a trading strategy based on Elliott Wave Theory - page 129

 
Yurixx wrote 10.08.06 23:02
Rosh I see that you have started to build channels on a slightly different principle?
I think it works better that way.


I read lately various manuals (this book as well) and found a simpler way to search for channels with a new criterion of single choice of a channel on condition of potentiality, using variance estimates.
But it's all in my head, I haven't done anything programmatically yet, I need to deploy all calculations mechanics, and it requires some air before diving in (now I better imagine volume of work on the second preliminary stage).
So for now I'm just reading the thread, I can't say anything clever :)
 
<br / translate="no"> But it's all in my head, I haven't done anything programmatically yet, I need to deploy the whole mechanics of the calculations a bit


Rosh, let me give you some advice. At least this advice is based on my own experience. Use MathCAD or similar product to test your ideas, there are quite a lot of them. You will get results very quickly (much faster, and more clearly than MT programming). Saves time which is always lacking. :о)))
 
I put Matkad on that year, and even bought a book. But I haven't needed it yet. And the calculation mechanism will still have to be programmed, although I agree - it is necessary to master it. I have never used anything but Excel, and to build a three-dimensional diagram in it (for example, after the backtest) is a pain in the ass, and I have not been able to find an acceptable solution.
 
<br / translate="no"> Yes, and the calculation mechanism still has to be programmed


This is true, but my point is, if there are several variants and for each variant there are several more ideas :o)) to try, in my opinion, it is better to use Matcad to get the solution itself. But, of course, if you know exactly what you need, you'd better program in MT. :о)
 
2 Rosh
I've been reading all kinds of textbooks lately (including this book) and found a simpler way to find channels with a new criterion of unambiguous choice of a single channel on the condition of potentiality, using variance estimates.

You have intrigued me. To my shame I must admit that I still can't understand how potentiality of a price field can be used. Not for specific purposes, but in general, for anything. From my point of view it is obtained automatically by choosing potential energy as a scalar function. And since vector analysis makes no sense at all in the price-time coordinate system, it turns out that there are simply no other options: potential energy can ONLY be in the form of a scalar function.

So there's something I don't understand, if others manage to derive something else from this condition.
In this regard, could you please share the link to the book where you read this.
 
I've been reading books lately that have nothing to do with potentiality :)
It's just that we've confused ourselves. Although I still don't understand where thousands of lines of code come from.

The mechanism works easily on the history, but to make a calculation in the dynamics - you need to do research to develop the criteria (that is, the criteria seem to be there, but they are purely theoretical - was smooth on paper, but forgot about the ravines - as they say).

That's why first I need to teach the indicator to search for channels not in the depth of history as it is now, but to track the behavior of the criterion for a channel that has already been defined and at a channel reversal/break to display the criterion statistics.

If it is successful, then we may move to statistics of deals opened by the same criteria, i.e. what Vladislav has implemented - a sort of cold water test.
 
I put Matkad on that year, and even bought a book. But I haven't needed it yet. And the calculation mechanism will still have to be programmed, although I agree - it is necessary to master it. I have never used anything but Exel, and to build a three-dimensional diagram in it (for example, after the backtest) is a real pain in the neck, and I have not been able to get an acceptable solution. <br / translate="no">.


It turns out that I used Matlab, not Matcad. Now I've found a book on my bookshelf - http://www.ozon.ru/context/detail/id/1599739/

What's the difference between Matlab and Matkad - I don't know :)
 
It's all about knowing how to use the tool. I agree, it sounds trivial. :о)))
And how they differ - there are quite a few forums on the Internet, on which participants "fight", defending each of their favorite instrument.
 
What is the difference between Matlab and Matcad - I don't know :)


In Matcad, even someone who only knows how it looks on paper can figure it out, i.e. if you know how to write it down using mathematical formulas (integrals, differentials, matrices, etc.If you know how to do it with mathematical formulas (integrals, differentials, matrices, etc.) then there will be practically no problems, in Matlab you need to be able to work with a lower level of visualization, but it has some things that matcad can't even dream of, just a little different goals of the application (in particular when learning we used it to model electric circuits just connect the elements, and it all there calculated) poof therefore now I think matcad is more relevant.
 
2 Rosh
I've been reading all kinds of textbooks lately (including this book) and found a simpler way

I understood that phrase to mean that you found that way in the book you were reading.
And you meant yourself. Right ?
Reason: