|
PureBytes Links
Trading Reference Links
|
Dan and Owen,
No I haven't tried any of the code in TASC. Candle analysis is a very
subjective art form and it's impossible to quantify for testing (kinda like
certain pattern charting software). I think it actually helps to be a bit
open minded about the formations. Sometimes, something like a "morning
star" will take four days to form, instead of three. Nison discusses
combining two candles, into one, in his second book: Beyond Candlesticks.
Although I don't practice that exactly, I do like to give the candles a
little room to do their supply and demand "thing".
Over the years, I have come up with additional patterns that I believe are
reliable. It becomes mostly pattern recognition. I've had fun with the
names: "shit sandwich"; "tri-star rocket"; and "bird of paradise imprisoned
by two big white guys".
I think it helps to pay attention to the length of the "wicks". I've
written simple code to see if this is tradable. I call it the "wicked"
series of formulas. For example, here is Wicked3:
Sum(If(C>O,H-C,H-O)-If(C>O,O-L,C-L),3)
Excuse the code, I write code like I speak Hungarian (not well). I have
Wicked's 2, 3,4, & 5. I don't know if they add much to analysis. After
viewing tens of thousands of charts, every once and a while something really
"barks" from a candle perspective.
I've attached the December Oat chart and scribbled a few things on it.
Again, it's always going to be somewhat subjective.
Oh, if you have a fantasy football team, pick up Eddie Kennison. At the end
of the year, all the Bronco fans (and the league) will be talking about his
season. Also, I'm moving back to Denver in October, just to be closer to
the stadium...the weekly commutes to the game, from Idaho, have been way too
time consuming.
Take care,
Steve Karnish, CTA
Cedar Creek Trading
http://www.cedarcreektrading.com
----- Original Message -----
From: "Owen Davies" <owen@xxxxxxxxxxxxx>
To: <metastock@xxxxxxxxxxxxx>
Sent: Thursday, September 20, 2001 8:51 AM
Subject: Re: hammer time?
> Among other questions to Steve Karnish, Dan Harels asked:
>
>
> > Have you tried to apply any of the
> > candle code indicators that have appeared in the last couple of issues
of
> > TASC?
>
> Hasn't anyone else noticed that so-called CandleCode is
> indistinguishable from a simple oscillator? Compare it with
> Mov(C,3,E) - Mov(C,10,E). If that isn't clear enough, set them up
> in Percent B format (as suggested in the recent CandleCode article),
> and see where they cross, say, the 70 and 30 levels.
>
> I can't decide whether the author did this deliberately, or just
> found a particularly complicated way to re-invent the wheel.
>
> Owen Davies
>
Attachment:
Description: "1.gif"
|