Customer Reviews

Regression Analysis by Example, 3rd Edition

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

 

 

Its strength is in the examples. It is true that the book also employs formulas, but most of them are informative and fairly easy to follow if you have a moderate background in statistics. I would recommend this book for an intermediate course in econometrics, mainly because it offers a comprehensive view of the subject and ways to apply analysis tools. I give it five stars for the presence of applied data.

Over the years, I have read many books on regression analysis. This book is one of the most lucidly written book on the subject. It achieves a fine balance between theoretical explanation and application of the method to real world problems. It requires some background knowledge, i.e. first college course of statistics, but it is not too dense. I am using this as a companion to a book on regression analysis by Douglas Montgomery (another excellent text). The best part is that the data and computer programs in SAS, R etc. are available on the UCLA website. All in all, a great resource.

I used this book for an introductory/intermediate course in regression.

Although the authors presented a lot of material and I did find it informative, I found myself having to refer to other texts for a better explanation of about half of the material presented.

To me it read like an academic paper. It seemed that all the variables were explained only once in the text and used throughout the book without a central glossary or formula page. There also seemed to be a lot of subscripts and superscripts to the variables, as well as bouncing back and forth between data examples, which made following the concepts difficult.

There was a lot of information packed in the book's pages, but it was real difficult reading and trying to comprehend past the second chapter.

I used the first edition of this book (Chatterjee and Price, 1975) when I was a graduate student in the late '70's. The text was not as easy to read as the title might suggest, but it was much more accessible than commonly used alternatives, such as Johnston's introduction to econometrics and Draper and Smith's regression book. Chatterjee and Price's last chapter, however, seemed misplaced and was a bit of a mystery. It dealt with multicollinearity and leaned heavily on the work of a British statistician named Silvey. His prescription for dealing with multicollinearity was (to closely paraphrase) collect additional data in the direction of an eigenvalue that is small but not equal to zero. Fortunately, it was not necessary to master the last chapter to take nearly the full measure of Chatterjee and Price's offering.

In subsequent editions nothing quite so formidable as Silvey's account and prescripton has been included. The book remains accessible and informative. It includes useful examples, though for students of the social and behvaioral sciences, the examples may seem substantively unfamiliar. Nevertheless, the examples are understandable, and they do a reasonbly good job of clarifying the authors' explanations. My only concern is that some of the illustrative material inclued early in the book seems better served through use of time-series analysis rather than cross-sectional ordinary least squares, as applied by the authors.

Some readers may like the authors' intuitive approach to regression diagnostics. For example, rather than employ the usual formula to identify a Cook's Distance value that betokens an unduly influential observation, the authors' suggest visual examination of distributions of the Cook's Distance statistic to find unusually large values. I think their approach has merit, and I prefer it; I have found the simple formula used with Cook's Distance to be much too conservative, identifying as potentially troublesome observations that seem to me to be unproblematic. Nevertheless, others may find the authors' intuitive approach too informal.

For most readers who have had a first course in statistics, Regression Analysis by Example will be a useful book. Annoying typographical errors that were fairly numerous in the first printing have been corrected, and the authors, very much to their credit, respond promptly to emailed questions.

I have used the third edition of this venerable textbook, along with ancillary material on logistic and ordinal regression and generalized tobit, in teaching regression to graduate students in education. In spite of their initial trepidation, students responded favorably to the Chatterjee, Hadi, and Price's presentation, and were able to use multiple regression in analyzing data for theses and dissertations.

The book arrived on time, excellent packaging and great condition. It has the new book smell I absolutely love! Highly recommend these sellers!

The book never came in the mail and the shipper won't refund my money without the book being returned to him. I went to the post office and they didn't have the book. Now I'm out $80 without a book.