An Improved Conditional Inference on General Lifetime Model Parameters

Abstract

M. Maswadah

It is widely known that the conditional inference is usually efficient as the Bayesian inference based on the non-in- formative prior. However, it is less efficient than the Bayesian inference based on the informative prior. Therefore, the main objective of this paper is to introduce an improvement to the conditional inference by using the kernel prior distribution. The improved conditional inference has been used for estimating the general lifetime model parameters, based on the generalized progressive hybrid-censoring scheme, and compared with the Bayesian estimates, via the Monte Carlo simulations. The simulation results have been shown that the improved conditional inference is highly efficient and provides better estimates than the Bayesian estimates based on different loss functions. Finally, real data sets have been given to demonstrate the efficiencies of the proposed methods.

PDF