|
PureBytes Links
Trading Reference Links
|
After reading the charts, wavelet decomposition comes to my mind. And "sigma bands" seem to mean low freq line(bold line) +/- some std.
kr
-----Original Message-----
From: Gary Fritz [mailto:fritz@xxxxxxxx]
Sent: Wednesday, May 21, 2003 6:17 AM
To: omega-list@xxxxxxxxxx
Subject: Re: Band-its...?
Alex wrote:
> >Of course, you couldn't use that to trade, since you have no
> >knowledge of future data points (bummer!). If that's what those
> >charts really are, then they're just dummied-up pretty sales
> >tools that couldn't actually be used for trading.
>
> Not necessarily. You don't need future points to fit a polynomial
> to data.
No, you don't. But the results you get at the "right edge of the
chart" will be very different than what you'd get with knowledge
of future data. I don't think you could get the kind of curve
he's showing without future knowledge. They're just too perfect.
It's interesting to note that none of his "sample charts" have
been updated any more recently than early March. Maybe he just
doesn't update his site very often. Or maybe he wants 6 weeks of
future knowledge to curve-fit his charts....
> >he couldn't be using a polynomial-fit curve for his trading
> >decisions.
>
> I think he could in this context. Think of it this way: for markets
> with seasonal tendencies, the tendencies average out to a sinewave
> having a period of 1 year. A sinewave for a single period can be
> well-approximated by a polynomial.
Maybe, though I seriously question whether you could get results
as perfect as he's showing. And I don't think e.g. SP500 or IBM
(which are two of the charts he has in his "Sample Charts"
section) show that perfect a seasonal tendency.
Gary
|