Cumulative logit models for matched pairs case-control design: Studies with covariates

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Binary as well as polytomous logistic models have been found useful for estimating odds ratios when the exposure of prime interest assumes unordered multiple levels under matched pair case-control design. In our earlier studies, we have shown the use of a polytomous logistic model for estimating cumulative odds ratios when the exposure of prime interest assumes multiple ordered levels under matched pair case-control design. In this paper, using the above model, we estimate the covariate adjusted cumulative odds ratios, in the case of an ordinal multiple level exposure variable under a pairwise matched case-control retrospective design. An approach, based on asymptotic distributional results, is also described to investigate whether or not the response categories are distinguishable with respect to the cumulative odds ratios after adjusting the effect of covariates. An illustrative example is presented and discussed.

Original languageEnglish
Pages (from-to)513-522
Number of pages10
JournalJournal of Applied Statistics
Volume33
Issue number5
DOIs
Publication statusPublished - Jun 2006

Keywords

  • Cumulative odds ratio
  • Deviance statistic
  • Logistic model
  • Matched pairs
  • Odds ratio
  • Polytomous logistic model

ASJC Scopus subject areas

  • Statistics and Probability

Fingerprint Dive into the research topics of 'Cumulative logit models for matched pairs case-control design: Studies with covariates'. Together they form a unique fingerprint.

  • Cite this