Iterative Refinement

[Jander]

Quote from TIKITRADER:

This post here from Mr_Black

http://www.elitetrader.com/vb/attachment.php?s=&postid=2018487

Reminds me of this great post from PointOne

http://www.elitetrader.com/vb/attachment.php?s=&postid=1872322

http://www.elitetrader.com/vb/showt...72322&highlight=fractal+point+one#post1872322

I feel like I have a solid foundation of understanding the jokari window, however the application of it I find a bit lacking. I thought (perhaps incorrectly) that the cells flowed in a certain order (as in PointOne's animated gif). The Mr Black diagram (and his recent posts) have the cells drawn in a seemingly random order. I dont see the practical use of the the cell sequences if there is no certain order to them and you never know the next step of the sequence.

Perhaps someone could expound on the issue?

 

[romanus]

Quote from romanus:

The reference to lack of increasing volume after lateral (or formation) BO always meant to be used to determine whether or not the trend has changed. In other words the break out has to be in the direction opposite to the trend. If you still convinced that Guavaman's comment with that respect has any validity, try finding an example of the traverse which ended because of that particular Traverse level change 'signal'. I did and I failed. Every time one sees a change after lateral BO with failure to show increased volume it is because there's some other signal present that does the job or there's no traverse end point.

RETRACTION: I was wrong about that. Apology for the unintended confusion.
P.S. Guava rules!

 

[Jander]

Chart for today (hindsight)

 

[Jander]

Quote from TIKITRADER:

ES Daily Chart August 7 2008

Fun to see this unfolding, although it is much like watching grass grow...

With all the economic crap going on, if I had to guess I'd say that brown RTL is going to hold and we will have a nice down leg to work with through the fall

 

[Tums]

Quote from Jander:
Fun to see this unfolding, although it is much like watching grass grow...

Apply the JW to it and see... ;-)



p.s. Thanks Tiki !!! You are coming through laud and clear.

 

[Mr_Black]

Quote from ivob:

Waiting for more than 2 hours for a good opportunity.

You got excellent discipline mr Black.

regards,
Ivo

Actually I just came home after work and this opp pops out...

 

[ljyoung]

Quote from treeline:

Hello innersky,

As we know, both Peak Volume and Pace Acceleration are characterized by 3 consecutive bars with an acceleration of the Gaussian Slope.

I like your hypothesis; I was thinking something similar after Tums' comments about the "Range Volume relationship":


... Someone can work out the Range Volume relationship.
http://www.elitetrader.com/vb/showthread.php?s=&postid=2012530#post2012530


and


I am trying to get people to think from all angles...

What about the range? Should it be in the calculation?
http://www.elitetrader.com/vb/showthread.php?s=&postid=2012611#post2012611



Based on a recent hint by Spydertrader, I came up with another possibility.

After another accleration of Pace (perhaps, by now, those Pace lines across the Volume Pane hold a bit more meaning than before)
http://www.elitetrader.com/vb/showthread.php?s=&postid=2017732#post2017732

I currently hypothesize that to qualify as a Pace Acceleration, Volume must increase substantially over previous levels, moving up at least one Pace level, if not two, above other recent bars. I've noticed that some of Spydertrader's examples of Pace Acceleration show this jump up to the next Pace level (or further) when compared to all other bars within the same Traverse. I've also noticed that some of Spydertrader's examples of Peak Volume show a smaller difference relative to other bars in the same Traverse, and sometimes the Peak Volume bar is actually smaller than at least one other bar in the Traverse.

I haven't seen near enough examples to verify my hypothesis, but this is where I'm currently focused.

Great ideas treeline. I have been thinking aout pace movements in terms of a jump through an as yet unknown number of levels from the viewpoint of the description given by Jack that the pace levels can be likened to a catenary (this was discussed in the 'Do trendlines work thread?'). To get an acceleration one must move volume beyond the level predicted from the catenary
construction for a single change in pace levels. This is what a catenary looks like:

The point here is that as you move along the x-axis one unit at a time, the amount that the y-axis increases/unit x-axis increase, goes up. Don't know the math. To me it's just a picture. This is basically what you were saying but from a different angle.

The problem I have is a practical one and that is that the 'pace levels' that we use are a variable feast and unless one wants to go through and do something like a 20 day MA of the various levels or has a program which does an 'instantaneous' calculation of pace levels (which I believe Spyder has), you're up the creek. On the other hand the PAs that I've looked at are pretty dramatic but if you have examples where there isn't a jump to very high pace levels, it would be great to see them.

However if one must find a way to get the appropriate pace levels each day, then so be it. It sounds like you're already collecting data. I'll do the same. Good hunting.

lj

 

[Mr_Black]

Quote from Jander:

I feel like I have a solid foundation of understanding the jokari window, however the application of it I find a bit lacking. I thought (perhaps incorrectly) that the cells flowed in a certain order (as in PointOne's animated gif). The Mr Black diagram (and his recent posts) have the cells drawn in a seemingly random order. I dont see the practical use of the the cell sequences if there is no certain order to them and you never know the next step of the sequence.

Perhaps someone could expound on the issue?

see cell migration on tape level fractal in this sample they follow the order 3-4-1-2 .... they do so because the market is doing it...if some times it looks that the market jumps cells...1-2-0-0-1 or 2-3-0-1 .... this is because this jumps are not spot able on current fractal...but if you look at the same time on 3 levels 64 cells it looks from this perspective that the market moves are not so random

 

[Jander]

Very cool Mr Black. I had a feeling that the randomness appeared due to the market cycling faster through the cells (on a lower fractal). I will delve into this further, thanks for the explanation...

 

[Mr_Black]

so far