Covariance structure models for clustered proportions Journal Article


Authors: Yao, T. J.; Reinsel, G. C.
Article Title: Covariance structure models for clustered proportions
Abstract: Covariance structures as functions of the means and extra-variation and correlation parameters are presented to model proportions correlated due to random clustering effects. The mean structures are specified by logistic regressions with proportion-specific covariates. The maximum quasi-likelihood/moment estimating functions for all the parameters are constructed. The resulting estimators are asymptotically normal. A general quasi-score test procedure is developed for testing the significance of any subset of the parameters. The small sample properties are studied through simulations. An illustrative example is included.
Keywords: correlated proportions; estimating functions; quasi-score test; variance-covariance structure
Journal Title: Statistics & Probability Letters
Volume: 29
Issue: 1
ISSN: 0167-7152
Publisher: Elsevier Science, Inc.  
Date Published: 1996-08-15
Start Page: 61
End Page: 70
Language: English
DOI: 10.1016/0167-7152(95)00156-5
PROVIDER: scopus
DOI/URL:
Notes: Article -- Export Date: 22 November 2017 -- Source: Scopus
Altmetric
Citation Impact
BMJ Impact Analytics
MSK Authors
  1. Tzy-Jyun Yao
    59 Yao