The Sultonov system indicator - page 27
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 remember Axiom (Alexey Yudin) solving a system of equations on the Alpari forum a long time ago. He seemed to be good at finding equilibrium points. The same question was discussed by Sitnikova in her dissertation and by Maryasov
..................................................... an excerpt
As I said before, my system consists of two blocks.
The first one finds unstable states in the market, i.e. states of deviation from equilibrium.
The basic system of equations of the block model
C'(t)=x1(t)*C(t) + x2(t)*C(t)*V(t) + x3(t)*C(t)*I(t);
V'(t)=y1(t)*V(t)*C(t) + y2(t)*V(t) + y3(t)*V(t)*I(t);
I'(t)=z1(t)*I(t)*C(t) + z2(t)*I(t)*V(t) + z3(t)*I(t),
where C is theclosing price of the interval, V is the trading volume, I is the market interest,
and families of functions x, y, z are unweighted parameters that determine the degree of influence and interrelation of the main parameters on the market.
I will leave the method of its solution and analysis without description as it will be time consuming and of academic value only. I will note the main result. Of the five points of equilibrium of this system the most
valuable point is
C5=(-x2(t)*y3(t)*z3(t)-x3(t)*y2(t)*z2(t)+x1(t)*y3(t)*z2(t))/
(x2(t)*y3(t)*z1(t)+x3(t)*y1(t)*z2(t)),
V5=(-x3(t)*y1(t)*z3(t)+x3(t)*y2(t)*z1(t)-x1(t)*y3(t)*z1(t))/
(x2(t)*y3(t)*z1(t)+x3(t)*y1(t)*z2(t)),
I5=(-x1(t)*y1(t)*z1(t)+x2(t)*y1(t)*z3(t)-x2(t)*y2(t))/
(x2(t)*y3(t)*z1(t)+x3(t)*y1(t)*z2(t)).
During the flat it almost lies on the price curve and when the trend changes, it moves away from it, judging by the distance between the curve and the price curve. By the value of C5 I estimate the market instability factor. The system calculates C5 for one interval ahead, i.e. it makes a forecast. Of course, it can be calculated for two and three intervals, etc., but as the system is hyper- sensitive to changes in the initial conditions for each successive calculation step, only a one-step-ahead forecast will be of practical value.
The second block is used to accompany an already open position.
And here I absolutely disagree with Bars that digital filtering methods are useless in predicting market behavior. :grin:
He uses these methods precisely.
It cuts off market cycles in the interval of 10-40 days and constructs "overheat zones" as bands restricting fluctuations of the cycle curve. It is a well known idea of Vladimir Kravchuk. I have a correction for volatility. The decision to close a position is made based on the behavior of the cycle curve in these zones. This block allows calculating a potential "out-of-reach" stop at position opening.
You can read more details about digital filtering here http://fx.qrz.ru. I think this resource is very decent though I do not use their developments. I have my own filtering and spectral density estimation programs.
This is a brief description of my system. If you have any questions I will be happy to answer them. :grin:
Eugene, this is not the case when by some mental or non-mental ways one could simplify the computational formulas for determining all five unknown coefficients of SLAE by my method. All possible simplifications have reached their logical limit and minimum - all calculations within all loops are performed by one. linear, chain of formulas using one memory cell. This situation, as chess players put it, is called a tug-of-war state, when any attempt to simplify the situation inevitably leads to its complication. And an attempt to jump from my method to the Gaussian method leads to a threefold complication, and to Cramer's matrix method leads to a fourfold complication of calculations. Therefore, you should put up with the seeming complexity of the above method and try to master it. There are no other methods. I do not advise anyone to experiment on this.
I am not saying simplify. I said to write in human language.
And the attempt to jump from my method to the Gaussian method leads to a threefold complication of calculations, and to Cramer's matrix method leads to a fourfold complication of calculations. Therefore, you should put up with the seeming difficulties of the above method and try to master it. There are no other methods. I do not advise anyone to experiment on this.
There are no difficulties there! SLAU of the fifth order is solved by the usual matrix method in the same Excel. We take a 5x5 matrix of coefficients, find the inverse of it (using MOBR()), multiply by a 5x1 matrix of free terms (using MUMNAGE()) - and we get a vector of 5x1 results. Everything happens instantly. What "complications in calculations" ? I'm sure you can easily solve tenth-order SLAEs in Excel.
...This situation, as chess players put it, is called a tug-of-war state, where any attempt to simplify the situation inevitably leads to its complication...
Not to simplify, but to improve, and leads not to complication, but to deterioration.
crossing lines?
That's a bummer...
Now that I uploaded here the ToR for EA and the code of indicator on Ezel, I am waiting for a response from programmers to create codes of indicator and EA.
Let's get started, then.
Bummer! I'm very disappointed. Why did you agree to do your own programming then? I lost a couple of hours of my time because of it.
You didn't realize that learning to code in MQL5 was a matter of life and death (creative) for you in your current situation. You made your choice. Well, freedom of choice is sacred.
You have had the chance to see for yourself the invalidity of the path through SLAU. It's impossible to catch anything there. Absolutely!
Maxim was right.
Do you have any idea what you are looking for in terms of geometric interpretation?
For a system of 4 linear equations with 4 variables, you are looking for the intersection point of 4 three-dimensional spaces in 4-dimensional space.
With each new bar such a point will be in a completely new location in 4-dimensional space. The trajectory of such a point with each new bar will be completely chaotic.
It is easier to represent it in the system of 3 linear equations with 3 variables where we need to find the intersection point of 3 planes:
With each new bar(calculation) you remove one (the oldest) plane and add a new one. The intersection point will now be in a completely different place.
Feel free to trash your theory.
Good luck to you in your search for a naive novice programmer... :))
Bummer! I'm very disappointed. Why did you agree to do your own programming then? I lost a couple of hours of my time because of it.
You haven't understood that learning to code in MQL5 was a matter of life and death (creative) for you in your current situation. You made your choice. Well, freedom of choice is sacred.
You have had a chance to see for yourself the invalidity of the path through SLAU. It's impossible to catch anything there. Absolutely!
Maxim was right.
Do you have any idea what you are looking for in terms of geometric interpretation?
For a system of 4 linear equations with 4 variables, you are looking for the intersection point of 4 three-dimensional spaces in 4-dimensional space.
With each new bar, such a point will be in a completely new location in 4-dimensional space. The trajectory of such a point with each new bar will be completely chaotic.
It is easier to represent it in the system of 3 linear equations with 3 variables where we have to find the intersection point of 3 planes:
With each new bar(calculation) you remove one (the oldest) plane and add a new one. The intersection point will now be in a completely different place.
Feel free to trash your theory.
Good luck in your search for a naive novice programmer... :))
It's called a technical merger! Yusuf asked for help in the first place, and this is you forcing him to write the indicator yourself.
This thinker has no time to help. He's studying the Core-Worker. Don't distract me!
The trajectory of such a point with each new bar will be completely chaotic.
Why?