A comparison of tests for restricted orderings in the three-class case

Todd A. Alonzo, Christos T. Nakas, Constantin Yiannoutsos, Sherri Bucher

Research output: Contribution to journalArticle

12 Citations (Scopus)

Abstract

A variety of methods for comparing three distributions have been proposed in the literature. These methods assess the same null hypothesis of equal distributions but differ in the alternative hypothesis they consider. The alternative hypothesis can be that measurements from the three classes are distributed according to unequal distributions or that measurements between the three classes follow a specific monotone ordering, an inverse-U-shaped (umbrella) ordering, or a U-shaped (tree) ordering. This paper compares these tests with respect to power and test size under different simulation scenarios. In addition, the methods are illustrated in two applications generated by different research questions with data from three classes suggesting monotone and umbrella orders. Additionally, proposals for the appropriate application of these tests are provided.

Original languageEnglish
Pages (from-to)1144-1158
Number of pages15
JournalStatistics in Medicine
Volume28
Issue number7
DOIs
StatePublished - Mar 30 2009

Fingerprint

Monotone
Alternatives
Unequal
Null hypothesis
Scenarios
Research
Class
Simulation

Keywords

  • ANOVA
  • Area under ROC curve
  • Kruskal-Walis test
  • Non-parametric tests
  • Ordered alternatives
  • Umbrella alternatives
  • Umbrella ROC volume
  • Volume under the ROC surface
  • Wilcoxon-Mann-Whitney test

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability

Cite this

A comparison of tests for restricted orderings in the three-class case. / Alonzo, Todd A.; Nakas, Christos T.; Yiannoutsos, Constantin; Bucher, Sherri.

In: Statistics in Medicine, Vol. 28, No. 7, 30.03.2009, p. 1144-1158.

Research output: Contribution to journalArticle

@article{746de2bed6a2438fbdc609a72ffabb71,
title = "A comparison of tests for restricted orderings in the three-class case",
abstract = "A variety of methods for comparing three distributions have been proposed in the literature. These methods assess the same null hypothesis of equal distributions but differ in the alternative hypothesis they consider. The alternative hypothesis can be that measurements from the three classes are distributed according to unequal distributions or that measurements between the three classes follow a specific monotone ordering, an inverse-U-shaped (umbrella) ordering, or a U-shaped (tree) ordering. This paper compares these tests with respect to power and test size under different simulation scenarios. In addition, the methods are illustrated in two applications generated by different research questions with data from three classes suggesting monotone and umbrella orders. Additionally, proposals for the appropriate application of these tests are provided.",
keywords = "ANOVA, Area under ROC curve, Kruskal-Walis test, Non-parametric tests, Ordered alternatives, Umbrella alternatives, Umbrella ROC volume, Volume under the ROC surface, Wilcoxon-Mann-Whitney test",
author = "Alonzo, {Todd A.} and Nakas, {Christos T.} and Constantin Yiannoutsos and Sherri Bucher",
year = "2009",
month = "3",
day = "30",
doi = "10.1002/sim.3536",
language = "English",
volume = "28",
pages = "1144--1158",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "7",

}

TY - JOUR

T1 - A comparison of tests for restricted orderings in the three-class case

AU - Alonzo, Todd A.

AU - Nakas, Christos T.

AU - Yiannoutsos, Constantin

AU - Bucher, Sherri

PY - 2009/3/30

Y1 - 2009/3/30

N2 - A variety of methods for comparing three distributions have been proposed in the literature. These methods assess the same null hypothesis of equal distributions but differ in the alternative hypothesis they consider. The alternative hypothesis can be that measurements from the three classes are distributed according to unequal distributions or that measurements between the three classes follow a specific monotone ordering, an inverse-U-shaped (umbrella) ordering, or a U-shaped (tree) ordering. This paper compares these tests with respect to power and test size under different simulation scenarios. In addition, the methods are illustrated in two applications generated by different research questions with data from three classes suggesting monotone and umbrella orders. Additionally, proposals for the appropriate application of these tests are provided.

AB - A variety of methods for comparing three distributions have been proposed in the literature. These methods assess the same null hypothesis of equal distributions but differ in the alternative hypothesis they consider. The alternative hypothesis can be that measurements from the three classes are distributed according to unequal distributions or that measurements between the three classes follow a specific monotone ordering, an inverse-U-shaped (umbrella) ordering, or a U-shaped (tree) ordering. This paper compares these tests with respect to power and test size under different simulation scenarios. In addition, the methods are illustrated in two applications generated by different research questions with data from three classes suggesting monotone and umbrella orders. Additionally, proposals for the appropriate application of these tests are provided.

KW - ANOVA

KW - Area under ROC curve

KW - Kruskal-Walis test

KW - Non-parametric tests

KW - Ordered alternatives

KW - Umbrella alternatives

KW - Umbrella ROC volume

KW - Volume under the ROC surface

KW - Wilcoxon-Mann-Whitney test

UR - http://www.scopus.com/inward/record.url?scp=69949091341&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=69949091341&partnerID=8YFLogxK

U2 - 10.1002/sim.3536

DO - 10.1002/sim.3536

M3 - Article

VL - 28

SP - 1144

EP - 1158

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 7

ER -