Comparing the performance of modified Ft statistic with ANOVA and Kruskal Wallis test
ANOVA is a classical test statistics for testing the equality of groups. However this test is very sensitive to nonnormality as well as variance heterogeneity. To overcome the problem of nonnormality, robust method such as Ft test statistic can be used but the test statistic can only perform well wh...
Saved in:
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Natural Sciences Publishing Cor.
2013
|
Subjects: | |
Online Access: | https://repo.uum.edu.my/id/eprint/30898/1/AMIS%2007%2021%202013%20403-408.pdf http://dx.doi.org/10.12785/amis/072L04 https://repo.uum.edu.my/id/eprint/30898/ https://www.naturalspublishing.com/Article.asp?ArtcID=3228 http://dx.doi.org/10.12785/amis/072L04 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
id |
my.uum.repo.30898 |
---|---|
record_format |
eprints |
spelling |
my.uum.repo.308982024-06-23T09:09:08Z https://repo.uum.edu.my/id/eprint/30898/ Comparing the performance of modified Ft statistic with ANOVA and Kruskal Wallis test Md Yusof, Zahayu Abdullah, Suhaida Syed Yahaya, Sharipah Soaad QA Mathematics ANOVA is a classical test statistics for testing the equality of groups. However this test is very sensitive to nonnormality as well as variance heterogeneity. To overcome the problem of nonnormality, robust method such as Ft test statistic can be used but the test statistic can only perform well when the assumption of homoscedasticity is met. This is due to the biasness of mean as a central tendency measure. This study proposed a robust procedure known as modified Ft method which combines the Ft statistics with one of the popular robust scale estimators, MADn, Tn and LMSn. A simulation study was conducted to compare the robustness (Type I error) of the method with respect to its counterpart from the parametric and non parametric aspects, ANOVA and KruskalWallis respectively. This innovation enhances the ability of modified Ft statistic to provide good control of Type I error rates. The findings were in favor of the modified Ft method especially for skewed data. The performance of the method was demonstrated on real education data Natural Sciences Publishing Cor. 2013 Article PeerReviewed application/pdf en https://repo.uum.edu.my/id/eprint/30898/1/AMIS%2007%2021%202013%20403-408.pdf Md Yusof, Zahayu and Abdullah, Suhaida and Syed Yahaya, Sharipah Soaad (2013) Comparing the performance of modified Ft statistic with ANOVA and Kruskal Wallis test. Applied Mathematics & Information Sciences An International Journal, 7 (21). pp. 403-408. ISSN 2373-8944 https://www.naturalspublishing.com/Article.asp?ArtcID=3228 http://dx.doi.org/10.12785/amis/072L04 http://dx.doi.org/10.12785/amis/072L04 |
institution |
Universiti Utara Malaysia |
building |
UUM Library |
collection |
Institutional Repository |
continent |
Asia |
country |
Malaysia |
content_provider |
Universiti Utara Malaysia |
content_source |
UUM Institutional Repository |
url_provider |
http://repo.uum.edu.my/ |
language |
English |
topic |
QA Mathematics |
spellingShingle |
QA Mathematics Md Yusof, Zahayu Abdullah, Suhaida Syed Yahaya, Sharipah Soaad Comparing the performance of modified Ft statistic with ANOVA and Kruskal Wallis test |
description |
ANOVA is a classical test statistics for testing the equality of groups. However this test is very sensitive to nonnormality as well as variance heterogeneity. To overcome the problem of nonnormality, robust method such as Ft test statistic can be used but the test statistic can only perform well when the assumption of homoscedasticity is met. This is due to the biasness of mean as a central tendency measure. This study proposed a robust procedure known as modified Ft method which combines the Ft statistics with one of the popular robust scale estimators, MADn, Tn and LMSn. A simulation study was conducted to compare the robustness (Type I error) of the method with respect to its counterpart from the parametric and non parametric aspects, ANOVA and KruskalWallis respectively. This innovation enhances the ability of modified Ft statistic to provide good control of Type I error rates. The findings were in favor of the modified Ft method especially for skewed data. The performance of the method was demonstrated on real education data |
format |
Article |
author |
Md Yusof, Zahayu Abdullah, Suhaida Syed Yahaya, Sharipah Soaad |
author_facet |
Md Yusof, Zahayu Abdullah, Suhaida Syed Yahaya, Sharipah Soaad |
author_sort |
Md Yusof, Zahayu |
title |
Comparing the performance of modified Ft statistic with ANOVA and Kruskal Wallis test |
title_short |
Comparing the performance of modified Ft statistic with ANOVA and Kruskal Wallis test |
title_full |
Comparing the performance of modified Ft statistic with ANOVA and Kruskal Wallis test |
title_fullStr |
Comparing the performance of modified Ft statistic with ANOVA and Kruskal Wallis test |
title_full_unstemmed |
Comparing the performance of modified Ft statistic with ANOVA and Kruskal Wallis test |
title_sort |
comparing the performance of modified ft statistic with anova and kruskal wallis test |
publisher |
Natural Sciences Publishing Cor. |
publishDate |
2013 |
url |
https://repo.uum.edu.my/id/eprint/30898/1/AMIS%2007%2021%202013%20403-408.pdf http://dx.doi.org/10.12785/amis/072L04 https://repo.uum.edu.my/id/eprint/30898/ https://www.naturalspublishing.com/Article.asp?ArtcID=3228 http://dx.doi.org/10.12785/amis/072L04 |
_version_ |
1802980294081380352 |
score |
13.18916 |