Provides a unified account of the most popular approaches to nonparametric regression smoothing. This edition contains discussions of boundary corrections for trigonometric series estimators; detailed asymptotics for polynomial regression; testing goodness-of-fit; estimation in partially linear models; practical aspects, problems and methods for confidence intervals and bands; local polynomial regression; and form and asymptotic properties of linear smoothing splines.
"Praise for the first edition. . . …an excellent textbook for a graduate statistics course on nonparametric regression or as an introduction for someone wanting to learn about this area. "
---Journal of the American Statistical Association
"…an important contribution to the dissemination of knowledge about nonparametric regression….Eubank's effort is highly tractable and contains many pragmatic tips without unduly sacrificing theoretical rigor. …the second edition does include the many advances in nonparametric regression that have occurred since the eighties…. It is worthwhile to highlight the following merits of this book. Each chapter opens with a motivation for the topic understudy, and it identifies similarities with or differences from the classic parametric regression approach. Discussions of computation aspects are included wherever appropriate. Illustrative examples and exercises are abundant in each chapter…. …the book was very well written, and…recommend[ed] to serve as reference and/or a textbook. "
"The book is a masterpiece: it is very clearly written, easy to follow and summarizes prominent research in approximation theory, based on mathematical statistics. …strongly recommend[ed]…. "
What is a good estimator?; series estimators; kernel estimators; smoothing splines; least-square splines.