Forum Replies Created
-
Posts
-
Sounds like a good plan to start out Alan.
I’m sure you will adjust as you go of course.
I just meant with the correlation that you can open the EAs you have it EA studio (all for the same pair and timeframe) and make sure that they are not correlated using its tools but if you are not using it then that’s not an option.
Will be interesting to see how you go. Keep us posted on your results :-)
Hi Alan,
It would be logical to continue using that EA that you have faith in and had some good success with in demo for two months and it’s good your not going too fast. Might be an idea to test your new plan using demo also though, but scaling up with small money to start is also fine really (as long as can tolerate any drawdowns and losses with real money).
Remember that in using the EA from top 5 course that it was selected for you. Picking 10 from this selection means you are picking yourself so there is a difference.
I’d suggest also trying to pick less correlated EAs to combine also rather than just top 10 profit factor for example
Hi Alan,
Where did the original 50 EAs come from and why have you used the mt4 tester rather than just use EA studio?
Soumds sensible what you are doing to get round the FIFA rule.
How long have you been demo trading? It sounds like you are going straight to live? If so I would strongly recommend doing what you are doing with demo for at least a month or two first.
Of the three options you have given I would say profit factor but definately trade them for at least a bit on a demo account to make sure you are happy with their performance.
Hi Tia,
There are a few issues with your settings (I’ll expand on this later as on the train on way to work now). But your low generation number isn’t to do with your settings.
I believe that you have made your window inactive or shut your screen off on laptop maybe? EA studio needs to run on an active browser.
Can I ask a few questions to help you get the most generated EAs for your setup.
1- What browser are you using? (Chrome is by far the best)
2-Are you running on a desktop or laptop
3-What computer processor does your computer have?
We can tackle the low generation number first and then once sorted will discuss some of the other parts of your settings ;-)
Hi Mathew.
This seems like a big success really. Especially as you were just getting pure losses hen you entered this thread right?
Balance drawdowns are pretty inevitable. Have you been keeping an eye on your equity drawdown?
Also am I right in that you are moving EAs from demo to your mock live account monthly and still working with the same EAs on the original demo?
I would recommend keep doing what you are doing for another month or two but also maybe run a second system side by side using the EXACT same process.
It could be that this is just a natural drawdown and you may get another profitable overall three months OR perhaps the life cycle of your system starts to deteriorate after 2-3 months?
4+ grand profit on a 10k account over three months is a great result as far as I am concerned.
Hi Amby,
Time entry would work as Petko says. However I think it would be easier just to use the time settings in EA studio rather than the time entry indicator.
Just change the hours from 0:00 – 24:00 to avoid opening new positions during the high spread hours. This will all be included in the EA studio code if you change these settings.
Either way will do the job though ;-)
Hi Paul,
Using the validator OR fxblue is fine for managing the EAs but once they are on demo I would recommend using fxblue personally.
As to your question about how regularly to change the EAs. I can only say what personally make the most sense to me at the minute and that would be to look at BOTH its performance over this life of its time on demo and a shorter period such as week and make your decision based on that. IIan in one of his courses uses a month and a week which is also sensible.
It would also depend on the timeframe as I would be looking to change M15 Ea’s more regularly that H4 EA’s for example.
Hope that helps.
PS Of course I hope you mean a mock live account (i.e second demo) rather than a live account right?
It sound like you are saying that in your MT4 terminal you are not seeing data for the GBPJPY. If in the history centre you can just double click the symbol to download data. However I wouldn’t recommend this as it will download from metaquotes rather than your broker.
I would suggest opening a chart for the GBPJPY on the timeframe you want and then holding down the HOME key on your keyboard until as much data as your broker allows is downloaded.
Exporting data to csv can be done easily by just clicking file/save as in MT4 with the appropriate chart open.
For exporting from MT4 to FSBpro it is best to open FSBpro and go to the MT4 Hist Import in the History Centre (of FSB Pro) and click the link “Mt4 Hist Import Help” which will give you clear instructions on how to do this.
Hi Hikaru,
Yes backtest on MT4 should give extremely similar results to EA studio. Have a read through the “portfolio-calculations-vs-reality” thread and watch the “How to compare EA studio and MT4 backtester video” that is linked on the “lets Create a Brand-New Youtube Channel” thread. That’s should clear everything up quickly :-). I have a few additional videos made that aren’t uploaded yet and will get back to creating content in the future but have become overloaded with work recently so this is currently on hold. Luckily this video addresses your question exactly however 😉
I can see that you have used the tick model in MT4. As EA studio ea’s are coded to open positions on the bar open you should be using the open price model in MT4 backtester as this will best match EA studio and also most accurately represent how the EA would have traded.
I suspect a few other issues but if you follow through that video then I’m sure it will all fall into place.
Let me know how you go.
Hey David,
You are very welcome regarding my support. Happy to help.
What a turnaround! I totally understand the long hours having been through these frustrations myself. You will go far for sure!! This is not an easy game and like anything requires perseverance and experience to succeed. From going through what you have done with these tests you have gained depth of knowledge and understanding that really is best earned this way in my opinion (although guidance and discussion definitely helps hugely!). I don’t doubt you are very relieved that its done though haha
I can tell you are lightyears ahead compared to when you started this post and its all been through your own hard work! Well done once again.
Now you can focus on practicing and developing your systems with confidence in your tools etc :-)
I’m sure you will also agree at this point that EA Studio is truly fantastic software! We just need to become skilled in using it to develop profitable systems as it can only do what its told right ;-)
Happy trading! Very interested to see how your results go over the coming months, keep us posted
Oh also. In my opinion it kind of depends on how often your EAs are trading and how likely they are to have open positions at the same time.
For example if you have 22 EAs on the hourly timeframe and each makes trades 1x per week on average with a 24hr average position length then it’s gonna be way less likely to have a high amount of open positions at a time than if you were trading say 22EAs of M5 timeframe that typically makes say 10 or more trades a week.
Definately do a little work on your money management and get very clear with how to determine the right lot size on your account to manage your risk appropriately.
Hi Achmidius,
Glad you are getting more EAs in the collection. Always good to have more in collection and then filter from there.
I’m not 100% sure what you mean by principle of EAs but I’ll try and explain.
Each of your 22 EAs can have an open a position at the same time yes. So if you have 22 EAs running then you could have 22 open positions.
What lot size on what size account are you intending to trade and with what leverage.
For example if you are trading on a 10k account using a 0.1 lot size and 22 EAs, then you could be potentially opening 2.2 lots at a time which is 220k worth right.
On a 10k account with 100 leverage you would have 1 million buying powers so would be using about 1/5 of your margin.
I personally think this is reasonable and would not be looking to use more of my margin than this.
I would recommend checking out Petkos course on Money Management and making sure you are clear with the amounts you are risking when you are trading.
Starting out with a 10k demo using 22 EAs as you have done, 0.05-0.1 lots size per EA seems sensible to me on 10k account with 100 leverage, keep an eye on your margin levels in MT4 and get a feel for this and how many trades are open at the same time.
Your conclusions match my own David. I started out in a very similar way, trying to run as many browsers as I could and always wanting more computer power. These days I’d be pretty happy with just a single mid range computer with a 6-8 core processor that can run 6 reactors at the same time comfortably overnight :-)
In a nutshell:
1. Use Chrome
2. Run EA studio on seperate browsers not seperate tabs (Again this is shown in Petkos courses)
3. Keep the Tabs visible
Kalin can I ask how many EA studio instances you are wanting run at a time? Also what is your computer processor and ram are you running (Nothing extravagant needed for EA studio in my personal opinion, just checking you are running an optimal number of EA studio instances for your hardware).
Also remove all the “minimum net profits =10 “ in the acceptance criteria.
This is redundant since you have set the profit factor about 1 in compete, is and oos.
I would say it is good practice generally if struggling to get enough strategies to make acceptance criteria less strict and then use performance filter.
Petko has said it before, much better to have lots of strategies in the collection that you can filter than only a handful of strategies right?
Okay perfect, now we are getting somewhere.
So basically you are working with 4 years of data on EURUSD M15. First point is that I would get in the habit of checking the box “use end date” on the data horizon tab.
This is as I expected, the combination of your time settings and all the boxes closing the sessions checked with such a high minimum trade count is definitely pretty strict.
So a few points:
Obviously 4 years of data with M15 requires more bars than say 4 years of H1 so that you can definitely expect less calculated strategies in the former case.
But I think that with your time restricted setting and such a high minimum trade count over 4 years that this is resulting in a pretty fair small amount of strategies passing this acceptance criteria.
If this is what you are after then fair enough but I would suggest reducing the minimum trade count to 250 for example and then once more strategies are in the collection you can use the performance filter.
I’m sure you will see in the generator that plenty strategies are being calculated but very few are being ascended. Essentially for your generator settings the acceptance criteria is too strict.