[ARCHIVE] Any rookie question, so as not to clutter up the forum. Professionals, don't pass by. Nowhere without you - 3. - page 405

New comment
 

Probably already raised my question here, but to re-read 405 pages of the forum alas I do not have so much time. The question itself - why the results of the optimizer and the tester do not converge? After optimizing the input parameters in the tester and get very different results...

 

Good afternoon!

Please advise.

The indicator does not recalculate (does not redraw) on every tick...

And by the close of the bar, too.

The i+1 is not used.

The bars that are not recalculated are defined

  int сounted_bars=IndicatorCounted();
    if (сounted_bars==0) int limit=Bars-1;
    if (сounted_bars>0) limit=Bars-сounted_bars;

Cycle

for(int i=limit-1;i>=0;i--) 

What may be the reasons, what should I look for?

Thank you!

 
igid:

Probably already raised my question here, but to re-read 405 pages of the forum alas I do not have so much time. The question itself - why the results of the optimizer and the tester do not converge? After optimizing the input parameters in the tester and get very different results...

MetaAbsolute just pointed you in the right direction - test the system in real time and output more messages to see what is actually happening. It's one thing to think it should work, and another thing to see in real time dozens of errors, which the tester won't show you. It's another thing to see dozens of errors in real time, which the tester won't show you.
 
forexnew:
It's just that MetaAbsolute has pointed you in the right direction - test systems in real time and output more messages to see what's really going on. It's one thing to think it should work, but it's another thing to see dozens of errors in real time, which the tester won't show you. It's another thing to see dozens of errors in real time, which the tester won't show you.
So the problem is in the EA I'm testing?
 
igid:
it turns out that the problem is in the EA I am testing?

No, of course not. Not a problem, and the best way to test any idea is with a demo account. This is my strictly personal opinion. Look at how many messages in my EA, and what do you see when you look at your screen?


 
forexnew:

No, of course not. Not a problem, and the best way to test any idea is with a demo account. This is my strictly personal opinion. Look at how many posts there are in my EA, and what do you see when you look at your screen?


You got me stumped. I need to find the optimal parameters. Otherwise I would be cheating myself. Which begs the question of why the optimizer and the tester are out of sync...
 
igid:
You've got me stumped. I need to find the optimum parameters. Otherwise I would be cheating myself. That's why I'm asking why the optimizer and the tester are out of sync.
Open 10-100 demo accounts and run them with different parameters - a long time, but no fitting.
 
forexnew:
Open 10-100 demo accounts and run with different parameters - long, but no tinkering.
This is the first I've heard of it. Can a terminal run multiple accounts at the same time?
 
igid:
This is the first I've heard of it. Can a terminal run multiple accounts at the same time?


Copies of terminals... :-)

Each one has its own unique trading account.

 

I am testing my TS on Limits and pullbacks of the price to its CC-value.

I am testing TS on Limits on price reversal to its MA value, advise flattish (conventional) instruments, beside USDCHF, EURGBP, EURCHF, CHFJPY...

Thank you.

1...398399400401402403404405406407408409410411412...652
New comment