A NOTE ON FITTING A MARGINAL MODEL TO MIXED EFFECTS LOG-LINEAR REGRESSION DATA VIA GEE

Authors
GROMPING U
Citation
U. Gromping, A NOTE ON FITTING A MARGINAL MODEL TO MIXED EFFECTS LOG-LINEAR REGRESSION DATA VIA GEE, Biometrics, 52(1), 1996, pp. 280-285
Citations number
15
Categorie Soggetti
Statistic & Probability","Statistic & Probability
Journal title
Biometrics → ACNP
ISSN journal
0006341X
Volume
52
Issue
1
Year of publication
1996
Pages
280 - 285
Database
ISI
SICI code
0006-341X(1996)52:1<280:ANOFAM>2.0.ZU;2-8
Abstract
Marginal generalized linear models for correlated data are often fit b y generalized estimating equations (Liang, K. Y. and Zeger, S. L., 198 6, Biometrika 73, 13-22), which requires the marginal expectation to b e correctly specified while the covariance structure is allowed to be misspecified. This note is concerned with the correct specification of the marginal mean for data from a mixed log-linear regression model a nd the relationship between the true subject-specific parameters and t he true marginal parameters.