coneixement obert de la UPC
Idioma:inglés
OAI Identifier:oai:upcommons.upc.edu:2099/8917
Acceso en línea:https://hdl.handle.net/2099/8917
Access Level:acceso abierto
Palabra clave:Mathematical statistics
Antedependence models
Bayes estimation Fisher scoring
Gibbs sampling
Fisher scoring
Estadística matemàtica
Estadística matemàtica -- Aplicacions
Classificació AMS::62 Statistics::62F Parametric inference
Classificació AMS::62 Statistics::62P Applications
Àrees temàtiques de la UPC::Matemàtiques i estadística::Estadística matemàtica
Àrees temàtiques de la UPC::Matemàtiques i estadística::Estadística aplicada
Descripción
Sumario:We consider the joint modelling of the mean and covariance structures for the general antedependence model, estimating their parameters and the innovation variances in a longitudinal data context. We propose a new and computationally efficient classic estimation method based on the Fisher scoring algorithm to obtain the maximum likelihood estimates of the parameters. In addition, we also propose a new and innovative Bayesian methodology based on the Gibbs sampling, properly adapted for longitudinal data analysis, a methodology that considers linear mean structures and unrestricted covariance structures for normal longitudinal data. We illustrate the proposed methodology and study its strengths and weaknesses by analyzing two examples, the race and the cattle data sets.