Automatic Polynomial Wavelet Regression for Signals with Non-Gaussian Noises
Alsaidi M. Altaher, Mohd Tahir Ismail · International journal of mathematics and computation · 2010
As treatment for boundary problem present in wavelet regression, polynomial wavelet regression is usually considered. However, this estimator is particularly efficient at estimating spatially inhomogeneous signals when the noise is Gaussian. In this paper, we extend the validity of polynomial wavelet regression for signals with non-Gaussian noises such as heavy and mixture normal noises by keeping the low order polynomial components and using the robust threshold procedure of Oh et al. (2009) instead of the classical ones found in literature. A simulation study and an application to real data are conducted to evaluate the superiority of the proposed method.