|
PureBytes Links
Trading Reference Links
|
Just making a quick jump into and out of the probability frey on price level
occurrance. From personal experience I have observed that there are three
levels each day that are visited by price crossings more than others. They
are a fraction of the day's high and low at that time. For example
0.333*(HighD(0)-LowD(0))+LowD(0) is one level. Another is the 0.5*
range+LowD(0) and another is 0.666*range+LowD(0). To prove this can be done
with a price distribution tool that counts the times price falls within a
price bin usually defined by the tick size, 0.1 or 0.25 or 0.5 etc.
Attached is a picture of the NQM2 for today showing the price distribution
lobes are longest at the 0.666 dynamic level in the morning and again in the
afternoon following the lunchtime seesaw. Times are in Pacific on the
chart. Fibo types like to use 0.382 and 0.5 and 0.618, but personally I see
the 0.333 and 0.666 dynamic levels are hit more often during each day. This
is one way to get a handle on the issue. Bullish day's are characterized by
a "P" distribution where the lobe is at the 0.666 level to 0.75. Bearish
days have a "b" distribution with the lobe at 0.333 to 0.25 and symmetrical
days or "normal" days have the lobe at the 0.5 level. Some days are
trending days with a more or less even distribution with a small lobe near
the upper or lower eighth levels.
bobr
----- Original Message -----
From: "ztrader" <ztrader@xxxxxxxxx>
To: <realtraders@xxxxxxxxxxxxxxx>
Sent: Wednesday, April 10, 2002 5:20 PM
Subject: Re[2]: [RT] random price moves in ES?
> On Wednesday, April 10, 2002, 4:42:21 PM, Ira Tunik wrote:
>
> IT> Nothing works perfectly all the time,
>
> This is a mathematical issue and is subject to mathematical analysis.
> Would you agree? Can't we try to establish the statistical occurence
> of some event we are describing? Sort of like the odds people
> calculate in some game of chance. Why can't we do the same kind of
> calculations about Fibo retraces? It's just math.
>
> IT> that is why someone invented the stop order.
>
> Yep. This is a *different* issue. This is a *trading* technique that
> gets invoked when you are on the wrong side of the *math*.
>
> I am ***NOT*** looking at trading systems. I am trying to keep the
> underlying math basis separate from the trading part. My question has
> nothing to do with trading - just math. I am just asking about the
> statistical occurrence of turning points at Fib levels. I am NOT
> including how to trade them.
>
> IT> If you are looking for a mathematical method that will give you a
> IT> 100% probability or predictability, there isn't one.
>
> Of course not. I'm not sure how you came to that conclusion.
>
> IT> I have a system that gives me an 80% probability and has
> IT> done so over the years.
>
> Does this mean that 80% of your trades are profitable? If so, that is
> NOT the kind of data I am discussing.
>
> If you are saying that the underlying *mathematical* basis of your
> system has an 80% probability of being correct, independently of any
> *trading* techniques, then that is an entirely different matter. Do
> you see the difference I am referring to? If so, which of these
> different things are you referring to - trades or math?
>
> IT> Being right all the time isn't the key to profitability.
>
> Of course. First, we're not talking about "all the time", just "how
> often". I think the stats of your underlying 'basis' do influence the
> design of the system, though. Would you not agree?
>
> IT> Someone wrote that if you took all the money and assets in world
> IT> and distributed them evenly among everyone on earth that those
> IT> that have all the money now would end up with all the money then.
>
> I can believe that. :-) Still, I don't see what that has to do with my
> question about a mathematical probability. It would seem that one
> could work through a problem in statistics class without having to
> take into account 'all the money in the world'. :-)
>
> ztrader
>
>
> To unsubscribe from this group, send an email to:
> realtraders-unsubscribe@xxxxxxxxxxxxxxx
>
>
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>
------------------------ Yahoo! Groups Sponsor ---------------------~-->
Save Time & Money
Outsource Your Database Design & Development with Elance
Post Your Project for FREE
http://us.click.yahoo.com/VdNj9C/b.zDAA/cosFAA/zMEolB/TM
---------------------------------------------------------------------~->
To unsubscribe from this group, send an email to:
realtraders-unsubscribe@xxxxxxxxxxxxxxx
Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
Attachment:
Description: "pricemoves.gif"
|