Correlated Binary Regression Using a Quadratic Exponential Model
Lue Ping Zhao, Ross L. Prentice · Biometrika · 1990
A model of quadratic exponential form is parameterized in terms of marginal means and pairwise correlations for the regression analysis of correlated binary data. Pseudo-maximum likelihood methods using a special case termed the multiplicative model are proposed, but are noted to be computationally unattractive if the ‘blocks’ of correlated responses are at all large. On the other hand score estimating functions for mean and correlation parameters are shown to be of a particularly simple form under the quadratic exponential family. Special cases of such estimating functions having attractive computational properties are identified and illustrated.
Cited by 483 works. doi:10.1093/biomet/77.3.642
Authors
References
- Longitudinal Data Analysis Using Generalized Linear Models
- Pseudo Maximum Likelihood Methods: Theory
- The Analysis of Multivariate Binary Data
- A REPRESENTATION OF THE JOINT DISTRIBUTION OF RESPONSES TO N DICHOTOMOUS ITEMS
- 15 Design and Analysis of Cross-Over Trials
- Gaussian Estimation for Correlated Binomial Data