The backtest tab not showing when using the strategy back tester - page 3

 
Colin Kimble #: Thank you for the advice! I have fixed the issue with that script, I have another script that uses a daily time frame and a one minute time frame. I have found if I use a EA that uses one time frame it is possible to run said EA in the Open Prices only model in the strategy back tester. Is it possible to run a EA that uses 2 different time frames (such as a Daily and one minute time frame) in the Open Price Only model in the strategy back tester? When I attempted this it kept on telling me how it is the wrong time frame (1 minute is the wrong time frame) request in a Open Prices testing mode.

I personally never use "Open Prices" modelling, ever. I only use the "Every tick based on real ticks", and on rare occasions when I have no other option, then I may use "Every tick" (virtualised ticks).

So, I am unable to offer you any further advice, on the "Open Price" modelling method.

 
Fernando Carreiro #:

I personally never use "Open Prices" modelling, ever. I only use the "Every tick based on real ticks", and on rare occasions when I have no other option, then I may use "Every tick" (virtualised ticks).

So, I am unable to offer you any further advice, on the "Open Price" modelling method.

Is it possible to use multiple time frames within an EA when using the "Every tick based on real ticks" modelling method in the strategy back tester? Thank you for all your previous help.

 
Colin Kimble #: Is it possible to use multiple time frames within an EA when using the "Every tick based on real ticks" modelling method in the strategy back tester? Thank you for all your previous help.

Yes, you can use multiple time-frames (and multiple symbols), on "Every tick", "Every tick based on real ticks" and "1 Minute OHLC".

As for "Open prices", I would assume that multiple time-frames would also work as long as they are higher than the default time-frame, but as I stated, I have no experience with this modelling method.

 
Fernando Carreiro #:

Yes, you can use multiple time-frames (and multiple symbols), on "Every tick", "Every tick based on real ticks" and "1 Minute OHLC".

As for "Open prices", I would assume that multiple time-frames would also work as long as they are higher than the default time-frame, but as I stated, I have no experience with this modelling method.

I understand, thank you for the wisdom!
 
Fernando Carreiro #:

Yes, you can use multiple time-frames (and multiple symbols), on "Every tick", "Every tick based on real ticks" and "1 Minute OHLC".

As for "Open prices", I would assume that multiple time-frames would also work as long as they are higher than the default time-frame, but as I stated, I have no experience with this modelling method.

I ran a back test using the, "Every tick based on real ticks" modelling system using a daily time frame for the ETHUSD, I tested it over a year being 2024, it gave me 365 total trades. I wanted to ask when looking at the Strategy Tester visualization in the History tab, there is a distinction between the Comment section either showing, "Enter Long" or "Enter Short", or it shows a "End of test". The "End of test" trades are said to have been placed manually where as the others were placed by the EA. How can trades be placed manually if it is back testing a EA? What is the actual difference between the two in the context of how it affects using this data to understand the performance of an EA?

Files:
 
Colin Kimble #:

I ran a back test using the, "Every tick based on real ticks" modelling system using a daily time frame for the ETHUSD, I tested it over a year being 2024, it gave me 365 total trades. I wanted to ask when looking at the Strategy Tester visualization in the History tab, there is a distinction between the Comment section either showing, "Enter Long" or "Enter Short", or it shows a "End of test". The "End of test" trades are said to have been placed manually where as the others were placed by the EA. How can trades be placed manually if it is back testing a EA? What is the actual difference between the two in the context of how it affects using this data to understand the performance of an EA?

When the test reaches the end, there may still be positions open. The tester, then forcibly closes them to end the test, and they are considered a "manual" operation, not done by the EA itself.
 
Colin Kimble #: I ran a back test using the, "Every tick based on real ticks" modelling

When using "real ticks", make sure that there is tick data available for the whole test period, or else the strategy tester will just internally switch to the virtualised ticks when they are missing.

Have a look at the modelling quality bar—It needs to be green across the board to verify that only real ticks were used.

 
Fernando Carreiro #:

When using "real ticks", make sure that there is tick data available for the whole test period, or else the strategy tester will just internally switch to the virtualised ticks when they are missing.

Have a look at the modelling quality bar—It needs to be green across the board to verify that only real ticks were used.

Thank you for the wisdom!
 
Fernando Carreiro #:

When using "real ticks", make sure that there is tick data available for the whole test period, or else the strategy tester will just internally switch to the virtualised ticks when they are missing.

Have a look at the modelling quality bar—It needs to be green across the board to verify that only real ticks were used.

My script closes all trades at the end of the day specifically at 23:00, what would then still be causing these trades to still be open? Is there a blind spot that I may be missing? Thank you for all your previous help.

Files:
DailyTest.mq5  24 kb
 
Colin Kimble #: My script closes all trades at the end of the day specifically at 23:00, what would then still be causing these trades to still be open? Is there a blind spot that I may be missing? Thank you for all your previous help.

Not according to the screenshots. Some positions were still open at 23:59. So, as stated before, put in some prints or use the debugger to find out.

That is what programming is about. When something is not as you expect it to be, you debug it until you find the issue and fix it. I'm old school. In my day, there were no forums to ask questions. You had to grind it out. You only had the physical manual as reference, and you debugged the code until you got it right. It may take a longer, but you learn a whole lot more than when someone spoon-feeds you the answer. That is how you gain experience and skill.