Stochastic resonance - page 36
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
I would like to ask if this development in 2007 is no longer subject to secrecy, or if it is still a promising topic.
What is this parameter "X" and how it is calculated, I would like to run it through the NS.
I suppose you mean this https://forum.mql4.com/ru/8855/page23. This topic is still promising, except that further research took me into a completely different area. So, I gave it up for a while. Including for the reason that this dependence has a significant value only between (n) and (n+1) segments of the zigzag. In "practice" it means that we can relatively accurately determine the end of the current wave, and even then, not always and not precisely.:o))
As for the parameter "X" - I will not tell you about the most interesting, but the simplest and most obvious one - easily. For clarity: wave A is a completed wave (everything is known about it), B is the wave that follows it (for the waves following it, the dependencies completely fade out). For parts of a time series, limited by a segment of the zigzag is calculated:
(1)
(2)
This is what R(A) and R(B) look like, respectively
And this is what it already looks like:
There are several clusters, you will figure it out. In addition, the zigzag itself is important. It is possible to build a system and calculate the end of the wave on it, though not so easy. You need to "remove" unnecessary things, and it depends on your specific goals. So, keep me posted on your progress :o)
PS: And now for the main point. Wave B is actually at the end of its formation and some information is already known about it. This means that the dependencies can be refined by calculating the current parameters of wave B. This is a bit like a "system with memory". A very good example is "Russian roulette" (a revolver with a randomly unwound cylinder with 1 round in it). Each pull of the trigger changes the probability of firing (the "system memory" changes, or rather the state :o). It's the same with the B wave.'Is this possible or is it a glitch ?...... '
If the question is rhetorical, then yes, almost anything is possible.
There are several clusters there, you'll figure it out. Also, the zigzag itself is important. It is possible to build a system and calculate the end of the wave on it, but not so easy. You need to "remove" unnecessary things, and it depends on your purpose. So, keep me informed about your progress :o)
PS: And now for the main point. Wave B is actually at the end of its formation and some information is already known about it. This means that the dependencies can be refined by calculating the current parameters of wave B. This is a bit like a "system with memory". A very good example is "Russian roulette" (a revolver with a randomly spinning cylinder with 1 round in it). Each pull of the trigger changes the probability of firing (the "system memory" changes, or rather the state :o). It's the same with wave B.
>>Thank you, I will try it.
I presume you mean this https://forum.mql4.com/ru/8855/page23.
...
And this is what it already looks like:
There are several clusters there, you'll figure it out. Besides, the zigzag itself is important. It is possible to build a system and calculate the end of ox on it, though not so easy. You need to "remove" unnecessary things, and it depends on your purpose. So, keep me posted on your progress :o)
Oops, I thought this point was still classified :).
Well then I can put a penny in.
All clusters disappear if you don't complicate things by introducing a module. That is, the dependence of R(B)-R(A) on R(A) stretches along a curve more or less like a straight line.
The nature of this effect is easy to understand. If we have expectation of zigzag segment size R0 and the last segment had size R(A), what will be the expectation of the difference of size of the next segment and that R(B)-R(A) if R(B) does not depend on R(A)?
...
Right, R(B)-R(A) = R0-R(A). This is a straight line and it is this straight line that we usually see on the graph. And the more random is the series by which the zigzag is plotted, the straighter will be this straight line.
In fact, we have a criterion of whether our zigzag uses some real laws.
Or, since any zigzag is an equivalent of some reverse trading system, we have a criterion of input randomness for this system.
I have a couple of pictures on this subject, only I don't have them handy at the moment, I may add them in the evening.
You will see exactly the same picture on the SB.
R(B) is really independent of R(A).
Oops, I thought that point was still classified :).
It's not classified at all. As to the content - extrema do not characterize the quotes. In general, they do not. All TA based on dancing with tambourines over extrema is nonsense, a complete illusion.
Promised pictures:
Here is the "classic" - HZZ. Here and elsewhere, blue dots are(R(A), R(B)-R(A)), red dots are(R(A),R0-R(A)). That is, the red dots correspond to a random (in the sense considered here) series
Very similar to random. If you concentrate the data a bit by taking median values for groups of points, the impression of randomness is even stronger
That is, "lines" are practically parallel, the shift is probably due to the fact that median values were taken instead of averages. But it may even have turned out more clearly.
But there are other Zigzags as well :)
P.S. Who wants to share similar pictures for my zigzags? The zigzags themselves need not be declassified, of course.
A very different point has been obscured. In terms of content - extrema do not characterise quotes in any way. Nothing at all. All TA based on dancing with tambourines over extrema is nonsense, a complete illusion.
Good, I have no intention of giving away any secrets :)
Yes! Let's just smile at each other, that's it :o)
P.S. No one wants to share similar pictures for their zigzags? The zigzags themselves need not be declassified, of course.