Smoothing Techniques: With Implementation in S
Synopsis
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.
Publisher information
- Publisher: Springer New York
- ISBN: 9781461287681
- Number of pages: 262
- Dimensions: 234 x 156 x 14 mm
- Weight: 423g
- Languages: English
