Open Access
December 1998 Discontinuous regression surfaces fitting
Peihua Qiu
Ann. Statist. 26(6): 2218-2245 (December 1998). DOI: 10.1214/aos/1024691468

Abstract

We suggest a three-stage procedure to recover discontinuous regression surfaces when noisy data are present. In the first stage, jump candidate points are detected using a jump detection criterion. A local principal component line is then fitted through these points in a neighborhood of a design point. This line provides a first-order approximation to the true jump location curve in that neighborhood. In the third stage, observations on the same side of the line as the given point are combined using a weighted average procedure to fit the surface at that point. If there are no jump candidate points in the neighborhood, then all observations in that neighborhood are used in the surface fitting. If, however, the center of the neighborhood is on a jump location curve, only those observations on one side of the line are used. Thus blurring is automatically avoided around the jump locations. This methodology requires $O(N(k^*)^2)$ computation, where $N$ is the sample size and $k^*$ is the window width. Its assumptions on the model are flexible. Some numerical results are presented to evaluate the surface fit and to discuss the selection of the window widths.

Citation

Download Citation

Peihua Qiu. "Discontinuous regression surfaces fitting." Ann. Statist. 26 (6) 2218 - 2245, December 1998. https://doi.org/10.1214/aos/1024691468

Information

Published: December 1998
First available in Project Euclid: 21 June 2002

zbMATH: 0927.62041
MathSciNet: MR1700229
Digital Object Identifier: 10.1214/aos/1024691468

Subjects:
Primary: 62G05
Secondary: 62G20

Keywords: Discontinuous regression surfaces , image processing , jump detection criterion , jump location curves , least squares coefficients , principal component line , threshold value

Rights: Copyright © 1998 Institute of Mathematical Statistics

Vol.26 • No. 6 • December 1998
Back to Top