PureBytes Links
Trading Reference Links
|
Hi Ming,
The problem with it is due to rounding. We plot only for a finite
number of points. Say you enter a period of 10, then we repeat the
wave every ten points, and we only plot 10 points for it. If we
plotted more points, say 0, 0.25, 0.5, 0.75, 1.0 etc, in stead of only
for 0, 1, 2 etc, then we would be able to more clearly see the peaks
and troughs. Here the max is at say 5.5, and at 4 and 5 we have equal
values since they are spaced equidistantly from the actual max.
We developed our own software to do Fourier, and it does things
exactly the way we want to. You can also do Fourier in something like
Excel. It will extract the coefficients of the Fourier series, from
which you can extract the most important cycles, the explanatory power
of each etc etc. If you can code then it is best to read up on it and
do it yourself.
If you have a long time series history and a fairly ok fit, then
things don't change that much as you add points. If they do, then
either the fit is too poor or you have too little history.
Regards
MG Ferreira
TsaTsa EOD Programmer and trading model builder
http://www.ferra4models.com
http://fun.ferra4models.com
--- In equismetastock@xxxxxxxxxxxxxxx, "tan ming"
<reminiscenostalgia@xxxx> wrote:
> Hi MG, tks for your great code on the sin wave. However when I plot the
> indicator and increase the resolution (by increase its frequency) I
noticed
> that the peak and tough have two equal values, i.e the there are two
points
> at the peak have equal value. Is there some computation error since
there
> suppose to have one highest and lowest point in the peak and tough
> respectively. If there are two points on peak and tough, would it
affect the
> period involved?
>
> Yes I am interested in fourier analysis and indeed did some research
on this
> area. I am trying to locate a program which can produce and save data,
> including the amp, period and phase after doing the analysis. Being
able to
> plot the fourier density graph will be a plus. Do u have any gd
> recommendation?
>
> As u know fourier analysis is not static in its analysis. When new data
> points are added, the analysis output will have variation. And this
seems to
> be a problem when trying to implement the result into metastock
platform.
> What do u think and any solution to it?
>
> Ming
>
> _________________________________________________________________
> Block pop-up ads with MSN Toolbar. http://toolbar.msn.com.my/
Yahoo! Groups Links
<*> To visit your group on the web, go to:
http://groups.yahoo.com/group/equismetastock/
<*> To unsubscribe from this group, send an email to:
equismetastock-unsubscribe@xxxxxxxxxxxxxxx
<*> Your use of Yahoo! Groups is subject to:
http://docs.yahoo.com/info/terms/
|