A Bayesian approach for the zero-inflated cure model: an application in a Brazilian invasive cervical cancer database

Hayala Cristina Cavenague de Souza; Francisco Louzada; Pedro Luiz Ramos; Mauro Ribeiro de Oliveira Júnior; Gleici da Silva Castro Perdoná

doi:10.1080/02664763.2021.1933923

A Bayesian approach for the zero-inflated cure model: an application in a Brazilian invasive cervical cancer database

J Appl Stat. 2021 Jun 1;49(12):3178-3194. doi: 10.1080/02664763.2021.1933923. eCollection 2022.

Authors

Hayala Cristina Cavenague de Souza¹, Francisco Louzada², Pedro Luiz Ramos², Mauro Ribeiro de Oliveira Júnior^{3

4}, Gleici da Silva Castro Perdoná¹

Affiliations

¹ Department of Social Medicine, Ribeirão Preto School of Medicine, University of São Paulo, Ribeirão Preto, SP, Brazil.
² Institute of Mathematical Science and Computing, University of São Paulo, São Carlos, Brazil.
³ UNIEURO Centro Universitário, Brasília, Brazil.
⁴ Department of Specialized Risks, Caixa Econômica Federal, Brasília, Brazil.

Abstract

This paper aims to discuss the Bayesian estimation approach for the zero-inflated cure class of models, which extends the standard cure model by accommodating zero-inflated data in the survival analysis context. A comprehensive simulation study is carried out to assess the performance of the estimation procedure. A new estimation methodology is illustrated using a real dataset related to women diagnosed with invasive cervical cancer in Brazil.

Keywords: 62Nxx; Bayesian estimation; Weibull distribution; cure rate; survival analysis; zero-inflation.

Grants and funding

The authors' researchers are partially supported by the Brazilian's Institutions, CNPq, CAPES, and FAPESP. Pedro L. Ramos is grateful to the Fundação de Amparo à Pesquisa do Estado de São Paulo (FAPESP Proc. 2017/25971-0).