EQUIDISTRIBUTED DESIGNS IN NONPARAMETRIC REGRESSION

In nonparametric regression estimation the best possible rate of convergence is attained when the design points are equidistributed. Whole sequences of equidistributed designs can be generated from vectors with algebraically independent components. Generators are determined which perform well alread...

Full description

Saved in:
Bibliographic Details
Published in:Statistica Sinica 2003-01, Vol.13 (1), p.129-142
Main Authors: Rafajłowicz, E., Schwabe, R.
Format: Article
Language:English
Subjects:
Continued fractions
Estimators
Experiment design
Faceted browsing
Least squares
Legendre polynomials
Linear regression
Mathematical independent variables
Observational studies
Sample size
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:In nonparametric regression estimation the best possible rate of convergence is attained when the design points are equidistributed. Whole sequences of equidistributed designs can be generated from vectors with algebraically independent components. Generators are determined which perform well already for moderate numbers of observations. The proposed design sequences are nearly optimal in classical settings and simultaneously appropriate for the nonparametric approach.
ISSN:1017-0405
1996-8507