What is the best book about generalized linear models for novices?

I'm still pretty new to generalized linear models, and I struggle with a lot of the notation in most of the GLM texts I've picked up. Are there extremely popular GLM books that lend themselves better to readability?

• regression
• generalized-linear-model
• references

$\begingroup$ You might try perusing this thread: advanced-statistics-books-recommendation, which includes some discussion of GLiMs. In general, I'm not sure if this question is answerable w/o more info. Do you want a mathematically dense book, eg? What is your background? Etc. $\endgroup$

$\begingroup$ I would not want a mathematically dense book. I'm a population geneticist, and my interest is very much an applied one. $\endgroup$

$\begingroup$ Try this book it covers many of the models. A text that uses "general linear models" in the title is likely to be mathematical $\endgroup$

$\begingroup$ In the linked thread, I recommended Agresti's Intro book. That has relatively little math. I suspect that might be the best book for you. What is the book you're reading now? $\endgroup$

$\begingroup$ @gung Agresti's book is excellent. A little more advanced that Long. I see the 3rd edition of Agresti is out now. $\endgroup$

4 Answers 4

For a new practitioner, I like Gelman and Hill.

Ostensibly the book is about Hierarchical Generalized Linear Models, a more advanced topic than GLMs; the first section, though, is a wonderful practitioners guide to GLMs.

The book is light on theory, heavy on disciplined statistical practice, overflowing with case studies and practical R code, all told in a pleasant, friendly voice.

I have read Agresti's Intro book but found it missing key interpretations for how generalized linear model is built and how it works. For example, you may not need to know how the binomial distribution and logit link work if you only want to fit a logistic regression. However it is annoying when you have read the chapter and started to wonder about it but couldn't find it in the book.

The McCullagh and Nelder GLM book is hard to read. It contains everything you need to know but lacks the derivation for the key results.

Luckily Agresti's Categorical Data Analysis presents a good balance.