Polynomial boundary treatment for wavelet regression

Authors: H-S. Oh¹; P. Naveau¹; G. Lee²

Source: Biometrika, Volume 88, Number 1, 1 February 2001 , pp. 291-298(8)

Buy & download fulltext article:

Price: $35.78 plus tax (Refund Policy)

Abstract:

To overcome boundary problems with wavelet regression, we propose a simple method that reduces bias at the boundaries. It is based on a combination of wavelet functions and low-order polynomials. The utility of the method is illustrated with simulation studies and a real example. Asymptotic results show that the estimators are competitive with other nonparametric procedures.

Keywords: Boundary problem; Polynomial regression; Polynomial-trigonometric regression; Wavelet regression

Document Type: Miscellaneous

Affiliations: 1: Geophysical Statistics Projects, National Center for Atmospheric Research, Boulder, Colorado 80307, U.S.A. e-mail: heeseok@ucar.edu e-mail: pnaveau@ucar.edu 2: Bank of Korea, Seoul 100-794, Korea e-mail: geunghee@unitel.co.kr

Publication date: 2001-02-01