Wavelets

How do you know when a trend has started? Ended? This forum is for discussions about trend indicators and signals.
Post Reply
bruce_wayne
Contributor
Contributor
Posts: 3
Joined: Tue Mar 16, 2004 6:10 pm

Wavelets

Post by bruce_wayne » Tue Mar 16, 2004 7:01 pm

Does anyone have Wavelet experience. I am wanting to reproduce a wavelt filter, specifically Symlet4, level 5. I have the specicifcation from Matlab but wish to reproduce this filter outside the Matlab environment.
Has anyone worked on this type of issue before.

Many thanks


:?:

Hiramhon
Roundtable Fellow
Roundtable Fellow
Posts: 98
Joined: Fri May 09, 2003 12:45 am

Post by Hiramhon » Tue Mar 16, 2004 10:52 pm

Dick Grayson / Commissioner Gordon,

You're going to need to get down and get nasty, starting with the Daubechies coefficients and working outwards. I'd recommend Numerical Recipes (of course), and also Barbara Burke Hubbard's text The World According To Wavelets (ISBN 1-56881-072-5).

bruce_wayne
Contributor
Contributor
Posts: 3
Joined: Tue Mar 16, 2004 6:10 pm

Post by bruce_wayne » Wed Mar 17, 2004 12:05 am

Thanks a lot - please calrify how 'down and dirty' we have to get to replicate the filter in the absence of matlab. Do you have a simple DB example to illustrate ?

many thanks

lperepol
Senior Member
Senior Member
Posts: 30
Joined: Thu Feb 26, 2004 12:50 pm
Location: Castlegar, BC, Canada
Contact:

Re: Wavelets

Post by lperepol » Wed Mar 17, 2004 1:34 am

bruce_wayne wrote:Does anyone have Wavelet experience. I am wanting to reproduce a wavelt filter, specifically Symlet4, level 5. I have the specicifcation from Matlab but wish to reproduce this filter outside the Matlab environment.
Has anyone worked on this type of issue before.
Bruce you might want to look at R http://www.r-project.org/
and then the CRAN package http://probability.ca/cran/ RWave

Here is the description:

Rwave is a library of R functions which provide an environment for the Time-Frequency analysis of 1-D signals (and especially for the wavelet and Gabor transforms of noisy signals). It was originally written for Splus by Rene Carmona, Bruno Torresani, and Wen L. Hwang, first at the University of California at Irvine and then at Princeton University. Credit should also be given to Andrea Wang whose functions on the dyadic wavelet transform are included. Rwave is based on the book: "PRACTICAL TIME-FREQUENCY ANALYSIS: Gabor and Wavelet Transforms with an Implementation in S", by Rene Carmona, Wen L. Hwang and Bruno Torresani, Academic Press (1998).

bruce_wayne
Contributor
Contributor
Posts: 3
Joined: Tue Mar 16, 2004 6:10 pm

Post by bruce_wayne » Wed Mar 17, 2004 9:00 pm

Thanks a lot. I am quite aware of R, as forerunner of S and S-Plus, as well as Cran - I replaced both with Matlab ! The issue here is not so much in the testing and specification, rather the reproduction of the filter once the coefficients have been tested for and are known via this testing.

I will have a look at r again -I have seen it for about2 years now.

thanks

Post Reply