• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Master's Dissertation
DOI
10.11606/D.55.2018.tde-06032018-113301
Document
Author
Full name
Josenildo de Souza Chaves
Institute/School/College
Knowledge Area
Date of Defense
Published
São Carlos, 1999
Supervisor
Committee
Rodrigues, Josemar (President)
Achcar, Jorge Alberto
Leite, Jose Galvao
Title in Portuguese
Inferência Bayesiana para Dados Clínicos Exponenciais com Variáveis Auxiliares
Keywords in Portuguese
Não disponível
Abstract in Portuguese
Apresentamos neste trabalho, uma análise bayesiana para dados clínicos exponenciais com variáveis auxiliares. Formulamos uma abordagem bayesiana com densidades a priori informativas, obtidas através das variáveis auxiliares sob o contexto de modelos lineares generalizados, para estimar os parâmetros de interesse, testar o modelo e prever a sobrevivência de pacientes com doenças graves. Diferentes funções de ligações são consideradas. O método que iremos examinar consiste na obtenção de informações a priori para a média das respostas, com correspondentes variáveis auxiliares fixas de modo que se possa induzir uma distribuição a priori sobre os coeficientes de regressão a partir de médias condicionais a priori. Esta abordagem utiliza os algoritmos computacionais do tipo Gibbs Sampling/Metropolis-Hastings e será comparada com a inferência bayesiana exata. Finalizamos com aplicações em dados clínicos exponenciais para pacientes com leucemia utilizando amostras completas e amostras censuradas.
Title in English
Not available
Keywords in English
Not available
Abstract in English
In this work, a Bayesian analysis for the exponential clinica] data with auxiliary variables is presented. This Bayesian approach, with informative priors obtained under the context of generalized linear models with fixed auxiliary variables, is formulated to estimate the parameters of interest, to test models and to predict the survival time of patients with serious diseases. Various link functions are considered. The method that we are going to study consists in obtaining prior information for the response mean corresponding to observable variables with fixed covariates, such that we are able to induce a prior distribution on the regression coefficients. This approach uses Gibbs Sampling/Metropolis-Hastings algorithms and it will be compared with the exact Bayesian inference. We end with applications based on censored and uncensored exponential clinica] data for patients with leukemia diseases.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2018-03-06
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2022. All rights reserved.