Statistical Modelling 2 (2002), 163–181
Generalized estimating equations: A hybrid approach for mean
parameters in multivariate regression models
Christoph Lange
Department of Biostatistics, Harvard School of Public Health,
655 Huntington Avenue,
Boston, MA 02155
U.S.A.
John C. Whittaker
Department of Epidemiology and Public Health, Imperial College School
of Medicine,
London
U.K.
Alex J. Macgregor
Twin Research & Genetic Epidemiology Unit, St. Thomas' Hospital,
London
U.K.
Abstract:
We propose an extension of the generalized estimating approach to
multivariate regression models (Liang and Zeger, 1986) which allows
the estimation of dispersion and association parameters in the
covariance matrix partly using estimating equations as in Prentice and
Zhao (1991), and partly by the direct use of consistent estimators.
The advantages of this hybrid approach over that of Prentice and Zhao
(1991) are a reduction in the number of fourth moment assumptions that
must be made, and the consequent reduction in numerical complexity. We
show that the type of estimation used for covariance parameters does
not affect the asymptotic efficiency of the mean parameter estimates.
The advantages of the hybrid model are illustrated by a simulation
study. This work was motivated by problems in statistical genetics,
and we illustrate our approach using a twin study examining
association between the osteocalcin receptor and various
osteoporisis-related traits.
Keywords:
GEE; GEE2; association mapping
Downloads:
Data and Software (using S-Plus) in
zipped archive
back