Hello, Group
I would like to know
whether
anybody has conducted a
research (statistical
research) for Fib
levels?
I did
it myself,
and the
results that
I have got
are in contradiction to the general
approach.
The idea is simple: we
calculate the zigzag to specify the turning points and analyse ratios, like
this:
To find the length of down swing D-E
(after 7.17.2007), we analyse all proportions
between Up swing (A-B) and Down swing (B-C), we do it for all historical
data.
If Fib levels work, we are
expecting to get the clusters around classical ratios (100%, 50%,62%, 38% ...),
because these levels according to
the theory are the most probable
levels where the price may change its trend. Mathematically there is no another
way.
I was really surprised when I got this
(Dow 1985-2007, 5% zigzag used):
The classical levels are marked by
purple vertical
lines.
For the whole Dow price
history (available from
1885), I have got
this distribution (it is covers 336 turning
points):
I played with different financial
instruments, zigzag %, different proportions. Practically everywhere I have
found 100% level, but everything else is pretty far from the classical
approach.
The positive result of my enterprise is that the
clustering exist. There
is some order in turning points distribution, not the chaos
only.
I think there is a kind of
quantum mechanizm working here (like
in quantum mechanics). The most
probable turning points levels look like
these (red
regions):
Visually it looks like
spectral lines.
These are links re my research:
Best
regards.
Sergi.
----- Original Message -----
Sent: Monday, August 20, 2007 6:39
AM
Subject: Re: [RT] DXU7
Ira-
Tks for your comments. Looks like I need to take more
time
to learn all about Fib levels and price. It
obviously
works well for
you and the students and that's good. I
could never work out if price was
supposed to stop at 38%
50% or 62%; so for me Fib levels have been a
dilema. Pice
your level and hope you're right about which level
you've
chosen. I know you must look at Fib levels along with
other
indicators as well as filters.
Chas
----
On Sun, 19 Aug 2007
12:14:27 -0700
"Ira" <mr.ira@xxxxxxxxxxnet>
wrote:
> The method is proprietary so I don't have an explanation
>for how it works. That would be unfair to past students.
> The
charts themselves, are as accurate as Ensign and
>DTNiq can make them.
I do put the levels on the charts
>when I post charts.
>
> I usually post charts on the week ends so that you can
>print
them and then mark them during the week to see if
>they do what they
are supposed to. The commentary and
>analysis are mine and if you have
a system that you are
>using you can use the information to reinforce
your
>findings or create a certain amount of doubt that will
>cause you to recheck your conclusions. It is you and
>your
system that is trading. I am not teaching a trading
>method on this
site.
>
> Thank you for your positive comments and I hope that
the
>site proves useful and profitable for you in the future.
>
> Ira
>
>
> ----- Original Message -----
>
From: Charles Meyer
> To: realtraders@yahoogroups.com
> Sent: Sunday, August 19, 2007 11:16 AM
> Subject: Re: [RT]
DXU7
>
>
> Ira-
>
> I don't offer a website
or tutoring; so maybe this
>isn't
> an appropriate response.
Having said that, here
> is my 2 cents about your website:
>
> * I love your comments. Good stuff for perspective,
>etc.
> * There is no commentary about how you arrive at the
> price levels on your various charts, so
> that isn't very
helpful; or useful.
>
> Tks...
>
> Chas
>
----
>
> On Sun, 19 Aug 2007 08:46:05 -0700
> "Ira" <mr.ira@xxxxxxxxxxnet>
wrote:
> > I guess that I will keep making spelling errors for
>the
> >web site. The correct address is
>
>www.thetradersguide.net.
> >
> > I have just
added the September Dollar contract to
>the
> >site. So the
indices, Crude oil and the dollars are
> >posted along with the
economic releases for the week.
> >
> > I will try to do
the bonds a little later.
> >
> > Ira.
>
>
>
>
