EA Factory | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Calendar as Fundamental Source at Forex Factory 14 replies

Forex Factory Store 0 replies

Journal Factory posters - how much money are you making? 17 replies

Newbie Factory Journal 12 replies

Attachments: EA Factory
Exit Attachments

EA Factory

  • #1
  • First Post: Sep 28, 2006 12:42pm Sep 28, 2006 12:42pm
I am opening this thread because of an article I read in PF about giving and receiving (authored by fijitrader if you still don't know about that article). I have been practicing the idea accordingly and I think it has been transforming my life ever since. Forex is just one aspect of it but let 's just focus on forex. Ever since I begin to make EAs and indicators and share them with the nice fellows on this forum, many experienced traders began to share their invaluable ideas and methods to me selfishlessly, or contributed their time on testing my EAs/indicators to make them better. In the process I have learnt a lot and at the same time I realized that greater results can be achived if we do it in a more organized way.

I guess I don't need to explain how a reliable EA can help ours trading life. The key word here is "reliable", or quality. Most of the EAs out there lack the quality because they are made by one or an extremely small group of volunteers, who usually do not have much time, much trading or programming experience and usually do not want to share their ideas with others (give) therefore can't get much help from others too. All these can be solved if we can begin to share and knows how to share.

Yes, open source EA development, that's what I am talking about. It's like building "linux" in EA world so we all benefit from it.

In fact in FF we are in a perfect situation here to make it happen. I will address what elements we need to build a reliable EA in the next post (I hate long posts, and this one is approaching my limit already).

BTW, my first language is not English. I will try to make my point clear but there is a "win/loss" ratio here I don't think I can overcome in this life.
Target: Trading naked.
  • #2
  • Sep 28, 2006 12:55pm Sep 28, 2006 12:55pm
To develop a successful EA, we need:

1. A successful system that is mechanical enough to be coded.
2. sufficient test to make sure the EA follows the system.

If the system on which the EA is built is a failure, there is no way the EA can be successful. If you 10 billion dollar system involves augury, I am afraid that we can't make a EA for it. If an EA is claimed to follow a successful system but have bugs that change the supposed behavour of the EA from the system, the EA is not based on the system anymore.
Target: Trading naked.
  • #3
  • Sep 28, 2006 1:13pm Sep 28, 2006 1:13pm
If you have a system, do you want to know if it's going to be profitable in long run? Have you ever felt that you could improve it but just lack of statistics to back the modifications you have been thinking about. Do you want to know what's the best ST or TP is for different pairs? And when the market sentiment changes, do you want to quickly sense it in your system and quickly adept to it?

You may do all these with manual back testing or forward testing in a long and fatiguing process. But if there is a RELIABLE EA available, you can do it in minutes.

Many members here don't trust EA backtesting because they saw too many unreliable results from unreliable EA or generated based on unreliable history data. If there is a team of volunteers to do it in an organized way, it can all be overcomed.

But again, there is no room in this thread to argue about the integrity of backtesting method. It's fine for me that you don't trust it. Is it fine for you that I trust it?
Target: Trading naked.
  • #4
  • Edited 2:33pm Sep 28, 2006 1:20pm | Edited 2:33pm
Have you felt exhausted physically and mentally in the process of finding a consistently profitable system? That process don't have to be like that if you are not doing it alone.

In an organized way, we can give what we can contribute, either to code the EA, or to review the code, or to test it, or to optimize it. We should be able to find out the quality of a system much sooner than we test it alone.

Yes, you can just jump in a system because of the amazing data being posted by others, and actually find out that's your dream system that suits your personality and style. That's fine for me. But personally I am a little tired of it.
Target: Trading naked.
  • #5
  • Sep 28, 2006 2:01pm Sep 28, 2006 2:01pm
Here is the EA I made for Mouteki system. It is currently being tested by many members on this forum. You may downloaded here and either backtest it or forward test it.

You may test it silently, there is no obligation to post your findings (bug report, trade setup or results, ideas to modify, etc.) here. But if you would like to share your findings, please do it in an organized but convenient way.
Attached File(s)
File Type: zip Mouteki EA.zip   4 KB | 3,506 downloads
Target: Trading naked.
  • #6
  • Sep 28, 2006 2:19pm Sep 28, 2006 2:19pm
Here is a EA I made based on Bago crosses. I already embedded some filters in the EA so you won't see as much as signals the original Bago crosses would produce.

I call it Bago crosses (EMA&RSI crosses) instead of Bago system because Bagovino applies his own filters on the signals generated by Bago crosses and those filters completes the entry condition of his system. Because I haven't found a good way to code his filter set yet so what this EA base on is actaully only the Bago crosses plus some of the filters I added on the top.

This EA is profitable for EURUSD GBPUSD USDCHF AUDUSD in the backtesting from 2004. However you are welcomed to challenge this conclusion by testing it on other pairs.

I have forward tested it for about a week and the result was positive. But since I begin the forward testing of Mouteki EA I have to stop this one because I don't have that many computers to test all the EAs I have. If someone can pick it up and post the result from time to time here that would be great.

The Bago_AHA indicator use the same rules as the Bago EA, but instead of placing a trade it put an arrow on the chart and send an email to alert you. You can set it up so the email is sent directly to your cell. In that way you won't even miss the signal at the night (providing that you can be woken up by your cell. Sometimes that fails on me). The signal is sent 2 minutes before the bar close, which should buy you some time to rush to your computer and maybe apply your own filters to decide to take the trade or not.
Attached File(s)
File Type: zip Bago_AHA.zip   8 KB | 3,065 downloads
Target: Trading naked.
  • #7
  • Sep 28, 2006 4:29pm Sep 28, 2006 4:29pm
Thanks for these EAs to test AHA.

I am mostly a believer in EAs and their possiblities to help computers help US!

I will be testing these. I am afraid I must have missed the BAGO Crosses. Is it in another thread somewhere? Thanks again and I will be posting my results back to the thread.
Don Life is expensive, but includes a free trip around the sun.
  • #8
  • Sep 28, 2006 5:19pm Sep 28, 2006 5:19pm
You are welcome.

Bago crosses is what I call the basic setup of Bago system. You may find it in the post #1 of Bagovino Method thread, which is here:

http://www.forexfactory.com/forexfor...ead.php?t=7603

It generates signals at 5/12 ema cross and rsi cross 50. Bago do not trade every signal it generates, but use discretion to filter out false signals. So the discretion he applies (TL and Fibo and maybe something else he is working on) plus the bago cross complete the bago system.

My EA haven't included his discretion part yet so it's not an honest Bago system follower. I thought this is important, that's why I specifically point it out that the EA is based on the bago cross, not the bago system.

I hope I haven't confused you even more...



Quoting traderone
Thanks for these EAs to test AHA.

I am mostly a believer in EAs and their possiblities to help computers help US!

I will be testing these. I am afraid I must have missed the BAGO Crosses. Is it in another thread somewhere? Thanks again and I will be posting my results back to the thread.
Target: Trading naked.
  • #9
  • Sep 29, 2006 1:04pm Sep 29, 2006 1:04pm
Quoting aha
Here is the EA I made for Mouteki system. It is currently being tested by many members on this forum. You may downloaded here and either backtest it or forward test it.

You may test it silently, there is no obligation to post your findings (bug report, trade setup or results, ideas to modify, etc.) here. But if you would like to share your findings, please do it in an organized but convenient way.
aha its still opening some trades 1 candle late.
  • #10
  • Sep 29, 2006 1:51pm Sep 29, 2006 1:51pm
great work- AHA THANKS when i go to load the ea on the chart- nothing happens- all my others work-any thoughts?


Quoting aha
You are welcome.

Bago crosses is what I call the basic setup of Bago system. You may find it in the post #1 of Bagovino Method thread, which is here:

http://www.forexfactory.com/forexfor...ead.php?t=7603

It generates signals at 5/12 ema cross and rsi cross 50. Bago do not trade every signal it generates, but use discretion to filter out false signals. So the discretion he applies (TL and Fibo and maybe something else he is working on) plus the bago cross complete the bago system.

My EA haven't included his discretion part yet so it's not an honest Bago system follower. I thought this is important, that's why I specifically point it out that the EA is based on the bago cross, not the bago system.

I hope I haven't confused you even more...
Trade what you see - Not what you think
  • #11
  • Sep 29, 2006 2:13pm Sep 29, 2006 2:13pm
Which one did you load?



Quoting summitfx
great work- AHA THANKS when i go to load the ea on the chart- nothing happens- all my others work-any thoughts?
Target: Trading naked.
  • #12
  • Sep 29, 2006 2:21pm Sep 29, 2006 2:21pm
I thought i have solved the problem....

There is one possibility though. Sometimes the price broke both trend lines, i.e. below the bottom TL and above the top TL at the same time. Since Mouteki did not say anything about this situation, I decide to let EA do nothing. I have seen that delays the entry before, but don't know if that's the case you encountered.

Could you post your chart here so that I can anlyze the setup and why the EA enter the trades late? Thank you!


Quoting Attiqul
aha its still opening some trades 1 candle late.
Target: Trading naked.
  • #13
  • Sep 29, 2006 3:37pm Sep 29, 2006 3:37pm
I'm able to load the bado but not bago-aha which is the one I'm after. When I attempt to open it or drag it onto a chart-it just doesn't open? Sounds like a great ea with the email notifier and such. Any help aprreciated!


Quoting aha
Which one did you load?
Trade what you see - Not what you think
  • #14
  • Sep 29, 2006 5:59pm Sep 29, 2006 5:59pm
Bago_aha is an indicator. it should label some arrows (entry) and crosses (exit) on your chart... it works here... too little information so I have to the goose now ... have you put it in the indicators directory?... have you compiled it?... the arrows are red and green what 's your background color?... right click on your chart and select indicators list do you see Bago_aha.... are you using MT4 or 3?



Quoting summitfx
I'm able to load the bado but not bago-aha which is the one I'm after. When I attempt to open it or drag it onto a chart-it just doesn't open? Sounds like a great ea with the email notifier and such. Any help aprreciated!
Target: Trading naked.
  • #15
  • Sep 29, 2006 6:08pm Sep 29, 2006 6:08pm
Quoting aha
Here is the EA I made for Mouteki system. It is currently being tested by many members on this forum. You may downloaded here and either backtest it or forward test it.

You may test it silently, there is no obligation to post your findings (bug report, trade setup or results, ideas to modify, etc.) here. But if you would like to share your findings, please do it in an organized but convenient way.
Aha thanks for sharing and starting up this great thread.

I'll download your mouteki EA and test it.
I am wondering have you taken into account two scenarios with this EA?

1) two adjacent candles with equal highs (for a resistance point), or two adjacent candles with equal lows (for a support point). mouteki says we're to use the most recent candle in that case

2) in determing a valid support or resistance (in real time) we have to wait for the two lower highs (for resistance point) or the two higher lows (support point) to form or close. in some cases they close, then establish a new trendline and simultaneously break the new trendline formed. does your ea trade this?

also in some cases they break both the support and the demand trendline (in which case a trade should not be taken). an example of this is shown in the attachment below

thanks AHA

Phoenix
Attached Image
  • #16
  • Sep 29, 2006 6:23pm Sep 29, 2006 6:23pm
Thank you for your post! Your questions are all on the points and yes, they have all been taken care of in the EA:

1) The EA use the most recent one.
2) In the first 10 seconds of a bar open, the EA will:
a. recalculate trendlines (no matter the TL moves or not)
b. check the setup in recent three bars and TLs
so yes, the EA trades on the latest setup.
3) when break both TL, EA will close existing trades and do nothing.

For #3, the rule is a little simple right now. I will polish it when I get time.

Hope this answers your questions.




Quoting phoenix
Aha thanks for sharing and starting up this great thread.

I'll download your mouteki EA and test it.
I am wondering have you taken into account two scenarios with this EA?

1) two adjacent candles with equal highs (for a resistance point), or two adjacent candles with equal lows (for a support point). mouteki says we're to use the most recent candle in that case

2) in determing a valid support or resistance (in real time) we have to wait for the two lower highs (for resistance point) or the two higher lows (support point) to form or close. in some cases they close, then establish a new trendline and simultaneously break the new trendline formed. does your ea trade this?

also in some cases they break both the support and the demand trendline (in which case a trade should not be taken). an example of this is shown in the attachment below

thanks AHA

Phoenix
Target: Trading naked.
  • #17
  • Sep 29, 2006 7:03pm Sep 29, 2006 7:03pm
anyone up for backtesting these?
  • #18
  • Sep 29, 2006 7:03pm Sep 29, 2006 7:03pm
Foolish me- i thought it was an Expert Advisor. That explains it.
Thanks and also for writing it!


Quoting aha
Bago_aha is an indicator. it should label some arrows (entry) and crosses (exit) on your chart... it works here... too little information so I have to the goose now ... have you put it in the indicators directory?... have you compiled it?... the arrows are red and green what 's your background color?... right click on your chart and select indicators list do you see Bago_aha.... are you using MT4 or 3?
Trade what you see - Not what you think
  • #19
  • Sep 29, 2006 7:37pm Sep 29, 2006 7:37pm
You mean the Mouteki EA? Yes, many people including myself are testing it. From backtesting result apparentely it's not holy grail, but it's working extremely well in this week.



Quoting bmwboyee
anyone up for backtesting these?
Target: Trading naked.
  • #20
  • Sep 29, 2006 8:39pm Sep 29, 2006 8:39pm
Regarding your comments lines from Mouteki EA.
// Entry Conditions:
//
// 1. Enter long trade
// 1) Enter long when price close higher than the top trend line. The
// entry price is the first open price above the top trend line.
2. Enter short trade
// 1) Enter short when price close lower than the bottom trend line.
// The entry price is the first open price below the bottom trend
// line.

Reading Mouteki's document, I understood that to open a position you don't have to wait until the candle is closed. I don't know if your EA open in this way the positions, or not . Or if this issue has been clarified somewhere outside the original document (forum) by Mouteki.
The trend isyour friend untilis about to end...
3 traders viewing now