Inverse unit compound Rayleigh distribution: statistical properties with applications in different fields

Abstract The present article presents a novel two-parameter distribution called inverse unit compound Rayleigh distribution (IUCRD), which has the support (1, $$+\infty$$ ). It is propounded via the inverse transformation of the unit compound Rayleigh distribution (UCRD). Different properties of the...

Full description

Saved in:
Bibliographic Details
Main Authors: Hatem E. Semary, Emmanuel W. Okereke, Laxmi Prasad Sapkota, A. S. Al-Moisheer, Abdirashid M. Yousuf, Eslam Hussam, Ahmed M. Gemeay
Format: Article
Language:English
Published: Nature Portfolio 2025-08-01
Series:Scientific Reports
Subjects:
Online Access:https://doi.org/10.1038/s41598-025-07915-5
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849763642829963264
author Hatem E. Semary
Emmanuel W. Okereke
Laxmi Prasad Sapkota
A. S. Al-Moisheer
Abdirashid M. Yousuf
Eslam Hussam
Ahmed M. Gemeay
author_facet Hatem E. Semary
Emmanuel W. Okereke
Laxmi Prasad Sapkota
A. S. Al-Moisheer
Abdirashid M. Yousuf
Eslam Hussam
Ahmed M. Gemeay
author_sort Hatem E. Semary
collection DOAJ
description Abstract The present article presents a novel two-parameter distribution called inverse unit compound Rayleigh distribution (IUCRD), which has the support (1, $$+\infty$$ ). It is propounded via the inverse transformation of the unit compound Rayleigh distribution (UCRD). Different properties of the IUCRD, namely, the quantile function, the mode, stochastic ordering, moments, and heavy-tailedness, among others, are explored. We notice that the distribution PDF is either unimodal or nonincreasing. It can also be left-skewed or right-skewed, depending on the values of the parameters of the IUCRD. The graph of the hazard rate function of the IUCRD is the upside-down bathtub shape or nondecreasing. Heavy-tailedness is also among the properties of the IUCRD determined in this work. We provide evidence of the relationship between the IUCRD and exponential distribution via the derivation of distributions of certain functions of one variable. Sixteen different estimation methods are compared employing the Monte Carlo simulation procedure. Numerical simulation evidence attests to the KE method being the best estimation methodology for the parameters of the IUCRD. Interestingly, according to the simulation results, the ML procedure assumes the second position. In demonstrating the usefulness of the IUCRD, we use the ML technique to fit the distribution to five real-world datasets and compare its fits with the fits of seven existing distributions to the data by employing goodness of fit statistics. For each of the data, the minimum value of each of the statistics corresponds to the IUCRD. This result makes it clear that in many data analysis circumstances, the IUCRD can be preferable to several continuous distributions, especially the UCRD and the inverse unit exponential, inverse Weibull, inverse Rayleigh, inverse Chen, inverse exponential, and inverse exponential distributions.
format Article
id doaj-art-2688432c73204f0f8e3efbdd80aa9312
institution DOAJ
issn 2045-2322
language English
publishDate 2025-08-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-2688432c73204f0f8e3efbdd80aa93122025-08-20T03:05:21ZengNature PortfolioScientific Reports2045-23222025-08-0115114510.1038/s41598-025-07915-5Inverse unit compound Rayleigh distribution: statistical properties with applications in different fieldsHatem E. Semary0Emmanuel W. Okereke1Laxmi Prasad Sapkota2A. S. Al-Moisheer3Abdirashid M. Yousuf4Eslam Hussam5Ahmed M. Gemeay6Department of Mathematics and Statistics, Faculty of Science, Imam Mohammad Ibn Saud Islamic University (IMSIU)Department of Statistics, Michael Okpara University of AgricultureDepartment of Statistics, Tribhuvan UniversityDepartment of Mathematics and Statistics, Faculty of Science, Imam Mohammad Ibn Saud Islamic University (IMSIU)Research and Innovation Center, Amoud UniversityDepartment of Accounting, College of Business Administration in Hawtat Bani Tamim, Prince Sattam bin Abdulaziz UniversityDepartment of Mathematics, Faculty of Science, Tanta UniversityAbstract The present article presents a novel two-parameter distribution called inverse unit compound Rayleigh distribution (IUCRD), which has the support (1, $$+\infty$$ ). It is propounded via the inverse transformation of the unit compound Rayleigh distribution (UCRD). Different properties of the IUCRD, namely, the quantile function, the mode, stochastic ordering, moments, and heavy-tailedness, among others, are explored. We notice that the distribution PDF is either unimodal or nonincreasing. It can also be left-skewed or right-skewed, depending on the values of the parameters of the IUCRD. The graph of the hazard rate function of the IUCRD is the upside-down bathtub shape or nondecreasing. Heavy-tailedness is also among the properties of the IUCRD determined in this work. We provide evidence of the relationship between the IUCRD and exponential distribution via the derivation of distributions of certain functions of one variable. Sixteen different estimation methods are compared employing the Monte Carlo simulation procedure. Numerical simulation evidence attests to the KE method being the best estimation methodology for the parameters of the IUCRD. Interestingly, according to the simulation results, the ML procedure assumes the second position. In demonstrating the usefulness of the IUCRD, we use the ML technique to fit the distribution to five real-world datasets and compare its fits with the fits of seven existing distributions to the data by employing goodness of fit statistics. For each of the data, the minimum value of each of the statistics corresponds to the IUCRD. This result makes it clear that in many data analysis circumstances, the IUCRD can be preferable to several continuous distributions, especially the UCRD and the inverse unit exponential, inverse Weibull, inverse Rayleigh, inverse Chen, inverse exponential, and inverse exponential distributions.https://doi.org/10.1038/s41598-025-07915-5Estimation methodsHeavy-tailed distributionSimulationUnimodal distributionUnit compound Rayleigh distribution
spellingShingle Hatem E. Semary
Emmanuel W. Okereke
Laxmi Prasad Sapkota
A. S. Al-Moisheer
Abdirashid M. Yousuf
Eslam Hussam
Ahmed M. Gemeay
Inverse unit compound Rayleigh distribution: statistical properties with applications in different fields
Scientific Reports
Estimation methods
Heavy-tailed distribution
Simulation
Unimodal distribution
Unit compound Rayleigh distribution
title Inverse unit compound Rayleigh distribution: statistical properties with applications in different fields
title_full Inverse unit compound Rayleigh distribution: statistical properties with applications in different fields
title_fullStr Inverse unit compound Rayleigh distribution: statistical properties with applications in different fields
title_full_unstemmed Inverse unit compound Rayleigh distribution: statistical properties with applications in different fields
title_short Inverse unit compound Rayleigh distribution: statistical properties with applications in different fields
title_sort inverse unit compound rayleigh distribution statistical properties with applications in different fields
topic Estimation methods
Heavy-tailed distribution
Simulation
Unimodal distribution
Unit compound Rayleigh distribution
url https://doi.org/10.1038/s41598-025-07915-5
work_keys_str_mv AT hatemesemary inverseunitcompoundrayleighdistributionstatisticalpropertieswithapplicationsindifferentfields
AT emmanuelwokereke inverseunitcompoundrayleighdistributionstatisticalpropertieswithapplicationsindifferentfields
AT laxmiprasadsapkota inverseunitcompoundrayleighdistributionstatisticalpropertieswithapplicationsindifferentfields
AT asalmoisheer inverseunitcompoundrayleighdistributionstatisticalpropertieswithapplicationsindifferentfields
AT abdirashidmyousuf inverseunitcompoundrayleighdistributionstatisticalpropertieswithapplicationsindifferentfields
AT eslamhussam inverseunitcompoundrayleighdistributionstatisticalpropertieswithapplicationsindifferentfields
AT ahmedmgemeay inverseunitcompoundrayleighdistributionstatisticalpropertieswithapplicationsindifferentfields