Homnormal: A comprehensive R package for testing the homogeneity of variances.

Saved in:

Bibliographic Details
Title:	Homnormal: A comprehensive R package for testing the homogeneity of variances.
Authors:	GÖKPINAR, Fikri¹, GÖKPINAR, Esra¹ eyigit@gazi.edu.tr
Source:	Sigma: Journal of Engineering & Natural Sciences / Mühendislik ve Fen Bilimleri Dergisi. Jun2026, Vol. 44 Issue 3, p1619-1634. 16p.
Subjects:	Likelihood ratio tests, Homoscedasticity, Monte Carlo method, Quality control, Statistical hypothesis testing
Abstract:	Testing the homogeneity of variances among normal populations is of interest in several research areas. It is widely applied to assess uniformity in quality control, biology, agricultural production systems, human performance studies, and even in the development of educational methods. For this reason, in this paper, we introduce an R package named homnormal, which includes some of the most powerful tests for homogeneity of variances, such as Bartlett's test, Levene's test, Brown-Forsthye's test, Bandairy Dai's test, generalized p-approach, computational approach test, likelihood ratio test based on the computational approach test, and standardized likelihood ratio test. The homnormal package is designed to be user-friendly, making it accessible for researchers across multidisciplinary fields. Additionally, we compare these tests based on their empirical power and type I error rates to determine the best-performing test in specific situations. Simulated type I error rates and powers of these tests are provided and evaluated through an extensive Monte Carlo simulation study. The simulation results indicate that, regardless of the number of groups or whether sample sizes are equal or unequal, small or large, the standardized likelihood ratio test and the likelihood ratio test based on the computational approach test consistently outperform other methods in terms of performance. [ABSTRACT FROM AUTHOR]
	Copyright of Sigma: Journal of Engineering & Natural Sciences / Mühendislik ve Fen Bilimleri Dergisi is the property of Sigma: Journal of Engineering & Natural Sciences / Mühendislik ve Fen Bilimleri Dergisi and its content may not be copied or emailed to multiple sites without the copyright holder's express written permission. Additionally, content may not be used with any artificial intelligence tools or machine learning technologies. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)
Database:	Engineering Source

Description
Abstract:	Testing the homogeneity of variances among normal populations is of interest in several research areas. It is widely applied to assess uniformity in quality control, biology, agricultural production systems, human performance studies, and even in the development of educational methods. For this reason, in this paper, we introduce an R package named homnormal, which includes some of the most powerful tests for homogeneity of variances, such as Bartlett's test, Levene's test, Brown-Forsthye's test, Bandairy Dai's test, generalized p-approach, computational approach test, likelihood ratio test based on the computational approach test, and standardized likelihood ratio test. The homnormal package is designed to be user-friendly, making it accessible for researchers across multidisciplinary fields. Additionally, we compare these tests based on their empirical power and type I error rates to determine the best-performing test in specific situations. Simulated type I error rates and powers of these tests are provided and evaluated through an extensive Monte Carlo simulation study. The simulation results indicate that, regardless of the number of groups or whether sample sizes are equal or unequal, small or large, the standardized likelihood ratio test and the likelihood ratio test based on the computational approach test consistently outperform other methods in terms of performance. [ABSTRACT FROM AUTHOR]
ISSN:	13047191
DOI:	10.14744/sigma.2025.00061