:: Volume 2, Issue 1 (9-2005) ::
JSRI 2005, 2(1): 39-52 Back to browse issues page
Distribution Free Confidence Intervals for Quantiles Based on Extreme Order Statistics in a Multi-Sampling Plan
M. Razmkhah *1, J. Ahmadi, B. Khatib
1- , razmkhah_m@yahoo.com
Abstract:   (2510 Views)

Extended Abstract. Let Xi1 ,..., Xini   ,i=1,2,3,....,k  be independent random samples from distribution $F^{alpha_i}$،  i=1,...,k, where F is an absolutely continuous distribution function and $alpha_i>0$ Also, suppose that these samples are independent. Let Mi,ni and  M'i,ni  respectively, denote the maximum and minimum of the ith sample. Constructing the distribution-free confidence intervals for quantiles of F based on these informations is the aim of this paper. Various cases have been studied and in each case, the exact non-parametric confidence intervals are obtained. First, we concentrate our attention to the maxima of the samples. Coverage probability of a confidence interval based on two different... [To Continue click here]

Keywords: Extreme order statistics, confidence interval, quantile, coverage probability, model, reversed hazard rate function.
Full-Text [PDF 396 kb]   (601 Downloads)    
Type of Study: Research | Subject: General
Received: 2016/02/9 | Accepted: 2016/02/9 | Published: 2016/02/9

XML   Persian Abstract   Print

Volume 2, Issue 1 (9-2005) Back to browse issues page