Who wants a strategy? Lots and for free) - page 2

 
Figar0 >> :

In the generator are limited) And to the absurdity in the presence of intelligence can be brought anything, and any program is a black box, if not written by yourself, or if in the presence of the code is not deeply into it. But that is not the point, for example, I'm not inclined to overestimate its capabilities, and in "everyday life" I don't use indicators at all. And I do not expect profit from it. If you do not know what's going on in your trading robot, don't worry, you won't get any tips. You should see it yourself, play with it.)

The point is not that the software fits the strategy to the history, but it does it very quickly. The tester in MT4 also does the fitting, but with terrible lags.

 
Reshetov писал(а) >>

The point is not that the software fits the strategy to the story, but that it does it very quickly.

It's fast, that's the point. It remains to be seen how reliable it is.

 
IMHO visibility is important. And in the MT itself, it is not so easy to see what goes in and out.
 
Interesting toy... I like it... (but is there any practical use...? we'll see...)
 
Figar0 >> :

Quickly, there is such a thing. It remains to be seen how reliable it is.

In this software there is "Indicators diagram", click on which opens a window and you can see how the events developed, where the positions opened and closed, what were the values of the indices and oscillators, what signals these indices and oscillators gave, etc.

 
sllawa3 >> :
>> interesting toy... I like it... but is it practical use...? We'll see... )

Generated three primitive strategies. Made Expert Advisors. All three are wonderfully optimised in MT4 even on minutes. And one of them successfully passes forward testing.


So this software is not a toy for fun at all.

 
Reshetov писал(а) >>

And one of the EAs is successfully passing the forwards tests.

So this software is not a toy for fun at all.

That's funny.

The funny thing is, Yura, that in half a year, you will still be on this forum discussing TCs and won't even remember about this NOT a toy.

 
Neutron >> :

That's funny.

The funny thing is, Yura, that in six months, you'll still be on this forum discussing TCs and won't even remember about this non-toy.

Funny thing is that I've found a second strategy, which is not even going to obudit, as there is nothing to fluster about - it also successfully passes the forward-tests.


>>. Two strategies with positive results on forwards in one day is a record in my practice. In the past, this has happened once every two months at best.

 

The only question is how long it will last in real life

P.s. The threads are rubbing off on selling EAs )))))

 
BARS >> :

The question is, how long will it last in real life?

P.s. This is the thread rubbing about selling EAs )))))

I am currently trying to modify strategies from tester variations into a code suitable for demo accounts. And then we will see.


The essence of the matter is that it is very convenient and most importantly, very visual, how all this stuff will work. If we tried to do it in the MT4 Strategy Tester, we would be wasting a lot of time:


1. In order to change something you have to use MetaEditor, edit the code. Then you have to look what you will get in the Strategy Tester. You select the variant with the mouse and immediately see the equity and balance diagram.

2. The code in MQL4 is terribly slow. All strategies are searched and run in machine code at once with very high speed.

Theoretically, we could parse results from XML directly into MQL EAs source code. But it is in the future. At the moment we do not even understand everything. Besides, I have a lot of different exotics which are absent in MT4 and therefore I have some interpretation problems. Therefore, I select only those versions, where the interpretation of trading signals is more or less clear.


I immediately disabled the Russian support, because it's not the Russian language, but a mockery of the rights of the Russian-speaking population.

Reason: