Please use this identifier to cite or link to this item: http://hdl.handle.net/123456789/650
DC FieldValueLanguage
dc.contributor.authorHeba A. Al-Muzainien_US
dc.date.accessioned2019-03-06T06:02:50Z-
dc.date.available2019-03-06T06:02:50Z-
dc.date.issued2017en_US
dc.identifier.urihttp://hdl.handle.net/123456789/650-
dc.description.abstractIn this thesis, different methods of prediction of future record statistics given the knowledge about some observed record statistics from the two-parameter exponential distribution were reviewed. The predictors discussed here are the best linear unbiased, best linear invariant, maximum likelihood, conditional median and median unbiased predictors. Comprehensive comparison studies for both one-sample and two-sample prediction problems were conducted to assess how much closer does one predictor get to another predictor in terms of Pitman’s measure of closeness. Then all these predictors were compared with respect to the median unbiased predictor in terms of Pitman’s closeness. Numerical computations for Pitman’s closeness of all predictors were performed, presented and discussed to examine how these predictors are compared. The performances of all these predictors were described and summarized in the sense of both optimality criteria: Pitman’s closeness and mean square prediction error.en_US
dc.publisher Kuwait university - college of graduate studiesen_US
dc.subjectExponential Record : Comparativeen_US
dc.titleComparative Study for Assessing the Pitman’s Closeness of Predictors of Exponential Record Statisticsen_US
dc.typethesisen_US
dc.contributor.supervisorProf. Mohammad Z. Al-Raqaben_US
dc.contributor.universityID214127387en_US
dc.contributor.emailthesis.feedback@grad.ku.edu.kwen_US
dc.description.conclusionsFirst, we summarize all PC comparisons among all predictors in Table 4.12. For completeness, the MSPE as another optimality criterion is considered here to assess the performances of these predictors of future record of 𝑌𝑈(𝑠) from 𝑌-sample based on 𝑋-sample. The values of MSPEs of all predictors addressed in Chapter 2 are presented in Table 4.13. From the tables, we can conclude the following points:  Generally, under PMC, the BLUP competes the BLIP and MLP well for small values of 𝑟 or large values of 𝑟 when 𝑠 being large. The opposite conclusion may be observed when one wishes to predict initial future record statistics from 𝑌-sample based on 𝑋-sample with large observed record statistics (large 𝑟). When comparing with MLP, the BLIP is preferred to MLP except for the cases when 𝑠=2 and 𝑟 is large.en_US
dc.date.semesterSpringen_US
dc.description.examinationYen_US
dc.description.gpa3.96en_US
dc.description.credits38en_US
Appears in Programs:0480 Statistics & Operations Research
Files in This Item:
File Description SizeFormat 
Heba Thesis.pdf2,25 MBAdobe PDFView/Open    Request a copy
Show simple item record

Page view(s)

15
Last Week
0
Last month
checked on Jan 25, 2020

Download(s)

5
checked on Jan 25, 2020

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.