Backtesting/Optimization - page 25
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
Try this http://ratedata.gaincapital.com/. The only problem is that the download bandwidth is limited, so it takes a long time to download files. Then you need to make it suitable for MT. I think I've seen a post how to that on some forum, but I forgot where...
EA with 90% modeling quality?
can someone show where i can find an EA which has 90% modeling quality?
Thanks
can someone show where i can find an EA which has 90% modeling quality? Thanks
Hi jong52yuara.
In fact, no EA have 90% modeling quality.
The % of modeling quality is related to your backtest data. (all ticks).
So you better try to find "How i can't get 90% modeling quality" from pairs.
Regards,
can someone show where i can find an EA which has 90% modeling quality? Thanks
NO HAD INDICATORS WORK AT MODELING 90%.
because you come in risky,when come in forex ...
EA failed with "Every Tick" / minimum StopLoss value
Hello,
1. In MT4 tester, why most EA failed using model "Every Tick" even if they perform using the model "Control Points" ?
2. What is the minimum Stoploss value we can use in function Ordersend() ?
3. I think this minimum Stoploss value could be smaller (without getting Error 130: Invalid stops), testing a EA offline.
Did you experiment the same ?
Thanks.
Is there a difference using PERIOD_15 instead of its integer value 15 ?
Hello,
A. I refer to this post http://forum.mql4.com/4965
and the irusho1 reply telling:
...
3. hardcode your timeframe in tester (i.e. use iTime, iHigh, etc. use explicit PERIOD_?? instead of 0 or Period() function)
and run your EA on 1 minute frame only)
...
B. Why is this point important to get a test closer to real life ?
C. Is there a difference using PERIOD_15 instead of its integer value 15 ?
Does this forum allow to contact another member ?
If so, how to do it ?
Thanks
Backtesting an EA (using a fixed timeframe value) provide different results !!!
Hello,
1. I put a fixed value PERIOD_M15 for the timeframe parameter for all indicators.
So we must get the same result for any timeframe selected in the tester setting.
Because, the Buy/Sell conditions are based only on those 3 indicators using a fixed timeframe value.
But I got different results! Why ??
diClose0=iClose(NULL,PERIOD_M15,0);
fMM=iMA(NULL,PERIOD_M15,5,0,3,2,0);
sMM=iMA(NULL,PERIOD_M15,7,0,3,2,0);
2. How to test multiple EA simultaneously on a real demo account (or on many demo account) ?
Is it possible to do it without installing many Metatrader Terminal on the PC ?
My goal is to test few EA on a real demo account.
Then I would easily compare their performance.
Thanks.
EA backtest or Forward testing?
Hello everyone,
Just need opinions from the programmers in this forum. I have heard enough about how backtesting can be misleading and not a true reflection of the system.
However, is Forward testing a true test of the potential of any EA??
comparison test for backtesters of MT4, Amibroker and Tradestation
Hi,
After all the questions about credibility of MT4 Tester I have made a simple comparison test for backtesters of MT4, Amibroker and Tradestation.
For the best comparison conditions I have backtested all the programs with:
- the same data ( originated form Alpari 1M data) and time period,
- the same level of spread(MT4) and corresponding commission ( Amibroker and Tradestation),
- the same simple sample EA( back-test formula):
Results:
As you can see ( in the attachment) all 3 applications are giving signals at the same time and levels ( there is some difference in net profit between MT4 back-Tester an two others because of swap points in MT4).
Conclusion for me is:MT4 Tester is OK
( giving the same results as the other programs).
Another question is the quality of testing data. Although the way described by Hendrick how to achieve a 90% modelling quality ( https://www.mql5.com/en/forum/general )seems to be the best, I am not convinced to Alpari 1M data quality. When we prepare 90% modelling quality from Alpari 1M data and compare back-tests results of EA’s of Hendrick, Zonker or Sashken ( participants of http://championship.mql4.com/2006/participants ) we will see 30-60% differences between those EA’s results and real results published on this site. Too much for Daily EA’s.
Maybe you know some trustful 1M test data ?
Regards
Backtesting
Is there any rule of thumb for the length of backtesting (6 months, 1 year, 2 years)? And is length of backtesting dependent on the number of trades per day, week, etc.? Thanks for your help, Les
leshammondpsf@yahoo.com