forex software

Create and Test Forex Strategies

forex software

Skip to forum content

Forex Software

Create and Test Forex Strategies

You are not logged in. Please login or register.


Forex Software → Expert Advisor Studio → Not scalper Expert Advisor: 99,99% modeling quality still mandatory?

Pages 1

You must login or register to post a reply

RSS topic feed

Posts: 10

Topic: Not scalper Expert Advisor: 99,99% modeling quality still mandatory?

Hi everyone,
it's my first post here after buying my monthly subscription for EaStudio, first of all thank you to Miroslav and his team for this tool.

I really like to have your help about this simple issue.

I downloaded Forex Pro-demo 15m data on EurUsd, copied to EaStudio and run a test on 300.000 bar from Nov 2004 to Dic 2016.

After some additional manual changes (at the beginning it only gave me one indicator to open position) and further optimizations, EaStudio gave me very good results (please look image attached)

I backtested on MT4 for checking accuracy and it gave 90% modeling quality.

Now: given that it's not a scalper EA, could I avoid to test 99,99% tickdata? I don't have tick data on Mt4 and it's quite complex for me to have it now (I have a MAC and mostly windows software for downloading tickdata doesn't work neither on PlayOnMac).

In addition, there's a way to futher optimize this EA on EaStudio? Any suggestion is really appreciated :-)

Thank you very much,
Angelo

Post's attachments

Schermata 2016-12-22 alle 16.06.32.png 165.77 kb, file has never been downloaded. 

You don't have the permssions to download the attachments of this post.

Re: Not scalper Expert Advisor: 99,99% modeling quality still mandatory?

torinfx wrote:

Hi everyone,
it's my first post here after buying my monthly subscription for EaStudio, first of all thank you to Miroslav and his team for this tool.

I really like to have your help about this simple issue.

I downloaded Forex Pro-demo 15m data on EurUsd, copied to EaStudio and run a test on 300.000 bar from Nov 2004 to Dic 2016.

After some additional manual changes (at the beginning it only gave me one indicator to open position) and further optimizations, EaStudio gave me very good results (please look image attached)

I backtested on MT4 for checking accuracy and it gave 90% modeling quality.

Now: given that it's not a scalper EA, could I avoid to test 99,99% tickdata? I don't have tick data on Mt4 and it's quite complex for me to have it now (I have a MAC and mostly windows software for downloading tickdata doesn't work neither on PlayOnMac).

In addition, there's a way to futher optimize this EA on EaStudio? Any suggestion is really appreciated :-)

Thank you very much,
Angelo

MonteCarloTest image

Post's attachments

Schermata 2016-12-22 alle 16.06.02.png 102.98 kb, file has never been downloaded. 

You don't have the permssions to download the attachments of this post.

Re: Not scalper Expert Advisor: 99,99% modeling quality still mandatory?

MonteCarloTest image

Confidence Table

Post's attachments

Schermata 2016-12-22 alle 16.05.37.png 114.46 kb, file has never been downloaded. 

You don't have the permssions to download the attachments of this post.

Re: Not scalper Expert Advisor: 99,99% modeling quality still mandatory?

torinfx wrote:

MonteCarloTest image

Confidence Table

MT4 90% backtest

Post's attachments

Schermata 2016-12-23 alle 14.41.01.png 137.12 kb, file has never been downloaded. 

You don't have the permssions to download the attachments of this post.

Re: Not scalper Expert Advisor: 99,99% modeling quality still mandatory?

Well, you are wasting your time to be testing in MT4..... of course the results will differ because the two softwares are different.

FsbPro is far more precise.

I can understand that you want to compare with MT4 and I can assure you that as you gain experience you will see the differences and gradually wean youself off MT4.

Examine the features of the program carefully and you will start to see what I am saying.

My 'secret' goal is to push EA Studio until I can net 3000 pips per day....

Re: Not scalper Expert Advisor: 99,99% modeling quality still mandatory?

Thank you for replying Mr MacKay, I'm not using FSBpro, only EA Studio. In some backtests I made on MT4, I found big differences in results compared to those provided by EaStudio for the same strategies.

Some were very similar to EAstudio, some other not. Let's say I don't want to use MT4 anymore for backtest, how do you deal with that using only EaStudio? Given that Robustness test on EAStudio (with same automatic settings used) gave in both cases good results for the 2 strategies.

Data on MT4 were the same used for creating strategies on EAstudio, timeframe were differents, but always on EurUsd.
Backtesting period = period used by EAstudio for creating strategy.

What should you suggest to know that robustness test is really valid? Thank you!

Re: Not scalper Expert Advisor: 99,99% modeling quality still mandatory?

The difference may come from the different length of the data series. Some indicators calculate their current value sby using the hole data from the beginning. For example "On Balance Volume" indicator.

The EA Studio experts are very fast and it is useful to make a MT test as a final check before the trading.

Torinfx, please send me the experts that show different results. I'll inspect them and my find a groundwork.

As always - the most important is to:

Trade Safe!

8 (edited by torinfx 2016-12-30 20:56:56)

Re: Not scalper Expert Advisor: 99,99% modeling quality still mandatory?

Hi Miroslav,
sure, please find attached the two expert advisors.

I do not have the collection file, I uploaded the .mq4 one.

These 2 EAs above are available for downloading to the EAstudio traders, so anyone who download them is very welcome if he/she shares with us how test went on its platform, wich broker and spread and period used etc.

Any further improvement and insight is highly appreciated for me and I think to the community here too, thank you! :-)

The first strategy gave +7.400$ net profit on M5 EurUsd in 2 years (2014-03-24 09:10, to 2016-12-23 23:55) on 200.000 bars (FxPro-demo broker), I don't remember robustness exactly but it was good.

Backtesting it on MT4 gave a loss of 1379$ with 90% ticks modeling quality in the same period same broker same data same timeframe same spread (10)...

Post's attachments

EA Studio EURUSD M5 +7.400 in 2 anni.mq4 10.46 kb, 8 downloads since 2016-12-30 

You don't have the permssions to download the attachments of this post.

Re: Not scalper Expert Advisor: 99,99% modeling quality still mandatory?

here's the second one, on timeframe H4, EurUsd, period 2001-01-02 20:00 to 2016-12-23 20:00.

It has long stagnation periods and around max 25-30% drawdown, with similar positive results both in robostuness and in MT4 backtesting.

After your analysis, please could you also share what you think about these 2 strategies?

Thank you!

Post's attachments

EA Studio EURUSD H4 +300%.mq4 10.66 kb, 5 downloads since 2016-12-30 

You don't have the permssions to download the attachments of this post.

10 (edited by torinfx 2016-12-30 20:56:01)

Re: Not scalper Expert Advisor: 99,99% modeling quality still mandatory?

these 2 EAs above are available for downloading to the EAstudio traders, so anyone who download them is very welcome if he/she shares with us how test went on its platform, wich broker and spread and period used etc.

Any further improvement and insight is highly appreciated for me and I think to the community here too, thank you! :-)
Angelo

Posts: 10

Pages 1

You must login or register to post a reply

Forex Software → Expert Advisor Studio → Not scalper Expert Advisor: 99,99% modeling quality still mandatory?

Similar topics in this forum