Scholarlink

<p>This paper proposes a new Ranked Set Sampling procedure using the Highest order statistics (RSSH) to estimate the population mean of Weibull distribution. Further, the estimator based on the Ranked Set Sampling using Lowest Order Statistic (RSSL) procedure is also developed for Weibull distribution and its efficiency has been compared with the proposed estimator over varying skewness. The expressions for Bias and MSE for both the estimators have been derived. Various forms of Weibull distributions based on positive and negative skewness are taken into consideration for efficiency comparison. The efficiency comparison is conducted in two scenarios, known and unknown parameters, wherein the unknown parameters are estimated using the Maximum Likelihood Estimation method. The analysis illustrates that the gains in the relative precisions of the estimators, based on both RSSL and the proposed RSSH method, consistently exceed the estimators based on Simple Random Sampling (SRS), Ranked Set Sampling (RSS) and Extreme Ranked Set Sampling(ERSS) procedures in the case of skewed populations. Simulation studies carried over randomly generated populations and an application of the proposed methods to a time-to-event data validated the theoretical results.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Extreme order statistics based inference for Weibull distribution under Ranked Set Sampling with application to time-to-event data

  • 期刊论文

  • 发表日期：

    2025年12月8日

查看全文
  • Tanushree Yadav,
  • Girish Chandra,
  • Priyanka Singh

摘要

This paper proposes a new Ranked Set Sampling procedure using the Highest order statistics (RSSH) to estimate the population mean of Weibull distribution. Further, the estimator based on the Ranked Set Sampling using Lowest Order Statistic (RSSL) procedure is also developed for Weibull distribution and its efficiency has been compared with the proposed estimator over varying skewness. The expressions for Bias and MSE for both the estimators have been derived. Various forms of Weibull distributions based on positive and negative skewness are taken into consideration for efficiency comparison. The efficiency comparison is conducted in two scenarios, known and unknown parameters, wherein the unknown parameters are estimated using the Maximum Likelihood Estimation method. The analysis illustrates that the gains in the relative precisions of the estimators, based on both RSSL and the proposed RSSH method, consistently exceed the estimators based on Simple Random Sampling (SRS), Ranked Set Sampling (RSS) and Extreme Ranked Set Sampling(ERSS) procedures in the case of skewed populations. Simulation studies carried over randomly generated populations and an application of the proposed methods to a time-to-event data validated the theoretical results.