Charlie Cornish (Department of Atmospheric Sciences, University of Washington) put together a MATLAB toolkit called WMTSA that implements much of the functionality described in the book (in particular, it provides very good support for analyses based upon the MODWT). Release 0.2.6 of the WMTSA Wavelet toolkit for MATLAB was put out on 23 Dec 2006. The toolkit is no longer under development. Among other features, the toolkit has MATLAB scripts for reproducing certain of the figures in WMTSA (as of 18 Dec 2003, these include Figures 99, 183, 184, 186, 187, 190, 194, 195, 328, 329, 330, 331 and 334).
There are plans to have a parallel implementation in the R language.
Shane Neph along with Michael Kuehn and John Stamatoyonnapoulos (Genome Sciences, University of Washington) put together a stand-alone program for computing the MODWT.
Software that is compatible with the material in WMTSA can be obtained from Brandon Whitcher's software Web page (hosted by Imperial College). His software includes
The 2.0 version of S+Wavelets has full support for the DWT, MODWT, wavelet varianace and many other analysis techniques described in WMTSA (the release date for this software has not yet been announced, but there is a rumor going around that it will appear in late 2003).
For those of you interested in Lisp (alas, not quite the null set, but getting close to it!), at some point in the future (promises, promises, promises!), we will be making available all of the wavelet code that was developed using Macintosh Common Lisp in the process of writing WMTSA (this is the code that was actually used to create nearly all the values for the figures and tables in WMTSA). This code is built upon a collection of Common Lisp routines called sapaclisp that was written for the book Spectral Analysis for Physical Applications by D.B. Percvial and A.T. Walden, Cambridge University Press, 1993. This collection is maintained on StatLib, which is hosted by the Department of Statistics at Carnegie Mellon University. While the Lisp code for wavelets certainly exists, it is sorely in need of documentation, which explains the long delay in getting it out (that and the fact that the intersection of the set of Lispers and the set of waveleticians might have just one element!).