The Sultonov system indicator

 

Dear forum members, as a basis for future indicator strategy let's consider and discuss the following hypothesis: The price of the current bar depends on 4 price values of previous bars according to the following relationship

C5 = C0 + a1C1 + a2C2 + a3C3 + a4C4

You may ask why it depends on 4? The point is that, so far, I am able to solve this equation up to 4 variables, the calculated formulas of which I have given before:https://www.mql5.com/ru/forum/86249/page3

Let's analyze the behavior of 5 coefficients, maybe we will be able to get some hints to the regularity.

  • The market should be described as a super-complex mechanism, not amenable to direct adequate analysis by the power of the human mind. To clarify the situation a little, let us choose a logically justified hypothesis and put forward and check the following global hypothesis by the power of a computer: LAST BAR PRICE IS FORMED BECAUSE ALL HISTORICAL DATA ON THE PRICE OF INSTRUMENT AT EACH HISTORICAL BAR is taken into consideration. Naturally, no one is able to check this hypothesis in full and we are forced to limit ourselves to a finite amount of information in order to assess the situation. We will compensate our digression by introducing the concept C0 - accounting for the pressure of historical data on the price at the beginning of the analysis, assuming that the market has a memory. Let us consider the regularity of price formation at the moment of opening of the current C5 bar on the basis of considering C0, C1, C2, C3 and C4 as follows: C5 = C0 + a1C1 + a2C2 + a3C3 + a4C4.

The computer gives the following results as food for thought, after introducing the term "virtual price" Tsi virtual. = aiCi:

C1 actual.C2 actual.C3 actual.Ts4 actual.a4a3a2a1a0C0 virtual historyC1 virtualC2 virtualC3 virtualC4 virtual.Tcvirt.Ц5 actual.
1,13761,13771,13751,1361-5,479871,1303931,375359-1,863376,6306826,630681616-2,119781,5647461,285822-6,225671,13581,1358
1,13771,13751,13611,1358-2,719060,2307690,635452-0,856194,2127944,212793788-0,974080,7228260,262177-3,088311,13541,1354
1,13751,13611,13581,13540,5588941,4507212,385817-0,8774-2,85908-2,85907782-0,998052,7105271,6477290,6345691,13571,1357
1,13611,13581,13541,13570,544521-0,489731,681507-1,883561,3034821,30348176-2,139911,909856-0,556030,6184121,13581,1358
1,13581,13541,13571,13580,949091-0,720910,41-3,097273,9287283,928727516-3,517880,465514-0,818741,0779771,13561,1356
1,13541,13571,13581,13560,422659-0,764780,341063-1,173292,4701732,470172562-1,332150,387345-0,868640,4799721,13671,1367
1,13571,13581,13561,1367-0,47611-0,67344-0,37034-0,904543,8908663,890866234-1,02728-0,42063-0,76476-0,541191,1371,137
1,13581,13561,13671,137-0,270980,044748-0,29526-0,337412,1118612,111861113-0,38323-0,33530,050865-0,30811,13611,1361
1,13561,13671,1371,13610,130820,10459-0,48492-0,236721,6885841,688583774-0,26882-0,551210,1189190,1486241,13611,1361
1,13671,1371,13611,1361-1,337951,684211-0,12465-1,603882,7070732,707073191-1,82313-0,141731,913432-1,520051,13561,1356
1,1371,13611,13611,1356-0,778931,468912-0,03022-1,297931,8617031,861702747-1,47574-0,034341,668831-0,884551,13591,1359
1,13611,13611,13561,1359-0,806391,512641-0,01497-1,287761,8143491,814348847-1,46302-0,017011,717756-0,915981,13611,1361
1,13611,13561,13591,1361-0,918111,03410,772669-1,073621,3471321,347132039-1,219740,8774421,174634-1,043061,13641,1364
1,13561,13591,13611,13640,2131862,7064492,317132-6,41532,4732282,473227751-7,285222,632033,0747960,2422641,13711,1371
1,13591,13611,13641,13710,7067743,2718481,97262-6,643461,9200151,920015419-7,546312,2410943,7181280,8036721,13661,1366
1,13611,13641,13711,13660,422892-0,60749-1,77213-3,235757,0379077,037906837-3,67614-2,01385-0,690780,4806591,13781,1378
1,13641,13711,13661,13780,951666-0,688570,496879-0,418490,7467930,746792813-0,475570,565001-0,782631,0828061,13641,1364
1,13711,13661,13781,1364-0,330530,632365-0,401450,6507720,506110,5061098780,739993-0,456290,719505-0,375621,13371,1337
1,13661,13781,13641,13370,2634460,36517-0,299530,5644730,1184830,118482630,64158-0,340810,4149790,2986691,13291,1329
1,13781,13641,13371,13290,178440,717824-0,417060,595678-0,08697-0,086971020,677762-0,473940,8137970,2021551,13281,1328
1,13641,13371,13291,13280,0645530,922948-1,101970,8036730,349780,349779910,913295-1,249311,0456080,0731261,13251,1325



It turns out that the market operates on the basis of a strict regularity, which no man is able to comprehend at the given stage of brain development and is perceived by him as a regularity, then as an absolute randomness. In short, the mind cannot understand the market, which is proved by researchers' efforts and traders' persistence.

Let me explain the first line: At the beginning of the experiment, i.e. at the moment of opening of the 1st bar, the pressure of virtual historical prices was +6.63 conventional units of the price that the market has to compensate for in the future. The effort of the 1st bar manages to soften the historical shock by -2.12 units, but the 2nd and 3rd bars hit by 1.56 and 2.12 units aggravating the situation. It remains for the 4th bar to strike a decisive counter-strike of -6.23 virtual price units at once. to stabilise the market by the time the current 5th bar opens! It seemed to everyone that, the market "accidentally" went down! All lines of the table can be analyzed in the same way. Conclusion: The terminal shows us only the tip of a huge iceberg called the market, without hinting at the fact that this market is mostly virtual, and it shows its real part in the terminal, which is used by uninitiated people. Moreover, this process of market self-regulation goes on forevery tick, minute, ....., month and years. To be fair, I have to admit that with this method of analysis I have not achieved any result that helps to profitably trade or predict the market. It just shows how complex the market mechanism is, that's all! The attempt to detect a pattern of price formation leads us to search for even more complex patterns, for example, a pattern of formation of C0 and ai, like a movement in a swamp. Although, logically, we can create and test the indicator, working by the following principle: if а4<1 and Ц04>0, then price tends to fall - SELL, otherwise - BAY. If anyone is ready to program and check this hypothesis, I'm ready to provide all calculations on Exel.

Let's try to develop here and now by common efforts of forum participants a sounded system indicator by principle:

If а4<1 and Ц0>0, then, the price is inclined to fall - SELL, otherwise - BAY:

We see that, during a flat, a4 does not exceed 1 and C0 remains positive - it means that the market will fall, which indeed happened soon. We will follow the situation as the data is processed. I will post right away.

Let's continue and see the efforts to stop the downward trend:


Continued:

For the first time the indicator advises to close all the sells as the a4>1 signal has appeared. We open BAY orders:

Continued:

Unexpectedly there is a signal to close the BAY orders and open the SELL orders, as again a4<1 has appeared, hence the upward movement proved false, which is confirmed by the further market movement:

Despite the sharp market movement upwards, the indicator remains unperturbed, considers this movement to be deceptive, firmly continues its SELL verdict and turns out to be right!

Назовите 4 фактора, от которых, на Ваш взгляд, зависит цена
Назовите 4 фактора, от которых, на Ваш взгляд, зависит цена
  • 2016.05.24
  • www.mql5.com
Уважаемые форумчане, назовите, пожалуйста, 4 фактора, от которых на Ваш взгляд, зависит уровень цены...
 
Yousufkhodja Sultonov:


Let's analyse the behaviour of the 5 odds, perhaps we can come up with some hints of a pattern.

There are already thousands of patterns in the market that are constantly changing each other. What's one more for?

 
Evgeniy Zhdan:

There are already thousands of patterns in the market that are constantly changing. Why one more?

If you read the first post carefully, you will see that I may have hit on the trail of a leading indicator.

 
What is he ahead of?
 
Vladimir Tkach:
What is it ahead of?

Market events. After looking at the charts and their comments do you have any doubts about this? Just as long as it is confirmed with other data. The work goes on.

 
Yousufkhodja Sultonov:

Market events. After looking at the charts and their comments do you have any doubts about this? Just as long as it is confirmed with other data. The work goes on.

Well you have just found the area where it "fits". It will work 50/50 like everything else.

 
Yousufkhodja Sultonov:

Let us consider and discuss the following hypothesis: The price of the current bar depends on 4 previous bar prices according to the following relationship

C5 = C0 + a1C1 + a2C2 + a3C3 + a4C4

Have you reinvented the AR model?

 
Evgeniy Zhdan:

Well you've just found the area where it 'fits'. It will work 50/50 just like everything else.

Please provide, from any timeframe, numerical price values to your taste and I will try to process them promptly. Believe me, I was not looking for a suitable plot on purpose, but I was happy with the U-shaped plot and did not know whether the indicator will cope with such a plot or not. How it coped brilliantly, see from the timeline of the charts.

 
secret:

Have you reinvented the AR model?

Point to the source of this model. There are many autoregressive models, which one do you mean? Why is there no such indicator or is there one?

 
Yousufkhodja Sultonov:

Point to the source of this model. There are many autoregressive models, which one do you mean? Why is there no such indicator or is there one?

There are a million sources as it was invented about 50 years ago. For example:Autoregressive model.

I was not looking for an indicator.

 
secret:

There are a million sources as it was invented about 50 years ago. For example:Autoregressive model

I wasn't looking for an indicator.

I agree, it turns out I used my version of the AR model and looked at it from a different perspective in terms of parameter analysis.