Please use this identifier to cite or link to this item: http://repositorio.unicamp.br/jspui/handle/REPOSIP/201
Type: Artigo de periódico
Title: Assessing influence in survival data with a cure fraction and covariates
Author: ORTEGA, Edwin M. M.
CANCHO, Vicente G.
LACHOS, Victor Hugo
Abstract: Diagnostic methods have been an important tool in regression analysis to detect anomalies, such as departures from error assumptions and the presence of outliers and influential observations with the fitted models. Assuming censored data, we considered a classical analysis and Bayesian analysis assuming no informative priors for the parameters of the model with a cure fraction. A Bayesian approach was considered by using Markov Chain Monte Carlo Methods with Metropolis-Hasting algorithms steps to obtain the posterior summaries of interest. Some influence methods, such as the local influence, total local influence of an individual, local influence on predictions and generalized leverage were derived, analyzed and discussed in survival data with a cure fraction and covariates. The relevance of the approach was illustrated with a real data set, where it is shown that, by removing the most influential observations, the decision about which model best fits the data is changed.
Subject: Cure fraction
Bayesian inference
local influence
generalized leverage
survival data
Country: Espanha
Editor: INST ESTADISTICA CATALUNYA-IDESCAT
Citation: SORT-STATISTICS AND OPERATIONS RESEARCH TRANSACTIONS, v.32, n.2, p.115-139, 2008
Rights: aberto
Address: http://apps.isiknowledge.com/InboundService.do?Func=Frame&product=WOS&action=retrieve&SrcApp=EndNote&UT=000262690800001&Init=Yes&SrcAuth=ResearchSoft&mode=FullRecord
http://www.idescat.cat/sort/sort322/32.2.1.ortega-etal.pdf
Date Issue: 2008
Appears in Collections:IMECC - Artigos e Materiais de Revistas Científicas

Files in This Item:
File Description SizeFormat 
art_ORTEGA_Assessing_influence_in_survival_data_with_a_2008.pdfpublished version325.27 kBAdobe PDFView/Open


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.