What are netizens disliked for? - page 6

New comment
 
And has anyone in the know described the principle of this very grid? How does it differ from the usual martingale, what are its advantages, and where should one put it? Aside from a lot of useless orders hovering on the chart, I do not see any advantages... I cannot see any advantages, because one can always place orders at a certain distance from the price in time, instead of dragging the entire grid back and forth - what difference will it make?
 
Maxim Dmitrievsky:
And has anyone who knows the principle of this very grid? How does it differ from a regular martingale, what are its advantages, and where should I put it? I don't see any advantages, apart from a lot of useless orders hanging on the chart. I cannot see any advantages, because one can always place orders at a certain distance from the price in time, instead of dragging the entire grid back and forth - what difference will it make?

What does Martin have to do with it? Martin is not a trading strategy, but a type of MM. Different variants of grids are strategies + MM. I have already invented about 20 strategies with grids, many variants. I'm finishing the class on working with grids now and then I will start experimenting. I will try Matlab on real tick data thanks to the abundance of my tester.

Why do I use MQL? Because Matlab connects C++ DLLs with no problems and MQL code is upwards compatible with C++, with a few exceptions (C++, on the other hand, is not fully backwards compatible with MQL). That is, if I follow some simple rules, I write a strategy class in MQL, then translate it into a C++ DLL and use it in Matlab tester. Why not make this class for Matlab tests from the very beginning? Because, alas, for all its monster-like and universal nature, Matlab is slow specifically in the area of OOP. In general the approach to OOP is miserable, we see that goggle-eyed mathematicians have invented a bicycle instead of copying the advantages (here MQ have done it correctly).

So, I'll load this class with strategy variants into Matlab, there are different optimization methods and roughly estimate what's better. By the way, concerning real trade - I already found out that ECN accounts are not suitable for HF grid variants since there is such bull as stoplevel. Here is a method from class that defines it.

    /*
        returns stoplevel in double
        string smb - symbol
        int stopL - stop level in point
        double &dsl - real stop level
    */
    bool GetStopLevel(string smb, int stopL, double &dsl) 
    {
        long stl = 0, dgt = 0;
        double pnt = 0;
        ResetLastError();
        if(SymbolInfoInteger(smb, SYMBOL_TRADE_STOPS_LEVEL, stl) && SymbolInfoDouble(smb, SYMBOL_POINT, pnt) && SymbolInfoInteger(smb, SYMBOL_DIGITS, dgt))
        {
            if(stopL < stl)
                stopL = (int)stl;
            dsl = NormalizeDouble(stopL * pnt, (int)dgt);
            return true;
        }
        return false;
    }
 
Alexey Volchanskiy:

What does Martin have to do with it? Martin is not a trading strategy, but a type of MM. Different variants of grids are strategies + MM. I have already invented about 20 strategies with grids, many variants. I'm finishing the class on working with grids now and then I will start experimenting. I will try Matlab on real tick data thanks to the abundance of my tester.

Why do I use MQL? Because Matlab connects C++ DLLs with no problems and MQL code is upwards compatible with C++, with a few exceptions (C++, on the other hand, is not fully backwards compatible with MQL). That is, if I follow some simple rules, I write a strategy class in MQL, then translate it into a C++ DLL and use it in Matlab tester. Why not make this class for Matlab tests from the very beginning? Because, alas, for all its monster-like and universal nature, Matlab is slow specifically in the area of OOP. In general the approach to OOP is miserable, we see that goggle-eyed mathematicians have invented a bicycle instead of copying the advantages (here MQ have done it correctly).

So, I'll load this class with strategy variants into Matlab, there are different optimization methods and roughly estimate what's better. By the way, concerning real trade - I already found out that ECN accounts are not suitable for HF grid variants since there is such bull as stoplevel. Here is the method from the class that defines it.

Well martin at that the grid, in my opinion, is used for averaging :) otherwise what is it for... 20 strategies... that's a bit much, I'll have to look it up...

STP accounts with zero stops, by the way...

 
Maxim Dmitrievsky:

Well martin at that the grid, in my opinion, is used for averaging :) otherwise what is it for... 20 strategies... that's a lot, I'll have to look it up...

STP accounts with zero stops, by the way...

Maxim, 20 grand - I'm just in the habit of dictating my thoughts into my phone, I suggest that )) I don't search the internet, I make it up myself, it's more fun to live that way )).

STP is a method of"Straight-Through Processing" and all the DCs use it because there is no protocol and no standard. It is a generic term.

HDMI 2.0 is a data transferprotocol that supports 4K TVs and monitors and it's a rigidly fixed protocol. Right down to the two types of HDMI cables. And if Samsung doesn't implement support for this standard, it's going to have a bad hiccup. And Samsung is no stranger to hiccups ))

And DC with STP support will not do anything as there is no document. This should be clearly understood.

ECN also has no standard, but it includes STP processing. Yes, there are zero stoplevels, at least there is some certainty.

 
Alexey Volchanskiy:

Maxim, 20 grand - I just have a habit of dictating my thoughts into my phone, I suggest )) I'm not looking for it on the Internet, I come up with it myself, it's more fun to live that way )).


As for the topic - I'll try to find time this weekend and describe some net strategies that have come to mind. I have no reason to think, that some strategies will be profitable, my experience as a programmer has taught me not to drool over them )) Everything must be tried and tested. After all, even in experiments worth tens of billions of dollars, you never know if the theory will be confirmed or not. I mean AK, of course. What to say about small speculators)).

By the way, I searched the English-speaking part of forum today. Friends, we have more fun, that's for sure ))

 
Alexey Volchanskiy:

...

By the way, today I searched the English part of the forum. Friends, we have more fun, that's for sure ))

- That's all, Sophochka! Tomorrow I will radically change my life!
- What are you going to do, Monia, lie down on the TV and stare at the sofa?!
 
Artyom Trishkin:
- That's it, Sofochka! Tomorrow I'm changing my life dramatically!
- What are you going to do, Mona, lie down on the TV and stare at the sofa?!
I looked at the level of questions and answers. Mentality is different, of course. I must start a thread, check the difference in answers and comments from Russian and English speakers. I've noticed that thinking depends on the language. I am not a polyglot, just a personal observation.
 
Alexey Volchanskiy:

By the way, about the real - I've already found out that non-ECN accounts are not suitable for HF grid variants, as there is such a bull as stoplevel. Here is the method from the class that defines it.

This will not be enough - e.g. at Alpari StopLevel in variables = 0, but closer than spread (in both directions) will not give
 
I think I know what netizens are disliked for, for not understanding their work.
 
Alexey Volchanskiy:
Yes, I looked at the level of questions and answers. The mentality, of course, is different. I'll have to start a thread to check the difference in answers and comments from Russian and English speakers.I've noticed that thinking depends on the language. I'm not a polyglot, just a personal observation.
Interesting idea, to measure the angle of thought between two linguistic cultures.
1234567
New comment