Non‐parametric regression with wavelet kernels

Applied Stochastic Models in Business and Industry (Impact Factor: 0.73). 03/2005; 21(2):153 - 163. DOI: 10.1002/asmb.533


This paper introduces a method to construct a reproducing wavelet kernel Hilbert spaces for non-parametric regression estimation when the sampling points are not equally spaced. Another objective is to make high-dimensional wavelet estimation problems tractable. It then provides a theoretical foundation to build reproducing kernel from operators and a practical technique to obtain reproducing kernel Hilbert spaces spanned by a set of wavelets. A multiscale approximation technique that aims at taking advantage of the multiresolution structure of wavelets is also described. Examples on toy regression and a real-world problem illustrate the effectiveness of these wavelet kernels. Copyright © 2005 John Wiley & Sons, Ltd.

11 Reads
  • Source
    • "[14] "
    [Show abstract] [Hide abstract]
    ABSTRACT: Over the last decade, the theory of reproducing kernels has made a major breakthrough in the field of pattern recognition. It has led to new algorithms, with improved performance and lower computational cost, for nonlinear analysis in high dimensional feature spaces. Our paper is a further contribution which extends the framework of the so-called kernel learning machines to time-frequency analysis, showing that some specific reproducing kernels allow these algorithms to operate in the time-frequency domain. This link offers new perspectives in the field of non-stationary signal analysis, which can benefit from the developments of pattern recognition and statistical learning theory.
    IEEE Transactions on Signal Processing 08/2007; 55(7-55):3930 - 3936. DOI:10.1109/TSP.2007.894252 · 2.79 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: The arithmetic of wind power prediction plays an important part in the development of wind power prediction. In this paper, based on the principles of support vector machine (SVM) and wavelet, the wavelet SVM model for short term wind power prediction is built up along with analyzing the characteristics of power curves of wind turbine generator systems. The operation data from a wind farm in North China are used to test the proposed model, the mean relative error of wavelet SVM model is 6.05% less than that of traditional RBF SVM model. For the time frame of one hour ahead, the average error of optimal wind turbine prediction method is 12.07%.

Show more

Similar Publications