Wavelets in statistics: beyond the standard assumptions

Abstract
This paper, attention has been focused onmethods that treat coefficients at least as if they were independent. However, itis intuitively clear that if one coefficient in the wavelet array is nonzero, then it ismore likely (in some appropriate sense) that neighbouring coefficients will be also.One way of incorporating this notion is by some form of block thresholding, wherecoefficients are considered in neighbouring blocks; see for example Hall et al.

This publication has 28 references indexed in Scilit: