Nonparametric Regression in Environmental Statistics

This article provides an introduction to the major types of nonparametric regression techniques, including kernel, spline and orthogonal projection methods. Practical aspects of the methods and their applicability in environmental statistics are emphasized through examples and discussion. Topics covered include bandwidth selection, non-parametric regression in multiple dimensions, and methods for handling data exhibiting correlated errors. Extensions to generalized regression models and semiparametric regression are also discussed.