Bayesian estimation strategy for multi-component geometric life testing model under doubly type-1 censoring scheme

This study develops a Bayesian approach for estimating the unknown parameters of the 3-component mixture of geometric (3-CMG) model under a doubly type-I censoring scheme (DT1CS). The derivations of the Bayes estimators (BEs) and Bayes risks (BRs) are presented under square error loss function (SELF), precautionary loss function (PLF) and DeGroot loss function (DLF) using Beta prior under DT1CS. The strategy is evaluated through extensive simulation and real-life data analysis, showing the strength and efficiency of the newly proposed model. The study recommends that the SELF is the optimal choice for accurately estimating the unknown parameters of the 3-CMG model. © 2024 The Author(s)