The Sultonov system indicator - page 67

 
Vladimir Baskakov:
Are you serious or are you kidding me?

He's serious.

 
Mikhail Dovbakh:
Thank you, Yusuf, for your informative answer.
I'll ask again - please don't send it again.
1. Is the summation done on all available history, by points or what?
2. Y as well as x can be any financial instrument, or is that wrong?

1. It is possible to sum over the whole history of an instrument, but. then we get MNC estimation of coefficients over the whole or, any period of history analysis. It still needs to be done in blocks of 5 bars and summing up for the entire history and gradually going deeper into the history. Now we do the same but summing up should be done for 5 bars and we can gradually cover the entire history.

2. Yes, all with one requirement - that all bars of this tool should have the same nature of trading activity. It appears that there are markets related to indices where 1-3 bars are quiet, and on the 4th-5th bars, the market goes astray. Such markets cannot be properly analyzed, you can only find out which bars were active and which ones did not look like the market.

For example, this series of history data:

Date Time Open
2018.04.06 04:43 1,4013
2018.04.06 04:44 1,4012
2018.04.06 04:45 1,4013
2018.04.06 04:48 1,4013
2018.04.06 04:51 1,4012
2018.04.06 04:52 1,4014
2018.04.06 04:54 1,4013
2018.04.06 04:55 1,4014
2018.04.06 04:56 1,4013
2018.04.06 04:58 1,4014
2018.04.06 04:59 1,4012
2018.04.06 05:00 1,4013
2018.04.06 05:01 1,4013
2018.04.06 05:02 1,4014
2018.04.06 05:03 1,4012
2018.04.06 05:04 1,4011
2018.04.06 05:05 1,401
2018.04.06 05:07 1,4011

Convert to this form:

Ц1 Ц2 Ц3 Ц4 Ц5
1,4013 1,4012 1,4013 1,4013 1,4012
1,4012 1,4013 1,4013 1,4012 1,4014
1,4013 1,4013 1,4012 1,4014 1,4013
1,4013 1,4012 1,4014 1,4013 1,4014
1,4012 1,4014 1,4013 1,4014 1,4013
1,4014 1,4013 1,4014 1,4013 1,4014
1,4013 1,4014 1,4013 1,4014 1,4012
1,4014 1,4013 1,4014 1,4012 1,4013
1,4013 1,4014 1,4012 1,4013 1,4013
1,4014 1,4012 1,4013 1,4013 1,4014
1,4012 1,4013 1,4013 1,4014 1,4012
1,4013 1,4013 1,4014 1,4012 1,4011
1,4013 1,4014 1,4012 1,4011 1,401

And we take 5 sets of data to make up each SLAU:

System 1:

Ц1 Ц2 Ц3 Ц4 Ц5
1,4013 1,4012 1,4013 1,4013 1,4012
1,4012 1,4013 1,4013 1,4012 1,4014
1,4013 1,4013 1,4012 1,4014 1,4013
1,4013 1,4012 1,4014 1,4013 1,4014
1,4012 1,4014 1,4013 1,4014 1,4013

2nd system:

1,4012 1,4013 1,4013 1,4012 1,4014
1,4013 1,4013 1,4012 1,4014 1,4013
1,4013 1,4012 1,4014 1,4013 1,4014
1,4012 1,4014 1,4013 1,4014 1,4013
1,4014 1,4013 1,4014 1,4013 1,4014

3rd system:

1,4013 1,4013 1,4012 1,4014 1,4013
1,4013 1,4012 1,4014 1,4013 1,4014
1,4012 1,4014 1,4013 1,4014 1,4013
1,4014 1,4013 1,4014 1,4013 1,4014
1,4013 1,4014 1,4013 1,4014 1,4012

4th:

1,4013 1,4012 1,4014 1,4013 1,4014
1,4012 1,4014 1,4013 1,4014 1,4013
1,4014 1,4013 1,4014 1,4013 1,4014
1,4013 1,4014 1,4013 1,4014 1,4012
1,4014 1,4013 1,4014 1,4012 1,4013

5-я

1,4012 1,4014 1,4013 1,4014 1,4013
1,4014 1,4013 1,4014 1,4013 1,4014
1,4013 1,4014 1,4013 1,4014 1,4012
1,4014 1,4013 1,4014 1,4012 1,4013
1,4013 1,4014 1,4012 1,4013 1,4013

only after solving all 5 systems. we get the 1st set of unknown coefficients for this instrument. Introduction of the 6th set:

1,4014 1,4013 1,4014 1,4013 1,4014
1,4013 1,4014 1,4013 1,4014 1,4012
1,4014 1,4013 1,4014 1,4012 1,4013
1,4013 1,4014 1,4012 1,4013 1,4013
1,4014 1,4012 1,4013 1,4013 1,4014

immediately leads to the next set of unknown coefficients and so on until the entire input dataset is entered, all the way through history.

 
Attention all participants - I have replaced the incorrect exel file with the correct one in the attachmenthttps://www.mql5.com/ru/forum/307935/page58#comment_11172808! Please use it.
Системный индикатор Султонова
Системный индикатор Султонова
  • 2019.03.31
  • www.mql5.com
Уважаемые форумчане, в качестве основы стратегии будущего индикатора рассмотрим и обсудим следующую гипотезу: Цена текущего бара зависит от 4-х зна...
Files:
02_04_19_q.zip  28 kb
 
Yousufkhodja Sultonov:

1. It is possible to sum over the whole history of the instrument, but then we obtain the MOC estimation of coefficients over the whole or any period of the history analysis. Still we need to use blocks of 5 bars, summing up the entire history and gradually going deeper into the history. Now we do the same but summing is performed for 5 bars and we can gradually cover the entire history.

2. Yes, all with one requirement - that all bars of this tool should have the same nature of trading activity. It appears that there are markets related to indices where 1-3 bars are quiet, and on the 4th-5th bars, the market goes astray. Such markets cannot be properly analysed, you can only find out which bars are active and which are not.

Thank you.
Obviously, you have some theory of such composition of "coordinated" behaviour of different assets or you just observe a "plausible" indicator?
Which of the resulting parameters or what combination of parameters should serve as a signal to us?
 
Mikhail Dovbakh:
Thank you.
Do you obviously have some theory of such a composition of "coordinated" behaviour of different assets or are you just observing a "plausible" indicator?
Which of the resulting parameters or what combination of them should serve as a signal to us?

For example:

this kind of construction of unknown coefficients suggests that the first two bars of the story are not marketable:

a4 a3 a2 a1 a0 C5 calculated
-2,3334 -4 0 0 6,0668 1,4001
 
Yousufkhodja Sultonov:

For example:

this construction of unknown coefficients suggests that the first two bars of the story are not marketable:

a4 a3 a2 a1 a0 C5 calculated
-2,3334 -4 0 0 6,0668 1,4001
This might also be an interesting effect, but I was asking about the "normal" interpretation.
Also, it's not obvious to me to replace the whole composition of different assets with one, but with a shift? Or am I misunderstanding the example?
 
Mikhail Dovbakh:
That might be an interesting effect too, but I was asking about the "normal" interpretation.
Also, it is not obvious to me to replace the whole composition of the different assets with one, but with a shift? Or am I misunderstanding the example?

Yes, everything is shifting. But, what is ".... replacing the whole composition of different assets with one...."? What do you mean by asset? I'm used to differentiating between instruments only.

 
Yousufkhodja Sultonov:

Yes, everything is shifting. But, what is ".... replacing the whole composition of different assets with one...."? What do you mean by asset? I am used to distinguishing between instruments only.

yes, it's the same thing) usually trade assets.
As you said earlier, x1, x2,... x4 and Y are different assets(instruments, symbols) but on the same time scale. Right?
That's why I'm surprised by the new interpretation.
Then, maybe instead of shifting you should use data from different timeframes, if you decided to get away from multicurrency...
 
Mikhail Dovbakh:
yes, it's the same thing) usually assets are traded.
As you said earlier, x1, x2,... x4 and Y are different assets(instruments, symbols) but on the same time scale. Right?
That's why I'm surprised at the new interpretation.
So, maybe instead of shifting you should use data of different timeframes, if you want to avoid multicurrency.

This is the principle that systems need to be created to follow the principles of MNCs, Gauss states, and nothing else, it's not a whim of ours, but a necessity.

 
Yousufkhodja Sultonov:

This is the principle by which systems need to be built to follow the principles of ISC, Gauss argues, and nothing else, it is not a whim of ours, but a necessity.

did I understand correctly that we are now only talking about data for one instrument?
all data for all x and y belong to the same instrument - right?