My Goal:
I want to make sure that result of MT4 and SQX is the same exactly.
How i test:
1. XAUUSD forex 2015-2019 & lmax broker MT4
2.Export data to MT4 with no swap & commission => you can download my property file
3.retest a XAUUSD strategies in 2015-01-01 ~ 2019-01-01 => H1 timeframe
4.Now i can compare the netprofit to check if the result is the same
5.found netprofit is different
6.because some trades is different (using excel to compare), the trade show in screenshot is one of them (please refer the attached file)
p.s. you can retest strategy during 2016-03-01 ~ 2016-03-20, there is a trade i found is different
finally, If backtest engine have issue, all robustness tests is not believable...
I have a suggestion,
prepare an unit-test for backtest engine,
if you had modified the related code, run the unit-test first before releasing a new version,
it's make sure the backtest engine is work fine
Status changed from New to Fixed
Attachment XAUUSD_Strategy 16170.sqx added
Attachment Strategy Tester_ SQTest.html added
Both retested SQX strategy and html report are attached, you can check that
I know there is slight different because of precise of value of indicator,
So the solution will release in build 121 ? Or the bug you can't duplicated it again?
i suspect that you fixed it, so change the status to fixed
We made some improvements in MQL and SQ code recently, so maybe it fixed that. But most of the time it's just about the correct data and settings.
um....you don't find the root of cause
it's very clear problem for me
The trade is no show, but the upper shadow of bar at 2006/03/11 01:00 must be triggered the buy stop order to open
it's no doubt...
If the build 121 is released and the bug would be existed...I still have to report it again then...And you have to check that again...
Have you checked the things as below?
1.check if the buy stop order is opened or not in MT4 engine
2.If it's existed and don't be cancelled, checking why that trade don't open at 2006/03/11 01:00?
i suspect that you already check the two i listed and still can't find the problem.
Finally, i just want to make sure the problem will no show at build 121...
Do you ever have compared the two versions (build 120 & the version you said that have some improvements), Is there any difference ?
Is the bug existed in build 120 in your side and fixed in improved version ?
Thanks for you take care about this issue
Finally, i just want to make sure the problem will no show at build 121...
Do you ever have compared the two versions (build 120 & the version you said that have some improvements), Is there any difference ?
Is the bug existed in build 120 in your side and fixed in improved version ?
Attachment SQ_TickDataExportEA.mq4 added
I tested the strategy again on H1 and got the same problem as you did. I digged deeper and took a look on individual ticks in MT and SQ and I found the ticks are somehow not correct.
I contacted the rest of our team and we found bug in our MT export function. Now it works better but for some reason the spread isn't applied for XAUUSD symbol.
We must investigate this problem further. But the important thing is that the strategy works correctly in MT4 and SQ. The problem lies in the data export to MT4.
You can export the exact tick data from your MetaTrader installation using our tick data export EA. Then when you import it into SQ and retest the strategy on this data, you should get matching results.
We are currently working on the export and will fix it asap.
If you need any further assistance, just let me know.