a trading strategy based on Elliott Wave Theory - page 95

 
:-D. Guys, I was answering this <br/ translate="no">
Somehow I'm sure the productions will be 'slightly' different from each other.


I have already got tired of typing(Yurich respect, he sent me half a page of formulas and hasn't missed anything).

About the energy from the news, well, I do not know something still has time and speed, and the change in speed, too, you can look, so in principle, so far no restrictions can be seen, I just have to try, and maybe it will work. As for your grasn idea, maybe I'm a slow learner, I'm still figuring it out.


I hope that the "mechanical" approaches will work. And integration can indeed be done in different ways. But what is more important, in my opinion, is what to integrate. And in this regard, I don't see much point in integrating the price series, if we want to get the potential energy functional, of course.

As for the definition of the functional, you are quite accurate in your formulation. According to it, min(f(x)) - is also a functional. I do not think that it is necessary to be so tightly attached to this term.

And as to my idea - it is only a hint, as you asked. Its correctness has been tested in laboratory conditions. So far so good.
 
And as for my idea - this is just a hint, as you asked. It has been tested under laboratory conditions. So far so good.

I don't need any more, thank you, but I am only interested in what you were guided by when testing what you invented. You managed to check something, but unfortunately I don't understand what. Perhaps you've already written an EA that shows profit during testing, or maybe you've written an indicator that shows reversal zones on the history or something else? I.e. how have you established that everything works, maybe you can post a picture?
 
According to it and min(f(x)) - is also a functional.

It is not. min(f(x)) - is a number, a scalar. A functional is a number function, i.e. a scalar function.
 
В соответствии с ней и min(f(x)) – это так же функционал.

This is not true. min(f(x)) - is a number, a scalar. And a function is a number, i.e. a scalar function.


This is written, as an example to the definition of a functional in my maths handbook. :о)))) I didn't come up with that.
 
If by f(x) you mean a whole class of functions, and not one particular function f(x), then this is correct.
Each function f(x) from the class will correspond to one number - min(f(x)).
 
In functional analysis, each function is associated with a number (scalar). The function of the function is the functional F(y). It can be path length, potential energy and so on (this is what I remember). The function's problem is to find, by variation (differentiation), such a function in analytic form that F(y) is minimal. I haven't solved this problem yet, just all sorts of thoughts spinning around.
 
В соответствии с ней и min(f(x)) – это так же функционал.

Это не так. min(f(x)) – это число, скаляр. А функционал - числовая, т.е. скалярная функция.


This is written, as an example to the definition of a functional in my maths handbook. :о)))) It wasn't my idea.


Obozhit, you start to argue, but as usual you don't agree on the meaning of the basic concepts.
I so min(f(x)) denote the minimum of FUNCTION, while functional is another concept, and what did you denote in this way?
It is also written in the reference book: A functional corresponds to each function with some number. That is, the minimum of a function is a number, and a function can minimise or minimise it.
 
And once again, burn, this is not what we're discussing here. What is a functional and how to minimize it you can read in the reference book, but the type of this functional is another matter. Firstly it is necessary to define in which "class of functions" :) we will search a minimum, a hint is given it's potential energy (it is certainly not a class, but only a hint), but on what it will depend I would like to say on the bar number, as it unambiguously separates one sample from another, but there may be other variants.
 
А по поводу моей идеи – это же только намек, как Вы и просили. Ее правильность проверил в лабораторных условиях. Пока все хорошо.

I don't need any more for now, thank you, but I'm just very curious what guided you in verifying what you invented. You managed to check something, but unfortunately I don't understand what. Perhaps you've already written an EA that shows profit when tested, or you've written an indicator that shows reversal zones on the history or something else? I.e. how have you established that everything works, maybe you can post a picture?


I wrote a little script. Due to the fact that I seem to have understood the real nature of the optimization problem, I cannot write such a fully working algorithm. For this reason, I tested it "in laboratory conditions" on a story with some restrictions, including restrictions regarding the algorithm.

What I found. Exactly what (well, at least I'm an optimist :o) you wrote about: another criterion of channel selection, based on some (within the assumptions made) potential energy. Plus the pivot zone. At least the results are correct. The most reliable channel is chosen. There is another point here. I did not connect all the system components yet; I was testing only my hypothesis separately.

I did not get to the Expert Advisor. By the way, there were some ideas, connected with signal energy, but in the DSP context.

And indeed Yurixx was absolutely right - there is no need to approximate the parabola and extend it at the existing price series.

Unfortunately, I'm off on a business trip again (I have them quite often and in a way interfere with participation in the forum :o). But I think, within 4-12 days, there will be something to show. In the meantime, I do not consider it necessary to post the current results.
 
Maybe you have found what you need, but so far I've seen some discrepancies with Vladislav's hints.
Your statement:
You got channels with confidence intervals. You have extended them into the future. When a lot of lines intersect beyond zero, what do you get? So this is the space where you have to minimize the quadratic function.

contradicts Vladislav's statement

That is, constructing projections comes down first to sampling, then to linear algebra.


He first selects samples and then extends them into the future (at least that's what I understood him to say). We first minimize and then build channels.