Linear Mixed Models for Longitudinal Data

Rating:
Summary: thorough treatment of linear mixed models
Review: This book is basically an update of their 1997 mongraph. Longitudinal data are important in biostatistics and particularly in the analysis of clinical trials. There are effective methods for handling longitudinal data using linear models with covariance structures that represent the time dependence of the repeated observations. There are many subtle issues in the analysis and many who analyze longitudinal data apply incorrect linear models and are often not aware of the consequences of their decisions. The authors were motivated to provide a reference source to remedy this problem. The book presents the theory and applications and uses SAS Proc Mixed as a vehicle for presenting many of the results in a clear and understandable fashion. An important feature of the book is its emphasis on how best to deal with the problem of missing data. This is covered in chapters 14 - 16. Although SAS is emphasized throughout the book other software tools are also illustrated in Appendix A (including SPlus). SUDAAN is a package produced by the Research Triangle Institute in North Carolina that also handles longitudinal data but is overlooked by the authors. Another great book on longitudinal data analysis is Diggle, Liang and Zeger "Analysis of Longitudinal Data" published in 1994. There have been many advances since 1994 and Verbeke and Molenberghs cover a great deal of it. You can find my review of Diggle, Liang and Zeger on Amazon. An updated second edition of their book is in the works and will probably appear in 2001.