Application of Equality Test of Coefficients of Variation to the Heteroskedasticity Test

Tovohery, Josoa Michel and Totohasina, André and Rajaonasy, Feno Daniel (2020) Application of Equality Test of Coefficients of Variation to the Heteroskedasticity Test. American Journal of Computational Mathematics, 10 (01). pp. 73-89. ISSN 2161-1203

[thumbnail of ajcm_2020030514475584.pdf] Text
ajcm_2020030514475584.pdf - Published Version

Download (1MB)

Abstract

The presence of heteroskedasticity in a considered regression model may bias the standard deviations of parameters obtained by the Ordinary Least Square (OLS) method. In this case, several hypothesis tests on the model under consideration may be biased, for example, CHOW’s coefficient stability test (or structural change test), Student’s t-test and Fisher’s F-test. Most of the heteroscedasticity tests in the literature are based on the comparison of variances. Despite the multiplication of equality tests of coefficients of variation (CVs) that have appeared in the literature, to our knowledge, the first and only use of the coefficient of variation in the detection of heteroskedasticity was offered by Li and Yao in 2017. Thus, this paper offers an approach to determine the existence of heteroskedasticity by a test of equality of coefficients of variation. We verify by a Monte Carlo robustness and performance test that our method seems even better than some tests in the literature. The results of this study contribute to the exploitation of the statistical measurement of CV dispersion. They help technicians economists to better verify their hypotheses before making a scientific decision when making a necessary forecast, in order to contribute effectively to the economic and sustainable development of a company or enterprise.

Item Type: Article
Subjects: STM Open Press > Mathematical Science
Depositing User: Unnamed user with email support@stmopenpress.com
Date Deposited: 15 Jun 2023 07:42
Last Modified: 19 Oct 2024 03:52
URI: http://journal.submissionpages.com/id/eprint/1556

Actions (login required)

View Item
View Item