|
PureBytes Links
Trading Reference Links
|
The Lomb periodogram is available for fee, but ready to use in TS from:
http://www.sirtrade.com/lomb.htm
This cycle analysis tool has a better reponse than MEM like algorithms, produces
a fairly clear main cycle period, and also provide a cycle probability
information.
Even if the Lomb periiodogram works with sparse data, it works well with equally
spaced data.
Sincerely,
Pierre Orphelin
www.sirtrade.com
TradeStation Technologies representative in France
Safir-X neurofuzzy logic trading system builder
To subscribe to our FREE trading system newsletter:
http://www.sirtrade.com/newsletter.htm
> -----Message d'origine-----
> De : Gary Funck [mailto:gary@xxxxxxxxxxxx]
> Envoye : dimanche 22 septembre 2002 00:26
> A : omega-list@xxxxxxxxxx
> Objet : RE: Lomb Periodogram
>
>
>
> Here's the description of a C program that'll do the job:
> http://www.physionet.org/physiotools/wag/lomb-1.htm
> (with a link to the C source code at the bottom of the page). With some work,
> this could be converted to a DLL that performs the analysis. It would be quite
> slow in EL, no doubt.
>
> What do you see as the advantages of the Lomb algorithm versus FFT or some
> other spectral analysis algorithms? From what I've read, there's some debate
> as to whether EOD data should be viewed as equally spaced, or whether
> it should
> be translated into the actual calendar time domain (requiring
> extrapolation for
> the missing points, or the use of an technique like the Lomb algorithm). I've
> never seen this topic addressed in the literature.
>
> This top-level page has several other useful signal processing algorithms.
>
>
|