Tests for the homogeneity of two binomial proportions in extremely unbalanced 2 × 2 contingency tables

Seung Ho Kang, Chul W. Ahn

Research output: Contribution to journalArticle

8 Citations (Scopus)

Abstract

Asymptotic tests such as the Pearson chi-square test are unreliable for testing the homogeneity of two binomial probabilities in extremely unbalanced cases. Two exact tests (conditional and unconditional) are available as alternatives and can be implemented easily in StatXact 6.0. In equal sample cases it is well known that the unconditional exact test is more powerful than the conditional exact test. However, in this paper, we show that the opposite result holds in extremely unbalanced cases. The reason is that the peaks of the type I error occur at the extremes of the nuisance parameter when the imbalance among the sample sizes becomes severe. After we show that the conditional exact test is more powerful than the unconditional exact test in extremely unbalanced cases whose sample ratio is greater than 20, we compare the conditional exact test with the Berger and Boos approach (J. Amer. Stat. Assoc. 1994; 89:1012-1016) in which the supremum is taken over a confidence interval for the nuisance parameter. The Berger and Boos approach turns out to be slightly more powerful than the conditional exact test in extremely unbalanced data. A real example is provided.

Original languageEnglish
Pages (from-to)2524-2535
Number of pages12
JournalStatistics in Medicine
Volume27
Issue number14
DOIs
Publication statusPublished - 2008 Jun 30

Fingerprint

Exact Test
Contingency Table
Chi-Square Distribution
Homogeneity
Sample Size
Proportion
Conditional Test
Confidence Intervals
Unconditional Test
Nuisance Parameter
Unbalanced Data
Asymptotic Test
Chi-squared test
Type I error
Supremum
Confidence interval
Extremes
Testing
Alternatives

All Science Journal Classification (ASJC) codes

  • Epidemiology
  • Statistics and Probability

Cite this

@article{868ffeb69f1745718749a2c196d3232b,
title = "Tests for the homogeneity of two binomial proportions in extremely unbalanced 2 × 2 contingency tables",
abstract = "Asymptotic tests such as the Pearson chi-square test are unreliable for testing the homogeneity of two binomial probabilities in extremely unbalanced cases. Two exact tests (conditional and unconditional) are available as alternatives and can be implemented easily in StatXact 6.0. In equal sample cases it is well known that the unconditional exact test is more powerful than the conditional exact test. However, in this paper, we show that the opposite result holds in extremely unbalanced cases. The reason is that the peaks of the type I error occur at the extremes of the nuisance parameter when the imbalance among the sample sizes becomes severe. After we show that the conditional exact test is more powerful than the unconditional exact test in extremely unbalanced cases whose sample ratio is greater than 20, we compare the conditional exact test with the Berger and Boos approach (J. Amer. Stat. Assoc. 1994; 89:1012-1016) in which the supremum is taken over a confidence interval for the nuisance parameter. The Berger and Boos approach turns out to be slightly more powerful than the conditional exact test in extremely unbalanced data. A real example is provided.",
author = "Kang, {Seung Ho} and Ahn, {Chul W.}",
year = "2008",
month = "6",
day = "30",
doi = "10.1002/sim.3055",
language = "English",
volume = "27",
pages = "2524--2535",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "14",

}

Tests for the homogeneity of two binomial proportions in extremely unbalanced 2 × 2 contingency tables. / Kang, Seung Ho; Ahn, Chul W.

In: Statistics in Medicine, Vol. 27, No. 14, 30.06.2008, p. 2524-2535.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Tests for the homogeneity of two binomial proportions in extremely unbalanced 2 × 2 contingency tables

AU - Kang, Seung Ho

AU - Ahn, Chul W.

PY - 2008/6/30

Y1 - 2008/6/30

N2 - Asymptotic tests such as the Pearson chi-square test are unreliable for testing the homogeneity of two binomial probabilities in extremely unbalanced cases. Two exact tests (conditional and unconditional) are available as alternatives and can be implemented easily in StatXact 6.0. In equal sample cases it is well known that the unconditional exact test is more powerful than the conditional exact test. However, in this paper, we show that the opposite result holds in extremely unbalanced cases. The reason is that the peaks of the type I error occur at the extremes of the nuisance parameter when the imbalance among the sample sizes becomes severe. After we show that the conditional exact test is more powerful than the unconditional exact test in extremely unbalanced cases whose sample ratio is greater than 20, we compare the conditional exact test with the Berger and Boos approach (J. Amer. Stat. Assoc. 1994; 89:1012-1016) in which the supremum is taken over a confidence interval for the nuisance parameter. The Berger and Boos approach turns out to be slightly more powerful than the conditional exact test in extremely unbalanced data. A real example is provided.

AB - Asymptotic tests such as the Pearson chi-square test are unreliable for testing the homogeneity of two binomial probabilities in extremely unbalanced cases. Two exact tests (conditional and unconditional) are available as alternatives and can be implemented easily in StatXact 6.0. In equal sample cases it is well known that the unconditional exact test is more powerful than the conditional exact test. However, in this paper, we show that the opposite result holds in extremely unbalanced cases. The reason is that the peaks of the type I error occur at the extremes of the nuisance parameter when the imbalance among the sample sizes becomes severe. After we show that the conditional exact test is more powerful than the unconditional exact test in extremely unbalanced cases whose sample ratio is greater than 20, we compare the conditional exact test with the Berger and Boos approach (J. Amer. Stat. Assoc. 1994; 89:1012-1016) in which the supremum is taken over a confidence interval for the nuisance parameter. The Berger and Boos approach turns out to be slightly more powerful than the conditional exact test in extremely unbalanced data. A real example is provided.

UR - http://www.scopus.com/inward/record.url?scp=47249121840&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=47249121840&partnerID=8YFLogxK

U2 - 10.1002/sim.3055

DO - 10.1002/sim.3055

M3 - Article

VL - 27

SP - 2524

EP - 2535

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 14

ER -