Marginal regression analysis of multivariate binary response
Marginal regression analysis of multivariate binary response
We propose the use of the mean parameter for regression analysis of a multivariate binary response. We model the association using dependence ratios defined in terms of the mean parameter, the components of which are the joint success probabilities of all orders. This permits flexible modelling of higher-order associations, using maximum likelihood estimation. We reanalyse two data sets, one with variable cluster size and the other a longitudinal data set with constant cluster size.
847-854
Ekholm, Anders
bc4d2421-5d0d-4250-b8e1-adddf60772e7
Smith, Peter W.F.
961a01a3-bf4c-43ca-9599-5be4fd5d3940
McDonald, John W.
9adae16e-e1e1-4ddf-bf4c-7231ee8c1c8e
November 1995
Ekholm, Anders
bc4d2421-5d0d-4250-b8e1-adddf60772e7
Smith, Peter W.F.
961a01a3-bf4c-43ca-9599-5be4fd5d3940
McDonald, John W.
9adae16e-e1e1-4ddf-bf4c-7231ee8c1c8e
Ekholm, Anders, Smith, Peter W.F. and McDonald, John W.
(1995)
Marginal regression analysis of multivariate binary response.
Biometrika, 82 (4), .
(doi:10.1093/biomet/82.4.847).
Abstract
We propose the use of the mean parameter for regression analysis of a multivariate binary response. We model the association using dependence ratios defined in terms of the mean parameter, the components of which are the joint success probabilities of all orders. This permits flexible modelling of higher-order associations, using maximum likelihood estimation. We reanalyse two data sets, one with variable cluster size and the other a longitudinal data set with constant cluster size.
Full text not available from this repository.
More information
Published date: November 1995
Identifiers
Local EPrints ID: 34310
URI: http://eprints.soton.ac.uk/id/eprint/34310
ISSN: 0006-3444
PURE UUID: cb74cf3a-7960-48b5-90d5-6c8f2b5c392f
Catalogue record
Date deposited: 18 Jan 2008
Last modified: 07 Oct 2020 01:34
Export record
Altmetrics
Contributors
Author:
Anders Ekholm
Author:
John W. McDonald
University divisions
Download statistics
Downloads from ePrints over the past year. Other digital versions may also be available to download e.g. from the publisher's website.
View more statistics