A Monte Carlo Study to Evaluate the Performance of Selected Tests of Homogeneity in Sparse Contingency Tables

Dale Kraemer, Robert F. Woolson

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

We address the problem of tests of homogeneity in two‐way contingency tables in case‐control studies when the case category is subdivided into k subcategories. In this situation, we have two cells with large frequencies and 2 X k cells with frequencies that become small as k increases. We propose two ad hoc statistics in which a statistic for the sparse cells is combined with a statistic for the cells with large frequencies. We will study these tests along with the Pearson test (using a chi‐square approximation) in a Monte Carlo simulation study. Two sets of null hypothesis models and two sets of alternative hypothesis models are considered. The best test for the models considered is the usual Pearson test (using an approximate chi‐square distribution) although the ad hoc models are more powerful under one alternative model considered.

Original languageEnglish (US)
Pages (from-to)763-778
Number of pages16
JournalBiometrical Journal
Volume30
Issue number7
DOIs
StatePublished - Jan 1 1988
Externally publishedYes

Fingerprint

Test of Homogeneity
Contingency Table
Monte Carlo Study
Evaluate
Cell
Statistic
Chi-square Distribution
Model
Case-control Study
Chi-square
Alternatives
Null hypothesis
Monte Carlo Simulation
Monte Carlo study
Homogeneity
Contingency table
Simulation Study
Statistics
Approximation

Keywords

  • Contingency tables
  • Epidemiology
  • Monte Carlo simulation
  • Product multinomial models

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this

A Monte Carlo Study to Evaluate the Performance of Selected Tests of Homogeneity in Sparse Contingency Tables. / Kraemer, Dale; Woolson, Robert F.

In: Biometrical Journal, Vol. 30, No. 7, 01.01.1988, p. 763-778.

Research output: Contribution to journalArticle

@article{8ef96f65520144d9991a4708eb78c54a,
title = "A Monte Carlo Study to Evaluate the Performance of Selected Tests of Homogeneity in Sparse Contingency Tables",
abstract = "We address the problem of tests of homogeneity in two‐way contingency tables in case‐control studies when the case category is subdivided into k subcategories. In this situation, we have two cells with large frequencies and 2 X k cells with frequencies that become small as k increases. We propose two ad hoc statistics in which a statistic for the sparse cells is combined with a statistic for the cells with large frequencies. We will study these tests along with the Pearson test (using a chi‐square approximation) in a Monte Carlo simulation study. Two sets of null hypothesis models and two sets of alternative hypothesis models are considered. The best test for the models considered is the usual Pearson test (using an approximate chi‐square distribution) although the ad hoc models are more powerful under one alternative model considered.",
keywords = "Contingency tables, Epidemiology, Monte Carlo simulation, Product multinomial models",
author = "Dale Kraemer and Woolson, {Robert F.}",
year = "1988",
month = "1",
day = "1",
doi = "10.1002/bimj.4710300702",
language = "English (US)",
volume = "30",
pages = "763--778",
journal = "Biometrical Journal",
issn = "0323-3847",
publisher = "Wiley-VCH Verlag",
number = "7",

}

TY - JOUR

T1 - A Monte Carlo Study to Evaluate the Performance of Selected Tests of Homogeneity in Sparse Contingency Tables

AU - Kraemer, Dale

AU - Woolson, Robert F.

PY - 1988/1/1

Y1 - 1988/1/1

N2 - We address the problem of tests of homogeneity in two‐way contingency tables in case‐control studies when the case category is subdivided into k subcategories. In this situation, we have two cells with large frequencies and 2 X k cells with frequencies that become small as k increases. We propose two ad hoc statistics in which a statistic for the sparse cells is combined with a statistic for the cells with large frequencies. We will study these tests along with the Pearson test (using a chi‐square approximation) in a Monte Carlo simulation study. Two sets of null hypothesis models and two sets of alternative hypothesis models are considered. The best test for the models considered is the usual Pearson test (using an approximate chi‐square distribution) although the ad hoc models are more powerful under one alternative model considered.

AB - We address the problem of tests of homogeneity in two‐way contingency tables in case‐control studies when the case category is subdivided into k subcategories. In this situation, we have two cells with large frequencies and 2 X k cells with frequencies that become small as k increases. We propose two ad hoc statistics in which a statistic for the sparse cells is combined with a statistic for the cells with large frequencies. We will study these tests along with the Pearson test (using a chi‐square approximation) in a Monte Carlo simulation study. Two sets of null hypothesis models and two sets of alternative hypothesis models are considered. The best test for the models considered is the usual Pearson test (using an approximate chi‐square distribution) although the ad hoc models are more powerful under one alternative model considered.

KW - Contingency tables

KW - Epidemiology

KW - Monte Carlo simulation

KW - Product multinomial models

UR - http://www.scopus.com/inward/record.url?scp=84982047403&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84982047403&partnerID=8YFLogxK

U2 - 10.1002/bimj.4710300702

DO - 10.1002/bimj.4710300702

M3 - Article

AN - SCOPUS:84982047403

VL - 30

SP - 763

EP - 778

JO - Biometrical Journal

JF - Biometrical Journal

SN - 0323-3847

IS - 7

ER -