Applied Nonparametric Regression (Econometric Society Monographs) [Paperback]

Wolfgang Härdle (Author)

Book Description

Publication Date: January 31, 1992 | ISBN-10: 0521429501 | ISBN-13: 978-0521429504

Applied Nonparametric Regression brings together in one place the techniques for regression curve smoothing involving more than one variable. The computer and the development of interactive graphics programs has made curve estimation popular. This volume focuses on the applications and practical problems of two central aspects of curve smoothing: the choice of smoothing parameters and the construction of confidence bounds. The methods covered in this text have numerous applications in many areas using statistical analysis. Examples are drawn from economics--such as the estimation of Engel curves--as well as other disciplines including medicine and engineering. For practical applications of these methods a computing environment for exploratory Regression--XploRe--is described.

Editorial Reviews

Review

"Professor Härdle has provided us with an important book, one that will be appreciated both by applied statisticians who want to implement nonparametric regression techniques and by theoreticians interested in becoming knowledgeable in this growing field. Applied Nonparametric Regression is a very welcome addition to the literature." Journal of the American Statistical Association

"Nonparametric regression analysis has become central to economic theory. Härdle, by writing the first comprehensive and accessible book on the subject, has contributed enormously to making nonparametric regression equally central to econometric practice." Charles F. Manski, University of Wisconsin, Madison

"This book represents an optimally estimated common thread for the numerous topics and results in the fast-growing area of nonparametric regression. The user-friendly approach taken by the author has successfully smoothed out most of the formidable asymptotic elaboration in developing the theory. This is an excellent collection for both beginners and experts." Ker-Chau Li, University of California, Los Angeles

"This monograph on nonparametric regression presents a particularly clear and balanced view of the methodology and practice of this very important subject, and so is of use to theoreticians and practitioners alike." Peter Hall, University of Glasgow

"This book makes the main ideas and methodologies of nonparametric regression easily accessible to nonexperts, and is a valuable reference source for experts as well because of its wide scope." J.S. Marron

"...Härdle has written an important book on NPR that will undoubtedly serve as one of the standards in this field for some time to come." R. L. Eubank, Technometrics

Book Description

The applications and problems of two central aspects--the choice of smoothing parameters and the construction of confidence bounds--are focused on in an original presentation of the techniques for regression curve smoothing involving more than one variable.

Product Details

• Paperback: 352 pages
• Publisher: Cambridge University Press (January 31, 1992)
• Language: English
• ISBN-10: 0521429501
• ISBN-13: 978-0521429504
• Product Dimensions: 8.9 x 6 x 0.9 inches
• Shipping Weight: 1.2 pounds (View shipping rates and policies)
• Average Customer Review: 4.0 out of 5 stars  See all reviews (1 customer review)
• Amazon Best Sellers Rank: #1,311,139 in Books (See Top 100 in Books)

More About the Author

Discover books, learn about writers, read author blogs, and more.

Customer Reviews

Most Helpful Customer Reviews

12 of 13 people found the following review helpful:

4.0 out of 5 stars Very clear and readable account of nonparametric regression, April 10, 2001

By 

Random Thoughts (Maryland) - See all my reviews

This review is from: Applied Nonparametric Regression (Econometric Society Monographs) (Paperback)

This book deals with nonparametric regression, mainly kernel regression, though other estimators are also covered. This is a very clear and succinct discussion of the theory and application of this very important area of modern statistics. This book is one of the first few books on the subject (I saw a preprint of this book while a graduate student and bought a copy when it first appeared). Kernel estimators are known to be less efficient and suffer from boundary effects when the underlying regression is very smooth (twice differentiable). Local polynomial regression is the natural remedy and has experienced rapid developments in last 10 years. Fan and Gijbel's book in 1996 has detailed discussions on local polynomial estimation, an area which I also publish a few papers. But this book still contains substantial materials not found elsewhere. For example, this book has excellent discussion of nonparametric methods in econometric applications. I highly recommend it for its good introduction and as a valuable reference book on this subject.