Customer Reviews

Applied Smoothing Techniques for Data Analysis: The Kernel Approach with S-Plus Illustrations (Oxford Statistical Science Series)

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

 

 

This book covers parametric and nonparametric regression, logistic regression, density estimation and smoothing, semiparametric and additive modeling. The underlying approach is to use kernel methods for smoothing and density estimation. They point to bootstrap methods as a way to put confidence bands on the density estimates. SPlus software is used to illustrate the techniques. The exposition is clear and concise and coverage is fairly broad. Interesting topics such as time series, longitudinal data analysis comparing regression curves and exploratory methods are all covered.

This book is probably my #1 reference for kernel smoothing. It covers enough theory (bias and variance calculations) without getting stuck in mathematical details. Also the code samples are very useful and are carefully interspersed throughout the text. I think this is a great book for the applied statistician who wants to get up and running with kernel smoothing methods.