Parameter estimation of inverse Lindley distribution for Type-I censored data |
| |
Authors: | Suparna Basu Sanjay Kumar Singh Umesh Singh |
| |
Affiliation: | 1.Department of Statistics,University of Burdwan,Burdwan,India;2.Department of Statistics, Institute of Science,Banaras Hindu University,Varanasi,India;3.DST-CIMS,Banaras Hindu University,Varanasi,India |
| |
Abstract: | In life testing experiments, Type-I censoring scheme has been widely used due to its simplicity and poise with considerable gain in the completion time of an experiment. This article deals with the parameter estimation of inverse Lindley distribution when the data is Type-I censored. Estimates have been obtained under both the classical and Bayesian paradigm. In the classical scenario, estimates based on maximum likelihood and maximum product of spacings coupled with their 95% asymptotic confidence interval have been obtained. Under the Bayesian set up, the point estimate is obtained by considering squared error loss function using Markov Chain Monte Carlo technique and highest posterior density intervals based on these samples are reckoned. The performance of above mentioned techniques are evaluated on the basis of their simulated risks. Further, a real data set is analysed for appraisal of aforementioned estimation techniques under the specified censoring scheme. |
| |
Keywords: | |
本文献已被 SpringerLink 等数据库收录! |
|