• 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
 
 
Doctoral Thesis
DOI
https://doi.org/10.11606/T.45.1999.tde-20210729-023455
Document
Author
Full name
Vicente Garibay Cancho
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 1999
Supervisor
Title in Portuguese
Métodos de Monte Carlo em análise de sobrevivência
Keywords in Portuguese
Inferência Estatística
Abstract in Portuguese
O principal objetivo deste trabalho é desenvolver técnicas Monte Carlo via cadeias de Markov para alguns modelos especiais considerados na teoria da confiabilidade e na análise de sobrevivência. Uma metodologia Bayesiana é desenvolvida para omodelo Weibull-exponenciada proposto por Mudholkar et al. (1995). Uma extensão deste modelo é também considerada adicionando covariáveis ao modelo originalmente proposto por Mudholkar et al. (1995). Reportamos um estudo comparativo do modeloproposto por Mudholkar et al. (1995) com várias famílias de distribuições usadas no ajuste de dados de tempos de vida com função de taxas de falha do tipo 'bathtub'. Finalmente desenvolvemos um procedimento de análise Bayesiana para o modelo demistura exponencial com ou sem covariáveis e, comparamos tal análise com os procedimentos clássicos de análise. Consideramos a distribuição Weibull-exponenciada para o ajuste de dados de tempo de vida na presença de indivíduos imunes e estudamoso uso deste modelo no teste de adequação do modelo de mistura Weibull e exponencial considerando as abordagens clássica e Bayesiana
Title in English
not available
Abstract in English
The main object of this study is to develop Markov chain Monte Carlo techniques in some special models typically considered in reliability theory and survival analysis. A Bayesian methodology is developed for the exponentiated-Weibullmodelproposed in Mudholkar et al. (1995). An extension of this model is also considered by adding covariates to the model originally proposed in Mudholkar et al. (1995). Are of the special features of the exponentiated-Weibull model is that itallows fitting bathtub failure rate models to data sets. Numerical studies are reported comparing several methods for fitting bathtub models to data. Next, we consider a Bayesian analysis for the exponential and Weibull mixture models forincorporatinglong-term survival data to the model. Comparisons with the classical maximum likelihood methodology are reported and extensions considered by using the exponentiated-Weibull model in this situation
 
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
2021-07-29
 
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.