|
PureBytes Links
Trading Reference Links
|
Ace / Dimitris,
I think what you are looking for is an FFT that will result in a
periodogram of cycle length and amplitude.
This is not unlike what Hurst suggests and you can find relevant code
at Numerical Recipes to perform these functions simultaneously.
Fred
--- In amibroker@xxxxxxxxxxxxxxx, "Dimitris Tsokakis" <TSOKAKIS@xxxx>
wrote:
> Ace,
> I have not written any AFL, I use built-in functions in
DADiSP/SE2000 [www.dadisp.com] for this purpose.
> For example :
>
> The Curvefit SPL directory contains functions for curve fitting.
The following functions are included:
> LFIT - Fits a line to a series using the end points.
> PFIT - Performs Least Squares Polynomial fitting with error
statistics.
> SINFIT - Fits y(x) = A + B * sin(C*x + D) using the FFT.
> SINTREND - Fits y(x) = A + B*x + C * sin(D*x + E) using the FFT.
> [Copyright © 1995-2000 DSP Development Corporation]
>
> The SINTREND(W1)-TREND(W1) is a pure sinusoidal, since we remove
the linear and the curvilinear trend.
> [DADiSP uses arrays as we do in amibroker] and may give the
period/frequency.
> For complicated lines you may easily see the FFT results with the
built-in SPECTRUM() function .
> The curves are nice BUT improper for any trading attempt.
[http://groups.yahoo.com/group/amibroker/message/59137]
> Dimitris Tsokakis
> PS If you do not receive mails let me know [tsokakis@xxxx] to post
the explanatory gifs directly to your address
>
> Dmitris,
>
> Did you ever post any FFT or DFT code like you mentioned in those
> past posts? Or the sinetrend() indictor you mentioned? I'd love to
> chew on them.
>
> I used a program called O-Matrix to perform a least squares
> regression fit of A1*sin(w1)+A2*sin(w2)+....An*sin(wn) which
produced
> beautiful dominancy envelopes, but I could never get something to
> calculate relevant periods very clearly. Your sinetrend idea sounds
> similar in nature.
>
> I used O-matrix to calculate the periods by counting time between
> peak to peak and trough to trough and compiling it into a cumulative
> probability distribution, but while it did appear to work, this data
> was a little too erratic and left a lot up to the user's choice.
>
> I'd love to see any cyclic codes you've developed as I've been
> tinkering with these ideas for a couple of years now myself.
>
> -ace
------------------------ Yahoo! Groups Sponsor ---------------------~-->
Buy Ink Cartridges or Refill Kits for your HP, Epson, Canon or Lexmark
Printer at MyInks.com. Free s/h on orders $50 or more to the US & Canada.
http://www.c1tracking.com/l.asp?cid=5511
http://us.click.yahoo.com/mOAaAA/3exGAA/qnsNAA/GHeqlB/TM
---------------------------------------------------------------------~->
Send BUG REPORTS to bugs@xxxxxxxxxxxxx
Send SUGGESTIONS to suggest@xxxxxxxxxxxxx
-----------------------------------------
Post AmiQuote-related messages ONLY to: amiquote@xxxxxxxxxxxxxxx
(Web page: http://groups.yahoo.com/group/amiquote/messages/)
--------------------------------------------
Check group FAQ at: http://groups.yahoo.com/group/amibroker/files/groupfaq.html
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/amibroker/
<*> To unsubscribe from this group, send an email to:
amibroker-unsubscribe@xxxxxxxxxxxxxxx
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
|