The backtest tab not showing when using the strategy back tester

 

Hi! when I use the strategy back tester to test a EA, the back test tab doesn't show. Does anyone know what causes this?

Files:
 
Your topic has been moved to the section: Expert Advisors and Automated Trading
Please consider which section is most appropriate — https://www.mql5.com/en/forum/172166/page6#comment_49114893
 
Colin Kimble: Hi! when I use the strategy back tester to test a EA, the back test tab doesn't show. Does anyone know what causes this?
It only appears for single back-test runs, not for optimisations.
 
Fernando Carreiro #:
It only appears for single back-test runs, not for optimisations.

Thank you for the information! How would one then test multiple pairs at the same time then?

 
Colin Kimble #: Thank you for the information! How would one then test multiple pairs at the same time then?

For a single run on each symbol, you can select "All symbols" in the type of optimisation.


However, this is only a single run per symbol, not full optimisation over all the symbols.

For full optimisations over multiple symbols, you will have to make your EA have a parameter for selecting a symbol from an indexed predefined table of symbols, or make make it fully multi-symbol EA, trading all the symbols simultaneously. However, the EA will not be able to rely on the use of the OnTick() event handler as it is no longer the default symbol being tested.

To do this you would need to code the EA to be fully multi-symbol with simultaneous trading on any symbol, irrespective of the new tick event, relaying on other event handling methods in combination with the OnTick() handler.

Another option is to have multiple MetaTrader installations and dedicate each one to a different symbol and run them concurrently.

And yet another method, is to use external command prompt script to automate the testing process and have it run for each symbol, but in this case you would need to make your EA also export it's own optimisations results report to make things easier.

In summary, it is not an easy procedure, so the best is to manually run optimisations for each symbol one at a time and then compile the reports for analysis.

 
Fernando Carreiro #:

For a single run on each symbol, you can select "All symbols" in the type of optimisation.


However, this is only a single run per symbol, not full optimisation over all the symbols.

For full optimisations over multiple symbols, you will have to make your EA have a parameter for selecting a symbol from an indexed predefined table of symbols, or make make it fully multi-symbol EA, trading all the symbols simultaneously. However, the EA will not be able to rely on the use of the OnTick() event handler as it is no longer the default symbol being tested.

To do this you would need to code the EA to be fully multi-symbol with simultaneous trading on any symbol, irrespective of the new tick event, relaying on other event handling methods in combination with the OnTick() handler.

Another option is to have multiple MetaTrader installations and dedicate each one to a different symbol and run them concurrently.

And yet another method, is to use external command prompt script to automate the testing process and have it run for each symbol, but in this case you would need to make your EA also export it's own optimisations results report to make things easier.

In summary, it is not an easy procedure, so the best is to manually run optimisations for each symbol one at a time and then compile the reports for analysis.

Thank you for the wisdom!

 
Hi! How come does the strategy back tester submit  39 286 amount of trades when using a daily time frame on a single FX pair? My understanding is that if it was set to a daily time frame back tested across one week, with their being 5 trading days it would only submit between 0 to 5 trades.
Files:
 
Colin Kimble #:
Hi! How come does the strategy back tester submit  39 286 amount of trades when using a daily time frame on a single FX pair? My understanding is that if it was set to a daily time frame back tested across one week, with their being 5 trading days it would only submit between 0 to 5 trades.
Don't assume that all EAs uses the "current timeframe".
Properly done EA are usually instructed to check specific timeframes, regardless of the chart one.
 
Fabio Cavalloni #: Don't assume that all EAs uses the "current timeframe". Properly done EA are usually instructed to check specific timeframes, regardless of the chart one.

My EA does check specific time frames. What else could have caused this issue? Thank you for your help.

[executable file removed]

 
Fabio Cavalloni #:
Don't assume that all EAs uses the "current timeframe".
Properly done EA are usually instructed to check specific timeframes, regardless of the chart one.
I only saw this now where by History quality it says 0% which obviously is atrocious, but I'd assume this would more so be related to gaps in market data and not necessarily a time frame, "mix up", would this also affect a time frame "mix up"? I am obviously going to change my historical data but I just wanted to double check if the issue lies within 1. My script 2. My MT5 settings or 3. The data provider.
Files:
 
Colin Kimble #: [executable file removed]
You have already been informed that executable files are not allowed. Only source code files should be attached.