Strategic foresight systems - page 6

New comment
 

Taki deposit is alive, not dead. Now I have 200 coons in profit (I opened minimum lots). Now the maximum task is to live with these trades till Friday. Besides:

 

There are clear problems with the identification of several patterns, especially

  • AUDJPY
  • GBPJPY

Somehow I didn't pay attention to the number immediately - the confidence interval is very small, it shouldn't be like that. Why does the model suddenly think it knows exactly where it is going to go. I'll have to check again at the weekend. (Here for example is the correct interval: https://www.mql5.com/ru/forum/126769/page1681 in the forecast)

 

Taki found an error in model identification (trivial error in the formula), and since there are as many models as quotes, it will take a few days to repair. Should now be ok and probability counts. Theoretically.

PS: on page 1 the table is not corrected. There are some glitches.

 
Farnsworth:
...PS: the table on page 1 is not corrected. There are some glitches.

Do it in a new post, the edit period has already expired there.
 

I'm posting a picture from a 4 hour TF, but it could be just as good for a timeframe like a week.

It's not a cluster indicator, but a simple indicator based on the MAH principle. It simply tracks the market mood. The higher the chart of the instrument, the more popular it is. We have the same goal to determine the market reversal, but Farnsworth also looks into the future. I am watching it with interest but I cannot recommend anything yet. I am figuring it out. It is a little difficult to understand)).

I will not describe the programmer in this article so as not to make the programmers laugh, since I am not one of them.)

Such a picture today.

 

togranit77

Делайте в новом сообщении, там уже срок редактирования истек.

I see, I thought I would add in one post, reduce the number of posts and make it easier to read. :о)

to ULAD

i can't recommend anything yet

In the near future, I will try to be more specific in my question.

The higher the chart of the tool, the more popular it is. We have the same goal to determine the market reversal.

If I understand correctly, if something appears to be on the top/bottom, you can theoretically make a trading decision.

to Graff

...are arranged in descending order, i.e. the most profitable ones are at the top.

is this a local evaluation or is it the result of analysis of all statistics?

 

Important changes. I switched to Open[]. It seems to me that the best mode for Daylight Saving Time trading is the opening price of a new bar at night and there is almost a whole day for analysis and making trading decisions. Besides, we can coordinate the decision on the basis of the received quotes.

I have made corrections in the first block. I think the percentage value of the trend change probability has started to be considered as normal, while there were zeros, it should not be like that. But it still needs to be checked. I also included description of the current trend for better visibility:

  • The values "Trend start date" and "trend duration" are "by eye", they may be inaccurate. I will include the component with their calculation later
  • Similarly with the deviation, I will include it later

23.02: Probability of change of the current trend

Quote Type of current trend Date when current trend starts Current trend duration till the forecast moment (days) Decline in pips until the forecast moment Forecast date 24.02
(%)		 Forecast date 25.02
(%) 		Forecast Date 28.02
(%)
AUDJPY rising01.2719
35.2
AUDUSDgoing down02.174
38.3
CHFJPYgoing up01.1910
11.6
EURCHFdecreasing02.119
7.6
EURGBPdownward01.2720
29.9
EURJPYgoing up02.0414
25.1
EURUSDascending02.157
47.5
GBPCHFdecreasing02.148
32.2
GBPJPYrising01.2817
19.6
GBPUSDascending02.148
35.9
NZDUSDdecreasing02.0216
18.8
USDCADdownward02.0910
29.8
USDCHFdownward02.148
24.4
USDJPYgoing up02.0316
18.7

I hope I have time and will post the trend stats over the weekend. They are very important to understand.

A little analysis.

  • On AUDJPY yesterday's twists and turns fit in with the general behaviour, fits in the 85% confidence interval, but probably will change trend soon.
  • With EURUSD the situation is complicated, there is a high probability that there was a small correction and it will not develop, i.e. the positive trend will continue. Such places are very difficult to analyze and ambiguous. The statistics will show it.
 

Hi Sergei!

It seems to me that for your model it's better to analyze not 100 pairs (more/less) of different instruments, but to get only 10 indices (like USDx, etc.). Because indexes aggregate all the current trends for all currency pairs, their analysis is the most informative and resource-saving.

 
Neutron:

Hi Sergei!

It seems to me that for your model it's better to analyze not 100 pairs (more/less) of different instruments, but to get only 10 indices (like USDx, etc.). Since indices aggregate all the current trends for all currency pairs, their analysis is the most informative and resource-saving, and I think it works better in your case.

Hi Seryoga!

This is the second phase. Now I want to see the behavior of the system on quotes. It is not clear yet how it works. Get involved, become a strategist :o).

 

No-o-o-o-o! - I'm a sniper.

(crooked really... but it's the feeling that counts.)

12345678910111213...50
New comment