You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
The other day I tried to make a custom testing of my strategy and faced the same problems, which have been previously described and solutions proposed by fxsaber(here) and Francuz(here).
During this time the Tester's automation has become so flexible and reliable that the problem can be considered as solved.
You only need to add 3-4 simple standard functions already written in WinAPI to make automated tester available in the Market.
Forum on trading, automated trading systems and strategy testing
Libraries: MultiTester
fxsaber, 2021.04.21 14:26
I myself use only four functions:
Forum on trading, automated trading systems and testing trading strategies
Libraries: MultiTester
fxsaber, 2021.04.21 15:02
And four more for more advanced use.
I do not use anything else.
From the point of view of minimal modifications you may be right. If we take the current IT architecture of MT5, we should start with the introduction of a new system function (event) such as OnTesterInit() and then move to the implementation of a complete set of built-in mql-functions.
There is no record of the following statisticalresults inthe testresults, which severely limits the selection of strategies.
https://www.mql5.com/ru/docs/constants/environment_state/statistics
LR Correlation:
Please add.
Also, e.g. the following is missing.
LR Standard Error:
AHPR
GHPR
Z-Score
Margin Level
You could add more analysis to the backtests, namely:
profits and losses by the opening hours of the trades that these profits and losses brought about.
Now there areprofits and losses by hours of closing trades:
Thank you for your hard work.
I support you. All the time it seems that among 500 profitable variants there is a variant, where the number of losing trends is minimal.
And somehow simplify the introduction of broker's commission (there is - but crookedly works probably))). That would remove a lot of unnecessary results.
You could add more analysis to the backtests, namely:
profits and losses by the opening hours of the trades that these profits and losses brought about.
Right now there areprofits and losses by hours of closing trades:
This is logical, but will not be done.