Abstract
A review of generalized linear mixed models
Molenberghs, G., Renard, D., and Verbeke, G.
A general framework for modeling repeated categorical data is pictured, with three main model families: marginal, conditional, and subject-specific. The primary focus is on subject-specific or random-effects model, with some emphasis on the generalized linear mixed model. Estimation and optimization algorithms are discussed, together with available software. Advantages and disadvantages are pointed out. These tools have been exemplified using a simple but illustrative analysis. Similarities and differences between linear mixed models and generalized linear mixed models are discussed in detail.
Last update: January, 24, 2003 - Contact : S. Malali