Regression Models for Categorical Dependent Variables Using Stata, Third Edition

The book itself is very useful. However, the Kindle edition has a problem displaying equations. Some are cut, and some are replaced by white boxes. Buy a hard copy.

This is an excellent primer for using Stata, and more generally for understanding regression and logistic models, which this book does an excellent job of explaining. To be honest, this book would be equally as useful as a general statistics book, as it often takes the time to explain all of the nuances of making highly advanced models in Stata, as well as how to avoid the most common pitfalls.It may feel like a massive tome, but the two indices (one for Stata commands and one by layman's terms) make it very navigable as a reference book.

Incredibly useful text. It is somewhat verbose, but it covers everything in good detail and includes a huge number of actual Stata commands and the relevant output, so you can follow along.

This is a very good book to both learn about limited dependent variables and to learn about Stata formidable tools.The book covers also some theoretical material for the regression models included but the best would be to get the 1997 book dedicated only to that.

If you still cannot understand regression models for categorical data after you've read this book, you would better off considering another line of job. Statistics doesn't fit you!

I learned a great deal from this book. Almost certainly valuable even if you are not a Stata user (I am).

good

ok, its clear

Brilliant book, I use it all the time

I use STATA SE14. The book is well written and takes you through the use of regression models with clear examples and ways of interpreting them. If you run these models this book is very useful and if you are so lucky that your data look perfect (far from reality in everyday life), this book gives you insight into most aspects of regression. So far so good then and I would have given it full marks if it not were for a non-negligible issue.Intact, I believe that a huge shortfall is that it does not provide any information and guidance to deal with issues of complete/quasi complete separation you might encounter in your data. The Authors only mention in a brief paragraph general ways for avoiding problems with MR. The above issue can affect even big datasets with adequate sample size (like mine) and detailed explanation of ways of dealing with the problem is lacking in the book. You will have to go through many papers and STATA blogs to find the solution to your data (in the case in point, by using Firthlogit). Even in the less specialised Field's "Discovering Statistics" the issue of separation is explained in some detail though also in that case Field fails to provide a guidance in MR cases literally "of difficult solution". This is most likely due to SPSS having limitations with the analyses of more complex regression models and the use of add-ons which are not even written by IBM but R developers instead.

A very nice and comprehensive book for understanding analysis of categorical variables using stata. Could have been better if it was bit more lucid in explanations.

I am glad I read reviews about this book as I cannot emphasise enough how good this book is. There are many tips about using Stata in general that really make the book a standalone resource. I would say that this book is my equal favourite with Hamilton's.