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

Autor(es)

Gallardo, Diego I. ^[1] ; Bolfarine, Heleno ^[2] ; Pedroso-de-Lima, Antonio Carlos ^[2]

Afiliação do(s) autor(es)

^[1] Univ Antofagasta, Fac Ciencias Basicas, Dept Matemat, Antofagasta - Chile

^[2] Univ Sao Paulo, Inst Matemat & Estatist, Sao Paulo - Brazil

• Tipo de documento: Artigo Científico
• Fonte: COMPUTATIONAL STATISTICS & DATA ANALYSIS; v. 98, p. 31-45, JUN 2016.
• Citações Web of Science: 1

Resumo

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. (C) 2015 Elsevier B.V. All rights reserved. (AU)

• Processo FAPESP: 13/23684-2 - Estendendo o modelo de cura Destrutivo Binomial Negativo com um esquema de ativação latente.
• Beneficiário: Diego Ignacio Gallardo Mateluna
• Modalidade de apoio: Bolsas no Brasil - Pós-Doutorado