Test of an EA based on ticks - page 4
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
Please speak up if you have any real knowledge.
How much of what the tester shows will correspond to actual trading?
I have an Expert Advisor working on ticks (only ticks are analyzed, TF is not used). To what extent the results correspond to real trading
. . .
A week ago I also tested some EA for 2010-2016 and got amazing results on real ticks with max drawdown ~3%.
And started to study this "perfection".
I discovered this: at the beginning of the test it thinks for a long time (of course after loading tick data), something turns inside, and then starts the test with a colossal speed with no drawdown, skips some days, sometimes half a month.
And those days tested after the last update of the Expert Advisor are not like a test before the update.
Although the test is running on real ticks, most likely this EA first analyzes all ticks for the specified test period, identifies (virtually) those days where profitability is high with a smaller drawdown, saves those days in an array or in a file, then starts the test using only those profitable days that were saved.
These doubts of course were not, if this Expert Advisor had a corresponding signal, so you could check. And unfortunately many naive users believe it.
P.S. As they say: "Not caught, not a thief".
I have long used self-written indicators that simulate trading instead of a tester.
This approach allows the test to be done instantly.
I advise
A week ago I also tested a certain EA for 2010-2016 and got amazing results on real ticks with a maximum drawdown of ~3%.
And started to study this "perfection".
I discovered this: at the beginning of the test it thinks for a long time (of course after loading tick data), something turns inside, and then starts the test with a colossal speed with no drawdown, skips some days, sometimes half a month.
And those days tested after the last update of the Expert Advisor are not like a test before the update.
Although the test is running on real ticks, most likely this EA first analyzes all ticks for the specified test period, identifies (virtually) those days where profitability is high with a smaller drawdown, saves those days in an array or in a file, then starts the test using only those profitable days that were saved.
These doubts of course were not, if this Expert Advisor had a corresponding signal, so you could check. And unfortunately many naive users believe it.
P.S. As they say: "Not caught, not a thief".
I have long used self-written indicators that simulate trading instead of a tester.
This approach allows the test to be done instantly.
I advise
Can you be more specific? As much detail as possible, for me to understand.
The indicator simulates the trading strategy, calculates Equity, drawdown and so on.
You can set the strategy input parameters in the indicator in the same way
But you have to write it yourself
That's probably all
The indicator simulates the trading strategy, calculates Equity, drawdown and so on.
You can set the strategy input parameters in the indicator in the same way
But you have to write it yourself
That's probably all
Add these lines to the source
New test. The results are better, because in the first case the trawl did not work correctly. I have added what you have asked for.
A new test. The results are better, because in the first case the trawl didn't work correctly. What you asked for has been added.
The easiest way to check if trades in the same period in the tester and on the demo are the same. If they don't coincide, then check if the profit in the tester and on the demo, at least approximately. If you want to check whether they coincide with each other. It seems it is not difficult to deposit it on the demo for a week.
Why do we use H1 period if we trade on ticks?
What you asked for has been added.
I can only see the report. Asked for the last lines of the tester log, where the slippage details are.
ZZZ The bug in the report is revealed - withdrawal is not shown at the end but at the beginning.