|
PureBytes Links
Trading Reference Links
|
-----Message d'origine-----
De : Patrick White <spy@xxxxxxxxxxxxxx>
À : Mark Brown <markbrown@xxxxxxxxxxxxx>; omega-list@xxxxxxxxxx <omega-list@xxxxxxxxxx>
Date : mardi 22 juin 1999 20:35
Objet : Re: Multi Data Tick Bar Charts are here!
>I think that Mark made it this way to avoid a nasty problem present in
>TS. If you do a regular multi-data chart with different minutes, the
>2nd, 3rd etc. data symbols have to be able to divide into the
>timeframe of the first subgraph in order to graph. This is no problem
>if you are charting a 5 min. and 30 min graph on the same graph, but
>TS will actually SKIP bars on data 2 for a 5 minute and 8 min graph
>when the 2 don't divide into each other!
>
If you push the reflexion further, you will see that multi data tick charts are a mathematical impossibility.
Because besides the time problem ( synchronism that is truely impossible with tick bars and non common divisible numbers, you have two solution that have hir own drawbacks:
Either create holes whant is stupid if you want to run trading systems on it.
Or to use variable tick bars compression to fill the space.
This is also stupid, because you rely on bars that are not constant and the synchronism compression algorithm is arbitrary.
The tick bar core will have a different number of ticks in each of them.
What do you ,want with this and wat is the advantage over multiple minute bars that are exploitable to run a system ?
Do you expect that real situation wll be the same by looking at these chars ?
No:
They will look the same, but will be different because you do not see the internal variable compression of the bars, INDEPENDANT of the price action, but DEPENDANT of the others data tick speed.
Ina nutshell, you displace the impossibility of synchronization into a transfer of some information of data1, data2 into data3 and so on.
Then your data1 2 3 are no more representing what they intended to be, and with an hidden variable rule that you cannot control.
Of course, the charts presended with a large compression may let you hink that the problem was solved.
In fact it is deeply buried in the low resolution of the screen.
This solution coming from someone who have not yet understood the bouncing tick approximation is stunning.
Next time we will discuss the "quadrature du cercle" solution that these geniuses will certainly produce very soon.
>This also solves the problem of the longer timeframe graphs in a
>multi-data chart having so much space in between the next bar.
>
This solves nothing:
It produces a pseudo mathematical solution to apparently solve a problem that cannot have one.
That said, unless you do not write trading systems on it, its is a nice chart.
And for ending, I'm always amazed to see that most of the critical sense of the contributors is used to hammer on Omega head.
Besides , you are able to swallow a whale without any difficulty if the beast is stamped "I do not like TradeStation".
Bon appétit, messieurs....
Rgds,
-Pierre Orphelin
Représentant exclusif de Omega Research en France.
web: http://www.sirtrade.com
|