This is a very recent and authoritative treatment of classical parametric models, starting with the general linear model and extending to generalized linear models, linear mixed models and finally to generalized linear mixed models. It also has applciations to longitudinal data analysis and prediction problems. Heavy on theory and matrix algebra but not loaded with applications. Good for a graduate course in statistics especially for Ph.D. students. It is concise covering a large range of topics in only 310 pages. An interesting feature is a chapter on computing that deals with Markov chain Monte Carlo methods in some detail. There is also a brief chapter on nonlinear models (only 5 pages) that includes an example of corn photosynthesis and also the important application to pharmacokinetic models.
The emphasis is on maximum likelihood estimation and its extensions (e.g. restricted maximum likelihood and penalized likelihood and quasi-likelihood). The authors provide an interesting perspective on the non-applicability of analysis of variance techniques in some mixed effects models.
