Smoothing Techniques [electronic resource] : With Implementation in S /
The author has attempted to present a book that provides a non-technical introduction into the area of non-parametric density and regression function estimation. The application of these methods is discussed in terms of the S computing environment. Smoothing in high dimensions faces the problem of data sparseness. A principal feature of smoothing, the averaging of data points in a prescribed neighborhood, is not really practicable in dimensions greater than three if we have just one hundred data points. Additive models provide a way out of this dilemma; but, for their interactiveness and recursiveness, they require highly effective algorithms. For this purpose, the method of WARPing (Weighted Averaging using Rounded Points) is described in great detail.
Main Authors: | , |
---|---|
Format: | Texto biblioteca |
Language: | eng |
Published: |
New York, NY : Springer New York,
1991
|
Subjects: | Mathematics., Applied mathematics., Engineering mathematics., Applications of Mathematics., |
Online Access: | http://dx.doi.org/10.1007/978-1-4612-4432-5 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | The author has attempted to present a book that provides a non-technical introduction into the area of non-parametric density and regression function estimation. The application of these methods is discussed in terms of the S computing environment. Smoothing in high dimensions faces the problem of data sparseness. A principal feature of smoothing, the averaging of data points in a prescribed neighborhood, is not really practicable in dimensions greater than three if we have just one hundred data points. Additive models provide a way out of this dilemma; but, for their interactiveness and recursiveness, they require highly effective algorithms. For this purpose, the method of WARPing (Weighted Averaging using Rounded Points) is described in great detail. |
---|