Scholarlink

<p>A modern family of distributions, called log-cosine power-generated (LCP-G) family, is formulated in this study. The cumulative distribution function (cdf) and the probability density function (pdf) are uniquely derived. The Taylor series expansion is used to extend this pdf, which facilitates obtaining important statistical properties. The new family is used to generate the LCP-Weibull (LCPW) distribution as a further appendage to the Weibull distribution. Eleven techniques are used to estimate the parameters of the LCPW distribution to determine which estimation technique is the most efficient. Monte Carlo simulation experiments are also carried out to compare the performance of these techniques. The suitability of the LCPW distribution for analyzing data sets is demonstrated, and the metrics for comparison persistently select the LCPW distribution over other candidate models. The results confirm that the LCPW distribution is suitable for analyzing data generated from different domains.</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

The Log-Cosine-Power Generated Family of Distributions

  • 期刊论文

  • 发表日期：

    2025年5月27日

查看全文
  • Suleman Nasiru,
  • Christophe Chesneau,
  • Selasi Kwaku Ocloo,
  • Abdul Ghaniyyu Abubakari

摘要

A modern family of distributions, called log-cosine power-generated (LCP-G) family, is formulated in this study. The cumulative distribution function (cdf) and the probability density function (pdf) are uniquely derived. The Taylor series expansion is used to extend this pdf, which facilitates obtaining important statistical properties. The new family is used to generate the LCP-Weibull (LCPW) distribution as a further appendage to the Weibull distribution. Eleven techniques are used to estimate the parameters of the LCPW distribution to determine which estimation technique is the most efficient. Monte Carlo simulation experiments are also carried out to compare the performance of these techniques. The suitability of the LCPW distribution for analyzing data sets is demonstrated, and the metrics for comparison persistently select the LCPW distribution over other candidate models. The results confirm that the LCPW distribution is suitable for analyzing data generated from different domains.