A Bayesian Cure Rate Model Based on the Power Piecewise Exponential Distribution

Autor(es)

de Castro, Mario ^[1] ; Gomez, Yolanda M. ^[2]

Afiliação do(s) autor(es)

^[1] Univ Sao Paulo, Inst Ciencias Matemat & Comp, Sao Carlos, SP - Brazil

^[2] Univ Atacama, Fac Ingn, Dept Matemat, Copiapo - Chile

• Tipo de documento: Artigo Científico
• Fonte: METHODOLOGY AND COMPUTING IN APPLIED PROBABILITY; v. 22, n. 2, p. 677-692, JUN 2020.
• Citações Web of Science: 0

Resumo

Cure rate models have been used in a number of fields. These models are applied to analyze survival data when the population has a proportion of subjects insusceptible to the event of interest. In this paper, we propose a new cure rate survival model formulated under a competing risks setup. The number of competing causes follows the negative binomial distribution, while for the latent times we posit the power piecewise exponential distribution. Samples from the posterior distribution are drawn through MCMC methods. Some properties of the estimators are assessed in a simulation study. A dataset on cutaneous melanoma is analyzed using the proposed model as well as some existing models for the sake of comparison. (AU)

• Processo FAPESP: 13/07375-0 - CeMEAI - Centro de Ciências Matemáticas Aplicadas à Indústria
• Beneficiário: Francisco Louzada Neto
• Modalidade de apoio: Auxílio à Pesquisa - Centros de Pesquisa, Inovação e Difusão - CEPIDs