A GEOMETRICAL METHOD FOR REMOVING EDGE EFFECTS FROM KERNEL-TYPE NONPARAMETRIC REGRESSION-ESTIMATORS Academic Article

abstract

• We introduce a simple geometric method for removing edge effects from kernel-type nonparametric regression estimators. It involves reflecting the data set in two estimated points, thereby generating a new data set with three times the range of the original data. The usual kernel-type estimator may be applied to the new, enlarged data set, without any danger of edge effects. This technique is applicable generally to both regularly spaced and randomly spaced designs and admits a natural analog of leave-one-out cross-validation. The new cross-validation algorithm may be extended to the very ends of the design interval, unlike its more conventional counterpart, which must be downweighted at the ends of the interval to avoid edge effects. 1991 Taylor & Francis Group, LLC.

authors

• Wehrly, Thomas

published proceedings

• JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION

author list (cited authors)

• HALL, P., & WEHRLY, T. E.

citation count

• 84

complete list of authors

• HALL, P||WEHRLY, TE

publication date

• September 1991

publisher

• Taylor & Francis  Publisher

published in

• JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION  Journal

keywords

• Boundary Effects
• Cross-validation
• Design
• Mean Squared Error
• Reflection

Digital Object Identifier (DOI)

• 10.1080/01621459.1991.10475092

start page

• 665

end page

• 672

volume

• 86

issue

• 415

URL

• http://dx.doi.org/10.1080/01621459.1991.10475092