Please use this identifier to cite or link to this item:
https://research.matf.bg.ac.rs/handle/123456789/1859
Title: | O raspodeli verovanja parametara pouzdanosti dvokomponentnog sistema sa nezavisnim eksponencijalnim raspodelama | Other Titles: | On a fiducial distribution of the reliability parameter of a two-component system with independent exponential distribution | Authors: | Lukić, Žikica Milošević, Bojana Obradović, Marko |
Affiliations: | Probability and Mathematical Statistics Probability and Mathematical Statistics Probability and Mathematical Statistics |
Issue Date: | 2021 | Rank: | M34 | Publisher: | Beograd : Matematički fakultet | Related Publication(s): | Zbornik radova - XLVIII Simpozijum o operacionim istraživanjima = XLVIII International Symposium on Operational Research, SYM-OP-IS 2021 | Conference: | Simpozijum o operacionim istraživanjima SYM-OP-IS(48 ; 2021 ; Banja Koviljača) | Abstract: | Parametar pouzdanosti dvokomponentnog sistema R = P{X > Y }, kod koga X označava snagu sistema, a Y stres, predstavlja verovatnoću da će sistem opstati. U radu je izvedena raspodela verovanja (engl. fiducial distribution) parametra R = P{X > Y }, pri čemu su X i Y nezavisne slučajne veličine sa eksponencijalnim E(λ) i E(μ) raspodelama. Dobijeni teorijski rezultati su ilustrovani empirijskim ocenama raspodele verovanja za uzorke različitog obima, i različite vrednosti parametara λ i μ. Dodatno, predložen pristup je upoređen sa frekvencionističkim i Bajesovim ocenjivanjem parametra R. Reliability parameter of a two-component system R=P{X>Y}, where X denotes the strength of the system and Y denotes the stress of the system represents probability that the system will prevail. Fiducial distribution of the parameter R=P{X>Y} is derived in this paper in a case that X and Y are independent random variables with exponential \mathcalE(λ) and \mathcalE(μ) distributions respectively. Theoretical results obtained in the paper are illustrated with empirical estimates of fiducial distributions for different sample sizes and different values of parameters and µ. The suggested approach was compared with frequentist and Bayes estimates of the parameter R. |
URI: | https://research.matf.bg.ac.rs/handle/123456789/1859 |
Appears in Collections: | Research outputs |
Show full item record
Google ScholarTM
Check
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.