wavethresh (version 4.6.1)

irregwd: Irregular wavelet transform (decomposition).

Description

This function performs the irregular wavelet transform as described in the paper by Kovac and Silverman.

Usage

irregwd(gd, filter.number=2, family="DaubExPhase", bc="periodic", verbose=FALSE)

Arguments

gd

A grid structure which is the output of the makegrid function.

filter.number

This selects the smoothness of wavelet that you want to use in the decomposition. By default this is 2, the Daubechies extremal phase orthonormal compactly supported wavelet with 2 vanishing moments.

family

specifies the family of wavelets that you want to use. Two popular options are "DaubExPhase" and "DaubLeAsymm" but see the help for filter.select for more possibilities.

bc

specifies the boundary handling. If bc="periodic" the default, then the function you decompose is assumed to be periodic on it's interval of definition, if bc="symmetric" then the function beyond its boundaries is assumed to be a

verbose

Controls the printing of "informative" messages whilst the computations progress. Such messages are generally annoying so it is turned off by default.

Value

• An object of class irregwd which is a list with the following components.
• C
  Vector of sets of successively smoothed versions of the interpolated data (see description of equivalent component of wd.object for further information.)
• D
  Vector of sets of wavelet coefficients of the interpolated data at different resolution levels. (see description of equivalent component of wd.object for further information.)
• c
  Vector that aids in calculation of variances of wavelet coefficients (used by threshold.irregwd).
• nlevelsWT
  The number of resolution levels. This depends on the length of the data vector. If length(data)=2^m, then there will be m resolution levels. This means there will be m levels of wavelet coefficients (indexed 0,1,2,...,(m-1)), and m+1 levels of smoothed data (indexed 0,1,2,...,m).
• fl.dbase
  There is more information stored in the C and D than is described above. In the decomposition ``extra'' coefficients are generated that help take care of the boundary effects, this database lists where these start and finish, so the "true" data can be extracted.
• filter
  A list containing information about the filter type: Contains the string "wavelet" or "station" depending on which type of transform was performed.
• bc
  How the boundaries were handled.
• date
  The date the transform was performed.

RELESASE

3.9.4 Code Copyright Arne Kovac 1997

Details

If one has irregularly spaced one-dimensional regression data (t,y), say. Then the function makegrid interpolates this to a regular grid and then the standard wavelet transform is used to transform the interpolated data. However, unlike the standard wavelet denoising set-up the interpolated data, y, values are correlated. Hence the wavelet coefficients of the interpolated will be correlated (even after using an orthogonal transform). Hence, in particular, the variance of each wavelet coefficient may well be different and so this routine also computes those variances using a fast algorithm (related to the two-dimensional wavelet transform).

When thresholding with threshold.irregwd the threshold function makes use of the information about the variance of each coefficient to modify the variance locally on a coefficient by coefficient basis.

See Also

makegrid, wd, wr.wd, accessC, accessc, accessD, putD, putC, filter.select, plot.irregwd, threshold.irregwd.

Examples

#
# See full examples at the end of the help for makegrid. 
#