The Sultonov system indicator - page 95

 
rjurip1:

And if you optimise it, it'll be fantastic.)

You'll make a mess :)

 
Igor Zakharov:

It's going to be a cakewalk :)

I guess porridge without porridge and without an axe. One water ))

 
Igor Zakharov:

Based on...

I have changed the equation to:

Ts5 = a0C0+ a1C1 + a2C2 + a3C3 + a4C4

as only in this case they (coefficients) will have a hint of geometric meaning (something like degree of significance). Without reinventing the wheel, solved the system by the Gaussian method.

Then, by way of nonsense, I decided to play with coefficients... And strangely enough, there's something about them: for example their sum is always greater than 0 :)

If I use them considering bar direction and averaging, I get a trend indicator:)


I am attaching the file for those who are curious:

file - missing
 
Roman Shiredchenko:
file - missing

was...

 
Igor Zakharov:

was...

wasn't

;)

 
Renat Akhtyamov:

wasn't

;)

Files:
 
Indicator work on real and demo
Files:
H8_a1b.zip  324 kb
 
Igor Zakharov:

Based on...

I have changed the equation to:

Ts5 = a0C0+ a1C1 + a2C2 + a3C3 + a4C4

as only in this case they (coefficients) will have a hint of geometric meaning (something like degree of significance). Without reinventing the wheel, solved the system by the Gaussian method.

Then, by way of nonsense, I decided to play with coefficients... And strangely enough, there's something about them: for example their sum is always greater than 0 :)

If I use them considering bar direction and averaging, I get a trend indicator:)


I am attaching the file for those who are curious:

Hello Igor, What does C0 mean here? Are the left and right parts of the equation equal after the calculations?

 
Yousufkhodja Sultonov:

Hello Igor, What does C0 mean here? Are the left and right sides of the equation equal after calculation?

C0 is the price at bar 0. I used your equation to make it easier to see the difference. If you number not "programmatically" :) my system of equations looks like this:

C1 = a1*C2+a2*C3+...+a(N)*C(N+1)

Ts2 = a1*Ts3+a3*z4+...+a(N)*Ts(N+2)

....

Ts(N) = a1*C(N+1)+...+a(N)*C(2N+1)

That is, we suppose that N previous prices determine the current value. By analogy to your suggestion, we use N equations to find N unknown coefficients, where prices are shifted by 1 to the past in each successive equation.

I checked it by printout to the log - it works:

 
Yousufkhodja Sultonov:
The work of the indicator on the real and demo

Here we have 1 hour before the market opens and I decided to comment on your best result.

First of all, in 4 years of trading you have only made 75% profit.

And secondly, you have a continuous profitless section, which occupies 1/4 of the entire chart.

In other words, you can see that you haven't made profit during the whole year: