Ambitious ideas !!!
Are you able to trade? You may write them in C# or Java, but one day you will realize how much time you wasted and give up using mql. Both mql's implement something, but not what you really need.
Trading is a secondary issue really. If there is a working strategy, which will be tested on history and backed up by test results, then it is no problem to write it for trading.
The main problem is finding a working and most importantly stable trading system.
Multicurrency strategies really work, if you make a system only for one currency pair, then such a strategy usually does not work for a very long time.
Trading is a secondary issue really. If there is a working strategy, which will be tested on history and backed up by test results, then it is no problem to write it for trading.
The main problem is finding a working and stable trading system.
MATLAB
Multicurrency strategies really work, if you make a system for one currency pair only, such a strategy usually does not work for a very long time.
An example?
See Simple multi-currency tester idea with an example of implementation
I had such an idea for a long time, then somehow it all fell apart. no one is willing to do this kind of work. there is no support or anything.
1. У меня случилась нелюбовь к MT5, а вот MT4 как старая добрая и верная подруга. Что не попроси, на все готова. Но MT4 нужно обучать новым трюкам.
2. Разный подход к торговле в MT5, а годами разрабатываемые эксперты, проверенные временем работают исключительно на MT4, в MT5 не смогут, идеалогия у них такая.
...
I'll continue on the topicstarter's behalf:
3. I am an evil locker and all my strategies are based on this pointless and harmful feature of MT4. But I am firmly convinced that the market shown in MT4 is not the same as in MT5, it is fundamentally different markets, in one of which you can earn (MT4), and in another one you cannot (MT5).
4. I don't like or even hate OOP. I do not know it and I honestly do not understand why people choose OOP, while there is an incredibly ugly but very simple MQL4. It's so simple and miserable that it's incredibly easy to write "Hello forex" type programs in it, which automatically means that it's much easier to write multi-currency/multi-time-frame/multisystem EAs.
5. Despite years of experience I still don't understand...
//This post should be interrupted, simply because the topic is pointless. Better occupy yourself with the invention of the wheel (in the literal sense, a wooden one to begin with), believe me such a pastime will be much more useful for both the head and the hands.
...
I'll continue on the topicstarter's behalf:
3. I am an evil locker and all my strategies are based on this pointless and harmful feature of MT4. But I am firmly convinced that the market shown in MT4 is not the same as in MT5, it is fundamentally different markets, in one of which you can earn (MT4), and in another one you cannot (MT5).
4. I don't like or even hate OOP. I do not know it and I honestly do not understand why people choose OOP, while there is an incredibly ugly but very simple MQL4. It's so simple and miserable that it's incredibly easy to write "Hello forex" type programs in it, which automatically means that it's much easier to write multi-currency/multi-time-frame/multisystem EAs.
5. Despite years of experience I still don't understand...
//This post should be interrupted, simply because the topic is pointless. You'd better busy yourself with reinventing the wheel (literally, wooden to begin with), believe me such a pastime will be much more useful for both your head and your hands.
I have to agree.
3. When tearing down the average MT4 is more loyal to "crooks".
4. you can write anything you want on it. But it cannot be tested and debugged. One has to put "money" into the jaws of a demo-machine (open a cent account, for example) - then the next level will be opened :)
5. Read my cry for help - examples are more instructive than rules!
;)
C-4:
4. I don't like or even hate OOP. I don't know it and sincerely don't understand why people choose OOP
OOP is just a petty PR trick to write more code scattered all over the place and load the cpu more at the same time. :)
This drives up the price of software and hardware resources, with almost the same final performance. But of course they are not fools and do not write their programs on OOP. :)
Sorento:
But it can't be tested and debugged.
You can, but it's complicated. :)
OOP is just a petty Moscow publicity stunt to write more code scattered all over the place and load the processor more at the same time. :)
This drives up the price of software and hardware resources, with almost the same final performance. But of course they are not fools and do not write their programs on OOP. :)
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
For a couple of years now I've been periodically tormented by the idea of implementing a multi-currency strategy tester.
I've seen several implementations, read ideas, and even looked at MT5.
Many people may say at once why if there is an MT5, where the multicurrency is already implemented in the terminal, there are many reasons, I will try to explain.
1. I do not like MT5, but MT4 is like a good old faithful friend. I cannot ask anything, I will do anything. But MT4 needs to learn new tricks.
2. The different approach to trading in MT5, and the time-proven experts developed over years work only on MT4, they cannot do it on MT5, that is their ideology.
I think these 2 reasons are affecting many for the most intimate of reasons, i.e. for the trading account balance ))))))).
So the essence of multicurrency strategy tester is directly in MT4 and in MQL4:
There is a base of quotes from 1999. It is in MySQL. There are many currency pairs. The same base is in MT4 but terminal cannot open all currency pairs and programmatically it is not possible to open or request the required bar, the result is memory overflow. To put it simply, you cannot open one-minute quotes for 10 currency pairs since 1999 without affecting the terminal.
Possible solution. Batch loading of necessary quotes from the database, for the required period and for the required currency pair into arrays directly in the Expert Advisor.
Further we proceed with virtual work, i.e. testing the strategy written in MQL4 on these data, when we ran out of data we uploaded some more, while the virtual deals, open and closed are also kept in the message, the memory will not plump up much with such an approach (though not checked), but theoretically everything will be OK.
You may use any currency pairs, periods as well, multicurrency and multitime virtual testing system may well work. We may make a report, draw charts and calculate everything we need, but this is a technical issue. You can also test or rather analyse and combine different strategies. Store everything in the database. In fact, this opens up the prospect of searching for a strategy not by one computer at a time, but creating a distributed computing network.
Actually I am interested in any thoughts on this idea. Perhaps in this thread will be collected materials that will be used in the development, i.e. what you advise.