Difference between result in Optimisation mode and in Normal mode
Hello,
I coded my EA (3 actually) and I ran a back test on the last few months with it with optimization enable. That way, I picked the best value for some of the parameters of my EAs.
Then, just to check, I ran a back test (on the same period, the same instrument, the same method "real tick based on real tick", everything is the same, I didn't touch anything between the two back tests) but with no optimization (I simply click to "disable").
The result I got then is entirely different from the result I saw on the optimization result tab. The profit isn't the same, even the number of trades isn't the same. For 2 of my EA, they should have taken around 100 trades if I look at the optimization result tab, but they took less than 10 in the back test without optimization, and the 3rd EA didn't take any trade at all when it should have taken around 30.
I insist I didn't change my code between those tests (I only applied the value given by the optimization result tab to my parameters) and disabled the optimization on the second back test.
Is that normal, or do I miss something? What the point of optimization if the result isn't the same from one test to another? Which one believe? It's particularly the number of trades that shock me, it seems they are not testing the same EA at all...
How do you guys proceed to run back test in optimization mode and in "normal" mode?
Thank you very much in advance for your help.
You have not provided any details, nor examples of differences, so we have no idea where to start.
All I can say, is that most of the time, the problem is in the EAs code, but since you did not provided any, we can offer no comments on that.
The difference is in the number of opened trade (very few on the normal back test, 3 to 10 times more on the optimization back test) and also in the final profit result (and in every other statistique).
What kind of thing in the code could produce that difference? I already made those EA run in real, and there was no bug on it, they worked as expected. But I want to optimize some parameters, that's why I keep doing back test.
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
You agree to website policy and terms of use
Hello,
I coded my EA (3 actually) and I ran a back test on the last few months with it with optimization enable. That way, I picked the best value for some of the parameters of my EAs.
Then, just to check, I ran a back test (on the same period, the same instrument, the same method "real tick based on real tick", everything is the same, I didn't touch anything between the two back tests) but with no optimization (I simply click to "disable").
The result I got then is entirely different from the result I saw on the optimization result tab. The profit isn't the same, even the number of trades isn't the same. For 2 of my EA, they should have taken around 100 trades if I look at the optimization result tab, but they took less than 10 in the back test without optimization, and the 3rd EA didn't take any trade at all when it should have taken around 30.
I insist I didn't change my code between those tests (I only applied the value given by the optimization result tab to my parameters) and disabled the optimization on the second back test.
Is that normal, or do I miss something? What the point of optimization if the result isn't the same from one test to another? Which one believe? It's particularly the number of trades that shock me, it seems they are not testing the same EA at all...
How do you guys proceed to run back test in optimization mode and in "normal" mode?
Thank you very much in advance for your help.