Customer Reviews

Applied Regression Including Computing and Graphics (Wiley Series in Probability and Statistics)

5 star:		(3)
4 star:		(0)
3 star:		(1)
2 star:		(1)
1 star:		(2)

 

 

Ray Carroll and David Ruppert are well known research statisticians who have published many joint articles on regression, weighted regression and transformation and they have also written an excellent book together on this research topic. Stefanski has recently published several papers on measurement error models with Carroll. Here they have teamed up to write a statistics text on a unique topic. Measurement error models are common and practical when dealing with covariates that have measurement error. Least squares estimation in linear regression is based on the assumption that the predictor variables are measured without error. There are many articles and an excellent text by Fuller "Measurement Error Models", published by Wiley in 1988 that deals with the linear case. Also look at a section in Chapter 5 of Miller's "Beyond ANOVA, Basics of Applied Statistics" that refers to the problem as the error in variables problem. For the nonlinear case this is the first treatment. Well written and well documented, this text provides an up-to-date account of the theory and methods and provides real applications (e.g. the Framingham Heart Study). This is a great reference as are many of the other monographs in this series by Chapman and Hall/CRC Press. Includes bootstrap approaches in the chapter on fitting methods and models.

I used Cook/Weisberg's book to teach a required graduate data analysis class. Students found the text easy to understand. In previous classes they had used S-Plus; they raved about how easy ARC was to use --- especially since they could download it for free and use it after graudation without licensing. I found the statistics and mathematics clearly presented and the examples useful. Several of the students used ARC for the data analysis for their theses.

This is a great teaching book with a fabulous software package supporting it, by two of the best statisticians in the world.

I disagree strongly with the previous reviewer: this is much much more than a software manual. It is first a world class regression text, and secondly it supports the new methods with first-rate freeware. Who could complain?

Disclaimer: I studied under both Cook and Weisberg at Minnesota, have taught using previous versions of their books, and contributed data sets for this book.

When you touch the cover of the book, you feel the recent trend of regression. I bet you will and you should. It covers most of recent topics in 90's. ARC is a superb software and it provides so many excellent features and tools to look into the real aspects of data analysis. It is also easy enough for beginners to follow the steps.

I tried this for the textbook for my class last semester. My students loved it so much that some of them decided to pursue their career as a statistician.

But many statisticians don't appreciate what is the real value of the book. Too bad.

Cook and Weisberg provide good coverage of such a large topic. Unlike Weisberg's Applied Linear Regression book, this book goes into more theoretical detail. However, the mathematical aspects are left to the end of each chapter so the reader that's not interested in details will not have to wade through pages of equations.

This book doesn't really go into depth into what you are doing or why it works. Further the information is scattered all over the book, and while that's okay when you read it the first time - it means it's never going to be any good for a reference manual.

There's also really no reason to use ARC outside of this class, seeing as how R is also available for free. True ARC is easy to use with the built in data sets and point-and-click interface, but it also doesn't have much flexibility and I'd like to see somebody who can't figure out R try and add a data set to ARC.

Maybe this book would be okay for an intro to statistics course or something; but they actually allow masters students to take this class for graduate credit at my school. Ridiculous. The bottom line is this book will not prepare you sufficiently to do statistical analysis on your own data sets.

I took a different class with Johnson's Applied Multivariate Statistical Analysis, and although it has its own short comings, it's much better than this book.

it works through lot of examples using its (free) program ARC. The program is good and offers functionalities not available in S+ or SAS. But much of the content is on HOW to OPERATE this program. Click this button. Pull down that menu... Since my class is using S+, these information becomes irrelevant and distracting. Mixed within the usage instructions are actually formulas. Very disorganized for a first-timer like me.

Basicly, this textbook is too confined in the context of the software Arc. And I think this software isn't useful at all beyond the textbook. This textbook is not well-organized logically, and it doesn't state well the reasons of some methods and algorithms and how they come, and it has some concluson that are built upon nothing. Besides, it takes a lot of time and efforts to sort through the many useless words, sometime you just feel that the authors are just trying to increase the volume. There are some good methods, but the textbook state it in a way that after some time, you won't even think of what they are about. I think it should be more general, more reasonable, more well-organized and clearly-narrated.

Anyway, it might be good for a beginner of regression.