Inference for the trivariate Marshall-Olkin-Weibull distribution in presence of right-censored data

Ricardo Puziol De Oliveira*, Marcos Vinicius de Oliveira Peres, Jorge Alberto Achcar, Nasser Davarzani

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

Abstract

Multivariate lifetime data are common in many applications, especially in medical and engineering studies. In this paper, we consider a trivariate Marshall-Olkin-Weibull distribution to model trivariate data in presence of right censored data.Maximum likelihood and Bayesian methods are used to get the parameter estimators of interest. An extensive simulation study was performed to verify the effectiveness of the maximum likelihood estimators. Reliability data sets related to fiber failure strengths were considered to illustrate the performance of the proposed model under the classical and Bayesian approaches. As a result, note that the trivariate Marshall-Olkin-Weibull model could be considered as a good alternative to model trivariate lifetime data, especially under a Bayesian approach which could be of interest for the reliability analysis, as observed with the real data application in industrial engineering presented in the study or any other area of interest.
Original languageEnglish
Pages (from-to)95-116
Number of pages22
JournalChilean Journal of Statistics
Volume11
Issue number2
Publication statusPublished - 1 Dec 2020

Keywords

  • Bayesian approach
  • Censored data
  • Maximum likelihood method
  • Monte Carlo simulation
  • Multivariate distributions

Fingerprint

Dive into the research topics of 'Inference for the trivariate Marshall-Olkin-Weibull distribution in presence of right-censored data'. Together they form a unique fingerprint.

Cite this