No. Private messages are for personal communication only. And errors in code must be discussed publicly, so that others do not make the same mistakes, or someone could then share a link to a ready-made solution.
Please clarify. When optimizing Expert Advisor, I noticed that setting different TPs with switched on genetic algorithm produces different results, i.e. if I set TP=400, I optimize in steps from 1 to 400, then I set 350, and optimize in steps from 1 to 350 and so on till TP=50. So it turns out that if I run only at TR = 400, then I "skip" the results that are at lower TP and better! Is it because of genetics ? (The article https://www.mql5.com/ru/articles/1347 (Created: 25.12.2009 Author: Rider) says: "...Genetics is certainly a useful thing, but within reasonable limits. The thing is that its algorithm may play a cruel joke - some advantageous, from its point of view, set of parameters will be identified, and all further optimization will go "around it" until the very end. I think it's clear what the consequences are. Most of really good variants that will work outside the optimization area will be left "outside" and will not appear in the "Optimization results" table.
There are several ways out. One radical solution is to reject using genetic algorithm in optimization. But this is not always a good solution for one reason or another. The next two are half-measures, but at least something:
- Conduct the optimization not once, but two or more times. The first time, say, by "Balance", the next by "Maximal Drawdown" or something else ..."). So, in order not to get the best result, I should run it with different TPs + by different "optimizable parameters" and plus, for example, 9 forward tests with 9 optimizations for each test for reliability?
Break in your case triggers and interrupts the loop only when the condition is fulfilled. If the condition is not fulfilled, the loop is infinite. Try it this way.
int sec=Seconds();
while(true)
{if(sec!=Seconds()) // new sec
{ int ms=GetTickCount(); }
break;
}
To change the spread for example there is a program TakeMySpread, but how or with what you can change the "Stop Level" to test the EA in different conditions. Help!
Allow the use of cookies to log in to the MQL5.com website.
Please enable the necessary setting in your browser, otherwise you will not be able to log in.
You are signed in. Two-factor authentication has been disabled. Set it up again and scan a new QR code, or continue with two-factor authentication disabled.
But it doesn't work :(
So it doesn't work elsewhere or I've missed something
Can I knock on your door in person?
Hello. this simple code works with an error. I can't figure out the reason myself.
The first order opens, modifies. The second opens and... Doesn't modify. Keeps it until it fails.
In the log it says:
"01:23:22 MarkTrade started for testing
01:23:22 2010.01.29 00:00 MarkTrade EURUSD,H1: open #1 buy 1.00 EURUSD at 1.3965 ok
01:23:22 2010.01.29 00:00 MarkTrade EURUSD,H1: modify #1 buy 1.00 EURUSD at 1.3965 sl: 1.3935 tp: 0.0000 ok
01:23:22 2010.01.29 00:28 Tester: stop loss #1 at 1.3935 (1.3935 / 1.3937)
01:23:22 2010.02.01 00:00 MarkTrade EURUSD,H1: open #2 buy 1.00 EURUSD at 1.3877 ok
01:23:22 2010.05.05 13:17 MarkTrade: stopped because of Stop Out
01:23:22 2010.05.05 13:17 Tester: order #2 is closed
Got my mistake! OrderSend returns non-boolean value!
good afternoon
How to create a tester that works on different reframes.
and run it through history.
Good morning !
Please clarify. When optimizing Expert Advisor, I noticed that setting different TPs with switched on genetic algorithm produces different results, i.e. if I set TP=400, I optimize in steps from 1 to 400, then I set 350, and optimize in steps from 1 to 350 and so on till TP=50. So it turns out that if I run only at TR = 400, then I "skip" the results that are at lower TP and better ! Is it because of genetics ? (The article https://www.mql5.com/ru/articles/1347 (Created: 25.12.2009 Author: Rider) says: "...Genetics is certainly a useful thing, but within reasonable limits. The thing is that its algorithm may play a cruel joke - some advantageous, from its point of view, set of parameters will be identified, and all further optimization will go "around it" until the very end. I think it's clear what the consequences are. Most of really good variants that will work outside the optimization area will be left "outside" and will not appear in the "Optimization results" table.
There are several ways out. One radical solution is to reject using genetic algorithm in optimization. But this is not always a good solution for one reason or another. The next two are half-measures, but at least something:
- Conduct the optimization not once, but two or more times. The first time, say, by "Balance", the next by "Maximal Drawdown" or something else ..."). So, in order not to get the best result, I should run it with different TPs + by different "optimizable parameters" and plus, for example, 9 forward tests with 9 optimizations for each test for reliability?
the cycle is not interrupted for some reason
Break in your case triggers and interrupts the loop only when the condition is fulfilled. If the condition is not fulfilled, the loop is infinite. Try it this way.
break in your case only works if the condition is met
To change the spread for example there is a program TakeMySpread, but how or with what you can change the "Stop Level" to test the EA in different conditions. Help!