GEE analysis of negatively correlated binary responses: A caution

James A. Hanley, Abdissa Negassa, Michael D.De B. Edwardes

Research output: Contribution to journalArticle

22 Scopus citations

Abstract

The method of generalized estimating equations has become almost standard for analysing longitudinal and other correlated response data. However, we have found that if binary responses have less than binomial variation over clusters, and are modelled using exchangeable correlations, prevailing software implementations may give unreliable results. Bounding the negative correlation away from its theoretical minimum may not always be a satisfactory solution. In such instances, using the independence working correlation structure and robust SEs is a more trustworthy alternative. Copyright (C) 2000 John Wiley and Sons, Ltd.

Original languageEnglish (US)
Pages (from-to)715-722
Number of pages8
JournalStatistics in Medicine
Volume19
Issue number5
DOIs
Publication statusPublished - Mar 15 2000
Externally publishedYes

    Fingerprint

ASJC Scopus subject areas

  • Epidemiology
  • Statistics and Probability

Cite this