This popular book covers generalized estimating equations (GEEs), an extension of the generalized linear model when correlation is unknown. It remains the only book to focus specifically on this important method used in the analysis of longitudinal data. This edition features new methods, particularly regarding correlation structure and the interpretation of GEE models. It also presents updated computational material and now includes both Stata and R code for all the examples as well as corresponding SAS code. All data sets and code are available at CRC Press Online.

INTRODUCTION

Notational Conventions

A Short Review of Generalized Linear Models

Software

Exercises

MODEL CONSTRUCTION AND ESTIMATING EQUATIONS

Independent Data

Estimating the Variance of the Estimates

Panel Data

Estimation

Summary

Exercises

GENERALIZED ESTIMATING EQUATIONS

Population-Averaged (PA) and Subject-Specific (SS) Models

The PA-GEE for GLMs

The SS-GEE for GLMs

The GEE2 for GLMs

GEEs for Extensions of GLMs

Further Developments and Applications

Missing Data

Choosing an Appropriate Model

Summary

Exercises

RESIDUALS, DIAGNOSTICS, AND TESTING

Criterion Measures

Analysis of Residuals

Deletion Diagnostics

Goodness of Fit (Population-Averaged Models)

Testing Coefficients in the PA-GEE Model

Assessing the MCAR Assumption of PA-GEE Models

Summary

Exercises

PROGRAMS AND DATASETS

Programs

Datasets

References

Author Index

Subject Index