Bayesian Prediction of the First Order Statistics from Frecht Distribution When Sample Size is a Random Variable (A Numerical Investigation) | ||||
The Egyptian Statistical Journal | ||||
Article 8, Volume 44, Issue 2, December 2000, Page 219-227 | ||||
Document Type: Original Article | ||||
DOI: 10.21608/esju.2000.313836 | ||||
View on SCiNiTO | ||||
Abstract | ||||
This paper considers Bayesian prediction of order statistics when the sample size of the future sample is a random variable. It has been shown (see Lingappaiah (1986)) that this situation arises in the field as Biology Agriculture and in Quality Control. Under the assumption that the underlying distribution for the old and future samples is a Type II Extreme value (Frecht) distribution, Bayesian prediction analysis for the first future order statistics will be investigated. A numerical comparison is made with the case when the sample size is fixed and random variable assuming that the random sample sizes have a Poisson and Negative Binomial distribution. These comparisons are made under basis of relative efficiency, sampling distribution for the first future order statistics will be obtained numerically in both fixed and random sample size cases. The results of Monte Carlo Simulation study are presented. | ||||
Keywords | ||||
Bayesian Prediction; Poisson and Negative Binomial Distribution; Sampling Distribution; Monte Carlo Simulation | ||||
Statistics Article View: 21 |
||||