About the journal

Cobiss

Thermal Science 2023 Volume 27, Issue 3 Part A, Pages: 2091-2098
https://doi.org/10.2298/TSCI2303091W
Full text ( 524 KB)


Bayesian inference for a simple step-stress accelerated dependent competing failure model

Wang Ying (Science College, Inner Mongolia University of Technology, Hohhot, China)
Yan Zai-Zai (Science College, Inner Mongolia University of Technology, Hohhot, China), zz.yan@163.com

The Copula approach can be used to describe the dependence structure between variables. In this paper, by using a Bivariate Clayton copula, we discuss the statistical analysis of a simple step-stress accelerated dependent competing failure model under progressively Type-II censoring sample. With the assumption of cumulative exposure, the Bayesian estimations of the model parameters are derived. Based on Monte-Carlo simulation, the precision of the estimates is assessed. Finally, the statistical analysis of an actual data set of a solar lighting insulation system has been presented for illustrative purposes.

Keywords: step-stress accelerated life test, dependent competing risk model, cumulative exposure model, bivariate clayton copula, Bayesian estimations


Show references