Why are there so few experts in the MQL5 database? - page 8

[TheXpert]

Here. True, there's only part of the leak there.

[Prival]

Rosh:

The links are interesting, hopefully in the future we will have powerful maths packages too. But to be fair I have to point out that:

• 99.9% of traders do not know how to use it in trading;
• The lack of these packages can in no way be considered crutches.
  

As Nassim Taleb said, having mathematical baggage does not guarantee success for a trader, but it allows him to go the extra mile. Make no mistake about it.

I am very glad that at least someone understands the value of such libraries. And the fact that 99.9% of traders do not know how to use it, that's too bad, many traders do not even realize that they are reinventing the wheel, that there is a whole science of DSP, that there are mathematical principles of adaptation and separation (signal detection), well, here's an example.

The only indicator represented in the terminal that has adaptation properties is Kaufman's AMA and its derivatives. And if I'm not mistaken, you were saying that a lot of people use this indicator. Do you think so? Even without knowledge of DSP, many come to the simple conclusion that it is more efficient by gut feeling and searching for ATS. The method of testing is trivially simple, take a plot of history, and build the ATC at the intersection of 2 mash, optimize it to the maximum, and then do the same with 2 AMA. The result is simple and straightforward... and has anyone ever compared 2 AMA with two Chebyshev or Butterworth filters ?

As for Taleb, I can say the following: yes - you can become a trader without mathematical baggage. I don't think he would have said the same thing about the creator of ATS, I highly doubt that having mathematical baggage and the ability to program interferes with that.

[Hide]

The great thing about all this is that for many mathematicians, having mathematical baggage prevents them from becoming traders. They remain mathematicians. The same can be said of programmers. The ability to program prevents them from becoming traders.

[Prival]

Renat:

None of these points are directly related to the MQl5 programming language (C++, Java, etc). You have specified libraries.

And I will never tire of telling you that you need to change your priorities. The terminal is created for traders, it is their demand that will give further life to the software product. At this stage, the terminal with its OOP is designed for programmers, not for ordinary traders. We need object-oriented programming for traders, not for anyone else. Do you honestly believe that 99.9% can easily calculate the lot... https://www. mql5.com/ru/articles/113 , i.e. all the users of your terminal are extra class programmers ?

If you want to fix it in English, you'll have to compile a detailed description of the function and the use of the programmer's interface. You can find it here http://www.pro-technologies.ru/product/MSPE and in java alsohttp://measure.feld.cvut.cz/usr/staff/smid/local/sigproc.html.

If you have searched around, you may find a library for C++ as well. But MQL5 has no libraries for lot calculation, no library of trawls, no library of ready-made ATC, etc. The language has only classes and structures... and a dumping ground in the form of Code Base.

[Rashid Umarov]

Prival:

If you dig around, you'll find some for C++ as well. But MQL5 doesn't, which makes us proud that the language has no lot calculation libraries, no trawl library, no off-the-shelf ATC library, etc. There are only classes and structures... and a dump in the form of Code Base

We are already working on this - there will be classes implementing ekperts, trawlers and so on. And it will be possible to create your own EA by writing a minimum of programming code.

[Prival]

Rosh:
We are already working on it - there will be classes implementing ekperts, trawlers and so on. And it will be possible to create your own EA by writing a minimum of programming code.

This is exactly what I need. thank you. The minimum of programming, maximum result (the result is at least a test of the trading idea without months of labour spent on code debugging).

If you are able, add the entry, exit and trade efficiency parameters to the report and optimisation. We are missing these parameters. I am ready to help with formulas.

https://www.mql5.com/ru/forum/126953/page10 it's about checking the result of a trading idea, from my point of view, this is the ATS result, the better these parameters - the better ATS.

[Rashid Umarov]

Prival:

If possible, add the entry, exit and transaction efficiency parameters to the report and optimisation. How the air is lacking in these parameters. I am ready to help with formulas.

https://www.mql5.com/ru/forum/126953/page10 it's about checking the result of a trading idea, from my point of view, this is the ATS result, the better these parameters, the better ATS.

This thread and the screenshot from Bulashev also seen, thank you for the ideas. We'll do it, probably in the form of an article first. In general, there are plans.

[Andrey Dik]

Rosh:
This branch and the screenshot from Bulashev have also been seen, thank you for the ideas. Let's do it, first, probably in the form of an article. Anyway, it's in the plans.

Thank you very much. It's a great thing. Although the ideas are not new, I've been using these techniques for quite some time when teaching nets.

[Yedelkin]

Yedelkin:
С точки зрения непредвзятого начинающего пользователя язык MQL5 - доступный и изящный. А провокации - они и в Африке провокации.

TheXpert:
As a professional systems programmer, MQL5 is an clumsy bastard, supported from all sides by crutches, that may be easy to write simple things, but it is very difficult to write something really fundamental.

Well, let's see whether "unbiased novice users" will be able to write "something really fundamental" in MQL5, in spite of "professional system programmers".

I hope by "fundamental" they mean not theoretical knowledge, but reliable EAs that regularly make profit and protect from huge losses.

[[Deleted]]

Renat:

Please list these very crutches.

And the men don't know...

Men know, but they pretend they should. I'm talking about the dullard, not the crutches.

struct MyStruct;

We have an error message:

'MyStruct' - forward declaration is not supported

And the men also say that it's done that way on purpose for safety. They hide behind safety in order not to do it.

After all, the most important thing is to say: "the task was completed successfully". Even if the design is wrong.

Or a deep beta where the tester is a deep alpha, declare a release. And then people start discussing topics like "Why are there so few experts in the MQL5 database? And why are they, these people, acting like this? Everything is wrong for them...

Not to mention the Championship 2010 announced only for Expert Advisors written in MQL5. The 2011 Championship for Expert Advisors written in MQL5 would be more logical, because at this rate the tester could reach the confident beta stage by the spring of 2011...