- Series: Chapman & Hall/CRC Texts in Statistical Science
- Hardcover: 555 pages
- Publisher: CRC Press; 1 edition (September 24, 2012)
- Language: English
- ISBN-10: 1439815127
- ISBN-13: 978-1439815120
- Product Dimensions: 7.1 x 1.2 x 10 inches
- Shipping Weight: 2.6 pounds (View shipping rates and policies)
- Average Customer Review: 6 customer reviews
- Amazon Best Sellers Rank: #736,893 in Books (See Top 100 in Books)
Enter your mobile number or email address below and we'll send you a link to download the free Kindle App. Then you can start reading Kindle books on your smartphone, tablet, or computer - no Kindle device required.
To get the free app, enter your mobile phone number.
Generalized Linear Mixed Models: Modern Concepts, Methods and Applications (Chapman & Hall/CRC Texts in Statistical Science) 1st Edition
Use the Amazon App to scan ISBNs and compare prices.
All Books, All the Time
Read author interviews, book reviews, editors picks, and more at the Amazon Book Review. Read it now
Frequently bought together
Customers who bought this item also bought
"The book focuses on data-driven modeling and design processes, and it provides a context for extending traditional linear model thinking to generalised linear mixed modeling. This is a very sound text which teachers of any course on GLMMs should consider adopting."
―Erkki P. Liski, International Statistical Review (2013), 81
"Walter Stroup is a leading authority on GLMMs for applied statisticians, especially as implemented in the SAS programming environment. He offers a thorough, engaging, and opinionated treatment of the subject … I found the ‘fully general’ GLMM approach to modeling and design issues (Chapters 1 and 2) to be quite illuminating. … it is best to use this text in conjunction with SAS. Prospective readers without current access to SAS will be pleased to know that a reasonable level of access to SAS is now available at no cost to students and teachers on the web … If the reader prefers to work with GLMMs in the free, powerful, and state-of-the-art R environment, then he/she should supplement this text with some others that are built around R. I myself had good luck using Stroup’s text along with Julian Faraway’s two books Linear Models with R and Expanding the Linear Model with R, both published by CRC Press."
―Homer White, MAA Reviews, June 2013
"… for SAS users concerned with the analysis of trials, it is a very good resource. There are excellent discussions on many important concepts such as likelihood ratio testing and model selection criteria. PROC GLIMMIX is a powerful procedure implementing the rich family of GLMMs, and this book gives coverage to a wide variety of models with ample software illustration."
―Gillian Z. Heller, Australian & New Zealand Journal of Statistics, 2013
Top customer reviews
Here's what's good:
1. The book starts with the view that linear regression is a special case of GLMM, and builds a simple, clear, and solid exposition from this assumption. The author's case is persuasive; I think students (and others) coming at *regression* for the first time will benefit from this approach.
2. The first part of the book (esp. chapters 2) develops GLMM in the context of experimental design, with application's to "nature's experiments" -- ecology, longitudinal designs, panel data, etc. The focus on developing a good and appropriate linear predictor with a clear appreciation of the peculiarities of each specific design has strongly improved my understanding of GLMMs (and more basically of *regression*). I especially appreciated Stroup's advice on sketching a plot plan for each design. There's also a plain and persuasive argument on the importance of distinguishing between units of observation and units of replication/randomization when constructing any regression model.
3. The text is written cleanly. Few words are wasted. Stroup's dry sense of humor makes the book a pleasure to read, and fun! No BS. You get the feeling you'd like this guy.
4. Even the SAS parts were helpful (hence the surprise!) This is because Stroup works through the issues of how the computations are performed in light of the specificities of a model. Again his point is that you need to understand how to construct a good and appropriate linear predictor. No receipts for that. Oddly then, I think the SAS parts are worth reading and understanding regardless of whatever platform you use.
1. As a newbie to GLMM, I found I had to re-read some parts, and I feel I could use another read through it. Those who are coming at GLMMs for the first time should expect to move slowly. Those who understand GLMMs (or think they do) but are not used to concentrating carefully about design questions, will also want to move slowly. No canned solutions.
It is still very helpful in describing the theory behind the model formation and selection. And I really like that the example SAS codes he uses in the book are available to download (for free) from the publishers website (http://www.crcpress.com/product/isbn/9781439815120).
The reason I'm only giving it 3/5 stars is because of how many mistakes and types are in the textbook (and in the SAS code available online, which as of now was last updated Oct 22, 2012), both which can seriously impair understanding and take a while to recognize/fix.
Some key examples are:
1) The data and output data set names give in the "Chapter_1_Table_1_1" SAS example file are incorrect. The code doesn't work as written; you have to fix the names for it to run correctly.
2) The SAS code provided for Example 3.5 (using data set 3.2) does not include the variables a or b. These need to be reconstructed based on the table at the bottom of page 85. However, following the outline of the table, the SAS output results for a and b are backwards. So to get the correct output as given in the book, the treatments need to be as follows:
treatment 0 = A0, B0; treatment 1 = A0, B1; treatment 2 = A1, B0; treatment 3 = A1, B1
These examples don't include the many typos in the actual text.
Also, since I wasn't as familiar with matrix algebra, I had to look online for a lot of help and clarification, especially early on.
But since this is forcing me to brush up on basic stats knowledge, as well as helping me understand more advanced concepts, I'm still giving it a conditional recommendation.