Analyzing Multivariate Data (with CD-ROM) (Duxbury Applied Series) | 
 enlarge | Authors: James Lattin, Douglas Carroll, Paul Green
Publisher: Duxbury Press
Category: Book
List Price: $164.95
Buy New: $50.21
You Save: $114.74 (70%)
New (22) Used (12) from $50.21
Rating:  5 reviews
Sales Rank: 566763
Media: Hardcover
Edition: 1
Pages: 560
Number Of Items: 1
Shipping Weight (lbs): 2.3
Dimensions (in): 9.3 x 7.4 x 1.1
ISBN: 0534349749
Dewey Decimal Number: 519.535
EAN: 9780534349745
Publication Date: December 3, 2002
Availability: Usually ships in 1-2 business days
Shipping: Expedited shipping available
Condition: Brand New Hardcover US Edition w/CD Free Tracking Ref.935
| |
| Similar Items:
|
| Editorial Reviews:
Product Description
Offering the latest teaching and practice of applied multivariate statistics, this text is perfect for students who need an applied introduction to the subject. Lattin, Carroll, and Green have created a text that speaks to the needs of applied students who have advanced beyond the beginning level, but are not advanced statistics majors. The text provides a three-part structure. First, the authors begin each major topic by developing students' statistical intuition through applications. Then, they providing illustrative examples for support. Finally, for those courses where it will be valuable, they describe relevant mathematical underpinnings with vectors and matrix algebra. Additionally, each chapter follows a standard format. This format begins by discussing a general set of research objectives, followed by illustrative examples of problems in different areas. Then it provides an explanation of how each method works, followed by a sample problem, application of the technique, and interpretation of results.
|
| Customer Reviews:
Excellent bridge between practical use and rigor. November 23, 2004
BlueDaisy (Northern US)
5 out of 5 found this review helpful
Analyzing Multivariate Data by Lattin, Carroll, and Green fills an important niche in practical statistics for the applied researcher. For a purely conceptual introduction, Sam Kash Kachigan (Multivariate Statistical Analysis) or Joseph Hair et al (Multivariate Data Analysis) are excellent references; for more rigorously mathematical approaches, Johnson & Wichern's and Donald Morrison's texts are excellent. Analyzing Multivariate Data bridges those two poles. The rigorous texts teach how to twirl matrices around like a flaming baton, but unless the researcher has insight into the art and science of the applications of the multivariate methods little will be understood. This is particularly important if the researcher uses point-and-click stat packages. For the applied researcher and grad student who want to know how and when to use the multivariate methods but want accessible supporting mathematics as well, this book is exceptional. I use it to explain the applications of the mathematically intensive methods I am learning in a grad multivariate class.
The first statistics book i've ever read for pleasure... July 1, 2005
Cyanide Bunny (中国)
1 out of 1 found this review helpful
I bought this book as a cheap international edition because, er, it was cheap. It turned out to be an excellent reference for understanding many useful concepts in multivariate analysis, and as another reviewer said, it clarifies many of the ideas with real examples. Because the style is so accessible, I ended up reading it as much out of interest as the need to learn about all the techniques. It lacks the rigor of some other texts, but there are plenty of them out there to refer to for more concrete proofs if you need them. Read this to understand the techniques...
A statistics book that is easy to read and understand. April 29, 2003
9 out of 13 found this review helpful
I had the great pleasure of learning from this book when it was still in manuscript form. For the first time in my long history as a student I found that I understood the text at first reading! The chapters generally present two approaches: the technical (scary!) and the commonsense -- with the latter truly written in a commonsensical fashion. In fact, it is so well written that it is possible to go over the material on one's own. The book is pricey, but it delivers on its promise.
In depth, but easy to understand. January 18, 2008
Genevieve Hayes (Australia)
"Analyzing Multivariate Data" is not just one the of the best multivariate data analysis texts available, but also one of the best structured text books that I have ever seen. After providing an introductory discussion of the prerequisite materials (namely matrices and regression analysis), Lattin, Carroll and Green cover all of the expected areas of multivariate analysis, specifically: principal components analysis, factor analysis, multidimensional scaling, cluster analysis, canonical correlation, structural equation models, analysis of variance, discriminant analysis and logit choice models. What makes this book stand out from other, similar, text books is how the material is structured. In each chapter, the authors explain all of the relevant mathematics behind the techniques, but also explain the concepts in laymans terms. Important questions are posed and answered, such as "when should this technique be used?" and worked examples are given. Furthermore, at the end of each chapter, the main points are summarized in point form and a large number of exercises are provided. The only drawback I can see is that the solutions to the exercises are not provided.
Even though this text covers material at a reasonably high level (this book would be suitable for later year undergraduates or graduate students), this is still an extremely user-friendly textbook. I just wish more texts were this well written.
 Review from a Biometric Student August 24, 2003
John Young (Peking,China)
1 out of 9 found this review helpful
Conclusion: useful but still not so strict . If you need a systematic proof and strcuture , you may need other statistics text books.Detail:This book is an application-oriented one:in each section, a method discuessed and then a case study of real data set followed by the technological concern ,such as significant and the the violation to the assumption of each method. Especially, the 'intuitive' explain of 'how it works' is also helpful and still remains systematic. However,since the lack of theoratic proof of method , the model is still not so strict ,thus the the completeness 'intuitive'explaim is weakened.
|
|
|
