Point Estimation for the Inverse Rayleigh Distribution under Type-II Left and Right Censoring

Uloženo v:
Podrobná bibliografie
Název: Point Estimation for the Inverse Rayleigh Distribution under Type-II Left and Right Censoring
Autoři: Sümeyra Sert, Coşkun Kuş
Zdroj: Volume: 46, Issue: 1179-184
Cumhuriyet Science Journal
Informace o vydavateli: Cumhuriyet University, 2025.
Rok vydání: 2025
Témata: Statistical Analysis, İstatistiksel Analiz, İstatistiksel Teori, Anderson-Darling statistic, Cramér-von mises, Left and right censoring, Kolmogorov-Smirnov statistics, Modified maximum likelihood estimation, Statistical Theory
Popis: The Inverse Rayleigh distribution is frequently utilized in reliability and survival analysis. This study focuses on deriving modified maximum likelihood estimators for the scale parameter of the Inverse Rayleigh distribution under Type-II left and right censoring. The efficacy of the proposed estimators is assessed through comparison with Anderson-Darling, Kolmogorov-Smirnov, and Cramér-von Mises type estimators via Monte Carlo simulations across various censoring schemes and parameter configurations. Additionally, a numerical example is presented to illustrate the proposed methodology. The simulation study demonstrates that the proposed estimators outperform the others. Additionally, given their explicit nature, the proposed estimators can serve as initial values for obtaining the maximum likelihood estimator.
Druh dokumentu: Article
Popis souboru: application/pdf
ISSN: 2587-2680
DOI: 10.17776/csj.1460135
Přístupová URL adresa: https://dergipark.org.tr/tr/pub/csj/issue/90982/1460135
Přístupové číslo: edsair.doi.dedup.....1f9ede4e309c3814bf684194aa38ea3f
Databáze: OpenAIRE
Popis
Abstrakt:The Inverse Rayleigh distribution is frequently utilized in reliability and survival analysis. This study focuses on deriving modified maximum likelihood estimators for the scale parameter of the Inverse Rayleigh distribution under Type-II left and right censoring. The efficacy of the proposed estimators is assessed through comparison with Anderson-Darling, Kolmogorov-Smirnov, and Cramér-von Mises type estimators via Monte Carlo simulations across various censoring schemes and parameter configurations. Additionally, a numerical example is presented to illustrate the proposed methodology. The simulation study demonstrates that the proposed estimators outperform the others. Additionally, given their explicit nature, the proposed estimators can serve as initial values for obtaining the maximum likelihood estimator.
ISSN:25872680
DOI:10.17776/csj.1460135