![MQL5 - Language of trade strategies built-in the MetaTrader 5 client terminal](https://c.mql5.com/i/registerlandings/logo-2.png)
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
somewhere in the thread should be "not a demo" but just the code... scroll above
ok thanks!
here's the input from the report
Where is the logic ?
What logic for the input did you set yourself in the indicator settings at the moment?
What logic for entry have you set yourself in the indicator settings at the moment?
I have shown everything on the chart. The lines of a0 (red) and a4 (orange) coefficients are displayed
dotted horizontal lines are levels -2.5 and -7
according to what is written in the report
Alexey, there is no mistake, I just have to admit that, in the calculations, I use not 5, but 9 historical price values and that's it! And from this array we create SLAU - 4. In the calculations did not change, I only changed the principle of selection of input data in the structure of SLAU - 4. You too, please change this principle, looking file exel, which I send. I will immediately replace the old file everywhere in the attachment with the new one. Well done for noticing such a discrepancy. We have plenty of historical data. From the fact that, I admitted that there are not 5, but 9 historical prices, there is nothing criminal. Now, I can firmly assert that, future prices are not being used 100% . Now, I need to alert Makana to the need to adjust the ICL code in this way. Basically, why change the code if everything counts correctly. It's up to you, you want to change it, you don't want to change it.
What could have been optimised ?
there are no external options in the method.
Clean tests without options were presented last weekend. And it is possible to optimise, for example.
Pure tests without options were presented as early as last weekend. And it is possible to optimize, for example
somewhere above TC mentioned about "theoretical value of threshold 1.0", do you want to check it ? :-)
The more free variables, the better optimization results. This is the main property of the optimizer.
You can nail any function to the story in this way and it will not give anything but self-deception.
If the method itself works, then on zero (minimum spread) it should give a result different from SB right away. There may not be a profit, but there should not be a SB... And on solid runs(threshold overshoots) it should show effects as it gets closer to theoretical values.
Invite physicists and other laboratory assistants to the thread - they will explain in more detail, with references and literature, how to set up an experiment and test hypotheses.
As for optimisation. Of course, the best result in a certain segment is not the best a priori (as a rule). But among all results, there must be sustainable ones. As an example,
optimization on the interval
test +f
test +b+f
Really, how to find these stable parameters...?
As for optimisation. Of course, the best result in a certain segment is not the best a priori (as a rule). But among all results, there must be sustainable ones. As an example,
optimization on the interval
test +f
test +b+f
Really, how to find these stable parameters...?
is going hard
You'll need a lot of nerve in the real world.
Yusuf, all a4=0,945418086 are the same in your Excel file and by substituting data from page 1 of this thread all a4 are also the same as in your equations, you have also got DIFFERENT ones on page 1, which seems to be right, so you have mistakes in equations in file 04_04_19_q__1. How do you calculate or do you have a different table for yourself?
Please send me your variant of file 04_04_19_q__1 or place it in an attachment right here. After I post my variant of file here, I subject my variant to changes and even, already posted files are subjected to changes for the better. That's why I need to take a look at your file to be able to talk about it properly. And so, presumably, to clarify. that, on the first page, I showed the results of runs on the M (5x5) matrix, displaying only one line, describing the process of forming the opening price using all 5 coefficients. Then, in the next loop, the calculations were performed with a new sample, changing coefficients in the next row. Therefore, in the example on the first page, all the coefficients must be different. By the time the file you are referring to was created. Then I started showing all 5 lines of calculation results like this:
Take a look at the table and see. that, all the opening prices of the source data matrix are equally accurately calculated by the same set of coefficients. This very phenomenon allows us to say that these ratios characterize the market situation at the moment and relying on their values, the indicator produces a BUY signal, as in this case, because a4 < 1, then, SELL, when a4 becomes > 1. I think I was able to explain you something. If you don't understand, don't hesitate to ask, I will answer all your questions.