Destructive weighted Poisson cure rate models with bivariate random effects: Classical and Bayesian approaches

详细信息    查看全文

• 作者：Diego I. Gallardo^a ; ^{diego.gallardo@uantof.cl" class="auth_mail" title="E-mail the corresponding author} ; Heleno Bolfarine^b ; ^{hbolfar@ime.usp.br" class="auth_mail" title="E-mail the corresponding author} ; Antonio Carlos Pedroso-de-Lima^b ; ^{acarlos@ime.usp.br" class="auth_mail" title="E-mail the corresponding author}
• 关键词：REML ; Dirichlet process ; Competing risks
• 刊名：Computational Statistics & Data Analysis
• 出版年：2016
• 出版时间：June 2016
• 年：2016
• 卷：98
• 期：Complete
• 页码：31-45
• 全文大小：567 K

文摘

In this paper, random effects are included in the destructive weighted Poisson cure rate model. For parameter estimation we implemented a classical approach based on the restricted maximum likelihood (REML) methodology and a Bayesian approach based on Dirichlet process priors. A small scale simulation study is conducted to discuss parameter recovery and the performance of the proposed methodology is illustrated with a real data example.