Generalized Linear and Nonlinear Models for Correlated Data: Theory and Applications Using SAS

Part IINonlinear Models

4Generalized Linear and Nonlinear Models

4.1 The generalized linear model (GLIM)

4.2 The GLIM for correlated response data

4.3 Examples of GLIM’s

4.4 The generalized nonlinear model (GNLM)

4.5 Examples of GNLM’s

4.6 Computational considerations

4.7 Summary

While normal-theory linear models find a wide range of use, there are numerous applications where the data are non-Gaussian in nature and, as such, involve distributions where the mean response is likely to be nonlinear in the parameters of interest. In Chapter 1 we considered several such examples including the respiratory disorder data involving repeated binary outcomes over time and the epileptic seizure data in which the primary outcome was a count of the number ...