A purely theoretical question for mathematicians. With the possibility of moving to the practical plane. - page 6

 
Yousufkhodja Sultonov:

Understand, finally, once and for all, only I have been able to extend the MOC to the domain of nonlinear dependencies, absorbing, among other things, the Gaussian MOC for the linear domain.

Hey, docent, don't be ridiculous, eh? You might burst with laughter.

 
Yousufkhodja Sultonov:

Understand, finally, once and for all, only I have been able to extend the MOC to the domain of nonlinear dependencies, absorbing, among others, Gauss's MOC for the linear domain.

If you lived in the 17th and 18th centuries, you'd either be glorified or burned.

but now everyone's just laughing, even the most retarded. Or rather, the retarded would laugh and the smart ones would be silent :)

 
Maxim Dmitrievsky:

If you lived in the 17th and 18th centuries, you'd either be glorified or burned.

but now everyone would laugh, even the most retarded. Or rather, the retarded would laugh and the smart ones would be silent :)

Come on, the retarded ones.)

 

Before YS drowns everything in raptures of arithmetic knowledge :-)

We need to get back to the topic.

At any time, for any zigzag it is possible to make a "covering"/"converging" ("snake", there is no established term)
zigzag having the following property:

- each new maximum is smaller than the previous one

- every new low is greater than the previous one

And the construction is unambiguous.

I use this kind of thing just visually, but
Someone on the forum has already conducted a stat.study of this subject, unfortunately the necessary message is difficult to find.

I think the author of that post will not be lazy to share the result or conclusions again.

 
Yousufkhodja Sultonov:

If someone claims that, there is no such magic formula, it does not mean to take them as the truth in the last instance .

Can you stop saying "good light" outside of your own thread? it's blatant blatant flooding, irrelevant to the TC's question.

You're a grown man and you're forcing me to press the "infringement" button.

 
Maxim Kuznetsov:

Before YS drowns everything in raptures of arithmetic knowledge :-)

We need to get back to the topic.

At any time, for any zigzag, it is possible to make a "covering"/"converging" ("snake", there is no established term)
zigzag having the following property:

- each new maximum is smaller than the previous one

- every new low is greater than the previous one

And the construction is unambiguous.

I use this kind of thing just visually, but
Someone on the forum has already conducted a stat.study of this subject, unfortunately the necessary message is difficult to find.

I think the author of that post is not too lazy to share the result or conclusions again.

Apparently that was my post. Topikstarter does not like this approach - he calls it gonor, although it is just a simple matestat.

By the way, for zigzag knee values, I haven't noticed any noticeable deviation from SB (for major currencies) either.

 
Aleksey Nikolayev:

Apparently that was my message. The topicstarter doesn't like this approach - he calls it gonor, even though it's just an ordinary matstat.

By the way, for zigzag knee values I haven't noticed any noticeable deviation from SB (for major currencies) either.

It's time to go now, so quick, brief and most likely with errors:

on the minima/maxima of this figure you can build a regression n*e^k+c (if memory serves - it is such) , in excel it is a "trend line". Because from the construction it is a decaying oscillation. And even take the average of the two obtained - this is sort of the current centre, the convergence line, or the price to which everything tends to.

Whether you can get anything worthwhile out of it, you have to think very hard and justify it.


 
Maxim Kuznetsov:

Now it's time to go, so it's quick, brief and most likely with mistakes:

You can build a regression n*e^k+c (if memory serves - it is such) on minima/maxima of this figure, in Excel it is a "trend line". Because from the construction it is a decaying oscillation. And even take the average of the two obtained - it's like the current centre, the convergence line, or the price to which everything tends to.

If it is possible to draw something worthwhile from it, you should think hard and justify it.


In case of SB the ratio of adjacent knees is evenly distributed on the interval [0,1]. Therefore each next knee will, on average, be half the size of the previous one, i.e. the regression will have the form: x(n)=x(n-1)(1/2+e(n)), where e(n) is the residual evenly distributed on [-1/2,1/2].

Since no profitable strategy is possible in the case of SB, one must look for price deviations from it. In this case it means to find an algorithm that allocates time moments for which the ratio of adjacent knees (in the future) is distributed by a law different from the uniform one.

 

Those who want to deal with forex with some formulas show that they do not know how to trade profitably with their hands.

This implies that they do not understand forex and the nature of price formation and movement.

It is an axiom.


 
Petros Shatakhtsyan:

Those who want to deal with forex with some formulas show that they do not know how to trade profitably with their hands.

This implies that they do not understand forex and the nature of price formation.

It is an axiom


Of course you can trade with your hands, but to sit in front of the screen all the time ... That's not very good. If more or less logic can be described in the Expert Advisor, then let the computer work by itself. It is iron, so let it work...