Perform wavelet shrinkage using data-analytic, hybrid SURE, manual, SURE, or universal thresholding.
da.thresh(wc, alpha = .05, max.level = 4, verbose = FALSE, return.thresh = FALSE)
hybrid.thresh(wc, max.level = 4, verbose = FALSE, seed = 0)
manual.thresh(wc, max.level = 4, value, hard = TRUE)
sure.thresh(wc, max.level = 4, hard = TRUE)
universal.thresh(wc, max.level = 4, hard = TRUE)
universal.thresh.modwt(wc, max.level = 4, hard = TRUE)wavelet coefficients
level of the hypothesis tests
maximum level of coefficients to be affected by threshold
if verbose=TRUE then information is printed to
the screen
threshold value (only utilized in manual.thresh)
Boolean value, if hard=F then soft thresholding is used
sets random seed (only utilized in hybrid.thresh)
if return.thresh=TRUE then the vector of
threshold values is returned, otherwise the surviving wavelet
coefficients are returned
The default output is a list structure, the same length as was input, containing only those wavelet coefficients surviving the threshold.
An extensive amount of literature has been written on wavelet shrinkage. The functions here represent the most basic approaches to the problem of nonparametric function estimation. See the references for further information.
Gencay, R., F. Selcuk and B. Whitcher (2001) An Introduction to Wavelets and Other Filtering Methods in Finance and Economics, Academic Press.
Ogden, R. T. (1996) Essential Wavelets for Statistical Applications and Data Analysis, Birkhauser.
Percival, D. B. and A. T. Walden (2000) Wavelet Methods for Time Series Analysis, Cambridge University Press.
Vidakovic, B. (1999) Statistical Modeling by Wavelets, John Wiley \& Sons.