wt.filter {wavelets}

Wavelet Transform Filter

Description

Generates a wavelet transform filter.

Usage

wt.filter(filter="la8", modwt=FALSE, level=1)

Arguments

filter

A character string indicating which wavelet transform filter to compute or a numeric vector of wavelet (high pass) filter coefficients (not scaling (low pass) coefficients). If a numeric vector is supplied, the length must be even.

modwt

A logical value indicating whether to compute the maximal overlap discrete wavelet transform filter.

level

An integer value indicating the level of the wavelet filter to compute.

Details

The character strings currently supported are derived from one of four classes of wavelet transform filters: Daubechies, Least Asymetric, Best Localized and Coiflet. The prefixes for filters of these classes are d, la, bl and c, respectively. Following the prefix, the filter name consists of an integer indicating length. Supported lengths are as follows:

Daubechies

2,4,6,8,10,12,14,16,18,20.

Least Asymetric

8,10,12,14,16,18,20.

Best Localized

14,18,20.

Coiflet

6,12,18,24,30.

Thus, to obtain the Daubechies wavelet transform filter of length 4, the character string "d4" can be passed to wt.filter.

This naming convention has one exception: the Daubechies wavelet transform filter of length 2 is denoted by haar instead of d2.

Value

Returns an object of class wt.filter, which is an S4 object with slots

L

An integer representing the length of the wavelet and scaling filters.

h

A numeric vector of wavelet filter coefficients.

g

A numeric vector of scaling filter coefficients.

wt.class

A character string indicating the class of the wavelet transform filter. Possible values are "Daubechies", "Least Asymetric", "Best Localized", and "Coiflet". If the wt.filter object is generated from a numeric vector of wavelet coefficients, wt.class is "none".

wt.name

A character string indicating the name of the wavlet filter as listed in the Details section, above. If the wt.filter object is generated from a numeric vector of wavelet coefficients, wt.name is "none".

transform

A character string indicating whether the resulting wavelet transform object contains DWT or MODWT coefficients. Possible values are "dwt" and "modwt".

Note

The notation h and g for wavelet and scaling coefficients, respectively, follows Percival and Walden (2000). In other texts and articles the reverse notation is often adopted.

Author(s)

Eric Aldrich. [e-mail].

References

Percival, D. B. and A. T. Walden (2000) Wavelet Methods for Time Series Analysis, Cambridge University Press.

See Also

wt.filter.qmf, dwt, modwt.

Examples

wt.filter("la14")

wt.filter(1:10, modwt=TRUE)