This topic has been reported for inappropriate content
- This topic has 455 replies, 41 voices, and was last updated 8 years, 10 months ago by
.
- Posts
Is there any volunteer here to start assemble the contributed code into ONE EA?
1. Everyone can contribute snippets of code, correct bugs in existing code, add more functionality to code.
2. Most functionality should be in the form of functions.
3. The volunteer will assemble all parts into one, coherent, working EA – with version number (xxx_v.01.mq4). Each change/addition will increase the version, and will be published here.
Looking forward to start contributing.
G.
Hi GG53
waiting for your further hints on TREND and then we can start the EA..
I believe there are many members here ready for the contribution..
I will try my best as I really have less experience on coding an EA..
tankhang
Hi GG53 waiting for your further hints on TREND and then we can start the EA.. I believe there are many members here ready for the contribution.. I will try my best as I really have less experience on coding an EA.. tankhang
I need a Volunteer to assemble code and make versions, according to the guidelines above.
The target is that this community will create a sturdy, profitable EA with max features to stand almost all market conditions.
It will be with auto SL, Auto TP based on pair Volatility and auto Lot calculation based on Risk per trade.
According to the HUGE ammount of Volunteers standing in line so far…. it looks like almost everyone went to stand in the other line, waiting to download it….
So, I’m waiting too…
G.
Maybe many participants expect too much. But maybe you expect too much, too. Not all Penguintraders seem to be interested in your (genious) idea of trading. And most of them who are, are maybe no good programmers – think that’s the main problem.
In previous post you already gave some great hints regarding further steps where you’ve said that we have to use currency strength. Not sure if we can really trust the currency strength indicators already published. I read some of your replies to some of those indicators for which you state that they look very different to yours. Any suggestion what this difference caused/what to make baetter?
Anti,
I already heard that remark before. I’m veteran here and know many coders here. I know that not ALL are coders.
I already have several working EA’s, including this one (finished, full featured, working on live for the past 20 months), which I hope the community will build together. That’s what I DO expect from such Forum and community: mutual share and work. I don’t expect from EVERYONE to contribute, I know most can’t.
I don’t need this EA – but I know others do or want it, and I’m willing to share knowledge, help and guide.
But, if no one step forward and do some work, I’ll just join the party…
Maybe I have high expectations, but that’s me and I don’t intend to change nor want to.
G.
I understand that it may be frustrating. Perhaps during next week we (all those who have a day job) have more time. Perhaps we first should focus upon the procedure of identifying trend first. Other recommendations or still ideas from readers here …?!
I understand that it may be frustrating. Perhaps during next week we (all those who have a day job) have more time. Perhaps we first should focus upon the procedure of identifying trend first. Other recommendations or still ideas from readers here …?!
No need to concentrate on that.
Trend is just a filter for this strategy, and as you can see it’s already working and profitable.
It should be just another function, and a replaceable one.
Current results are going to be our base-line, which should improve by at least one aspect with each addition/modifiction. Whether it’s win/loss ratio, profitability or DD.
I already checked it on more H4 charts and it’s working with very similiar results on GBPUSD and even on the very volatile GBPNZD. The Peak & Trough algo need some refinements, as there are some differences between my results and the current indicator.
G.
Yes, I know. But as you mentioned before, an idea on general trend could help to select most profitable trades/currencies and to adapt TP/SL. Therefore, I think it’s worth to start an discussion on it as I think most volunteers (including me) don’t have a full understanding on what next steps should be. Yes, you still gave a detailled guidelines, but there are dozens of ways implementing it.
Is there any volunteer here to start assemble the contributed code into ONE EA? 1. Everyone can contribute snippets of code, correct bugs in existing code, add more functionality to code. 2. Most functionality should be in the form of functions. 3. The volunteer will assemble all parts into one, coherent, working EA – with version number (xxx_v.01.mq4). Each change/addition will increase the version, and will be published here. Looking forward to start contributing. G.
I volunteer to do it. I’m not at this forum everyday though, so my reactions may take some time. But if it’s not a problem then I would like to be considered
I did not follow each detail of this topic, so maybe I’ll suggest something that has been suggested before: what about a volume filter, like discussed elsewhere with @gg53?
I should be able to turn my Volume_Per_Session indicator into a set of functions. I think it would consist of three functions:
- a function that determines which session is active (London, London & New York, New York, Asian)
- a function that calculates session-wise average tick volume over the last 5 days
- a function that calculates current normalized volume in relation to session average
Function 3 would run on every tick. Functions 1 and 2 can run when a new bar appears.
Input parameters for session start hours would be required.
Basic idea of the filter would be to avoid entries at volume below session average.
@gg53: does that make sense for you?
@vlanfx: if I post the functions, would you be willing to integrate them into your EA prototype?s.
A good trader is a realist who wants to grab a chunk from the body of a trend, leaving top- and bottom-fishing to people on an ego trip. (Dr. Alexander Elder)
[quote quote=10337]Is there any volunteer here to start assemble the contributed code into ONE EA? 1. Everyone can contribute snippets of code, correct bugs in existing code, add more functionality to code. 2. Most functionality should be in the form of functions. 3. The volunteer will assemble all parts into one, coherent, working EA – with version number (xxx_v.01.mq4). Each change/addition will increase the version, and will be published here. Looking forward to start contributing. G.
I volunteer to do it. I’m not at this forum everyday though, so my reactions may take some time. But if it’s not a problem then I would like to be considered
[/quote]Not a problem at all. I’m not expecting full time workers.
As a matter of fact, I’m glad that you are the driver of this wagon….
Later on I’ll provide the EA dummy skelleton, and from that point our job is to fill it with content.
Thanks !
G.
I did not follow each detail of this topic, so maybe I’ll suggest something that has been suggested before: what about a volume filter, like discussed elsewhere with @gg53? I should be able to turn my Volume_Per_Session indicator into a set of functions. I think it would consist of three functions:
- a function that determines which session is active (London, London & New York, New York, Asian)
- a function that calculates session-wise average tick volume over the last 5 days
- a function that calculates current normalized volume in relation to session average
Function 3 would run on every tick. Functions 1 and 2 can run when a new bar appears. Input parameters for session start hours would be required. Basic idea of the filter would be to avoid entries at volume below session average. @gg53: does that make sense for you? @vlanfx: if I post the functions, would you be willing to integrate them into your EA prototype? s.
Remember that our base line is the posted EURUSD H4 results posted before. Other pairs perform roughly the same with same params, no optimizations.
I’ll be more than glad, but more important is that I’m sure the EA will be happy !!
G.
@gg53: ok, I did not keep in mind that the EA should work on H4. On H4, session averaging with 4 session per day will hardly make sense. I would consider M30 the upper limit, better M15.
But anyway: maybe it would make sense to check on lower timeframe (M5 or M15) whether there’s sufficient volume for an entry. Better entry timing, lesser drawdown?
Maybe this would be interesting for a later version to come.
My function set is ready.
s.
A good trader is a realist who wants to grab a chunk from the body of a trend, leaving top- and bottom-fishing to people on an ego trip. (Dr. Alexander Elder)
@GG53 Great, looking forward to building this EA
@simplex yes, as I understand it, the EA will need volume data, so it will be very helpful if you can provide some volume functions. We’ll see later on what exactly they should do. Also, running the EA on H4 may make it a little harder to get those volume data as H4 bars don’t necessarily align with sessions start/end.V.
Hello, I backtested this EA with various parameters. I can say, that this EA is profitable, because of trailing stop. Without trailing stop and with same TP ir SL parameters, the system is 50 win % like others. I have coded many EA’s without trailing and getting about 50% winning and I have never thought adding trailing stop to it. It’s a gold nugget for me. Thanks for that :)
Suggestion:
The EA is the most profitable on H1 and H4 timeframes. So EA could trade on both timeframes, not on one single timeframe. :)
@vlanfx: that’s correct! Like I posted before, the volume function set is useless on pure H4. On H4 there are 6 bars per day, and dividing those 6 bars per day in 4 session per day makes no sense. Even H1 is too coarse, imo.
The function set I coded yesterday is based on @gg53’s idea of averaging volume per trading session, like discussed over there. Basically I took the volume averaging posted there and converted it into a set of functions, that can easily be implemented in an EA.
gg53 started this thread with an idea in his mind, I’m sure of that. So I’d suggest to wait for his approval that the volume averaging idea makes sense here, and if yes, we go for it. If not, I’ll find different things to contribute here.
s.
A good trader is a realist who wants to grab a chunk from the body of a trend, leaving top- and bottom-fishing to people on an ego trip. (Dr. Alexander Elder)
Thanks to all comments and remarks.
1. The EA currently (as is) is just a rough skeleton, and is CURRENTLY profitable on almost all pairs (with same params) on H4.
2. ALL contributions are more then welcome.
3. In order to check contribution (or addition/suggestion) validity, it must ADD value to current version, whether increase PROFIT, reduce RISK, eliminate LOOSING trades, etc.
4. The EA is NOT going to be limited to H4 and above, but to start with, we need some base-line to compare the addition value.
5. VOLUME should play a part in that EA, the effect and power of the VOLUME was already proven here in more than one way.
6. Checking lower TF’s is very reasonable and valid contribution.
7. The Peal & Trough algo presented still needs some modifications, as it’s not quite finished as presented in the current EA.
Good luck, and thanks again for commenting/contributing.
G.
@gg53: ok, so I’ll prepare my volume per session for posting.
@vlanfx: I reviewed your EA this morning, identified your functionIsPeakThroughChange()to be the one that decides when to go long or short. Did you use this function in an indicator to visualize its results? If yes, would you share the code for comparison?Background: I wrote a similar one and would like to compare it to yours visually. If mine brings new aspects on chart, I would share it here.
s.
A good trader is a realist who wants to grab a chunk from the body of a trend, leaving top- and bottom-fishing to people on an ego trip. (Dr. Alexander Elder)
Attached is the initial EA template with some functions example.
VlanFx, Add the current logic based on this, and we’ll call it _v0.1
G.
Guys,
(off topic)
I’m worried about Saver0.
We haven’t “seen” him for a long time, and he doesn’t respond to Skype msgs.
I sure hope everything is Ok with him, and he’s just busy.
Does anyone have contacted him lately or know other ways to confirm that everything is Ok with him?
G.
Has anyone put the code for the EA somewhere in a VC system so everyone that wants to contribute can gain easy access instead of passing the file around here as attachments when work gets done? Github is a good place to start and I’d like to contribute as much as I can as time permits.
Has anyone put the code for the EA somewhere in a VC system so everyone that wants to contribute can gain easy access instead of passing the file around here as attachments when work gets done? Github is a good place to start and I’d like to contribute as much as I can as time permits.
Contribute here, Uradon Keeryder.
If We wanted somewhere private, we all know several other places.
This is a Community project, so everyone can participate & contribute, and you are more then welcome.
VlanFx is responsible for versioning, after comments and discussions of each contribution.
G.
I’m worried about Saver0. … I sure hope everything is Ok with him, and he’s just busy.
I seem to remember him posting earlier this year that he would change to a new day job, which would not leave him much of spare time for the forum. I hope he’s well, too.
s.
A good trader is a realist who wants to grab a chunk from the body of a trend, leaving top- and bottom-fishing to people on an ego trip. (Dr. Alexander Elder)
Ok, as announced earlier my ‘volume per session’ contribution to start our baby.
For the purpose of this thread, I stripped my volume per session indicator down to the minimum and placed the function set in an MQH library file for convenient linking to the EA.
So the indicator file may just act as an example for using this little lib. @vlanfx: feel free to ask if anything is unclear!
On the pics provided you see examples of how I’m planning to document my contributions to this little project, especially file headers and function headers. Should be easier to integrate and understand the contributions of a bunch of authors if everybody approves to some minimum standard of inline comments.
One pic shows my file header and an enumeration header, the other one a typical function header, with a short description of every parameter. Special techniques used can be described under ‘Remarks’.
@vlanfx: feel free to modify or enhance
If you would suggest any naming conventions to avoid problems during compilation, please post!s.
A good trader is a realist who wants to grab a chunk from the body of a trend, leaving top- and bottom-fishing to people on an ego trip. (Dr. Alexander Elder)
i did a little work to this so here is version 0.1
GmtOffset – makes a gmt time that will be used on TradeTimes
TradeTimes – you can use StartHour/EndHour
Cs – added simple curency stenght function – it does put ob/os currencys into array. builds pairs of em and if they are allowed they are put in trade_symbol[].. ide is that it will find proper pairs based on ob os conditions and trade thoes.. also is very so far and need futher work.
VolOsc func – added function that signals ob os on VolOscilator that was shared here somewhere.
Added stuff to initalize so we get less errors if something is wrong.
it may contain bugs tho i didnt tryed it out fully yet..
not that it can contain bugs. the InitCs function also needs more work..
[/quote]
If you would suggest any naming conventions to avoid problems during compilation, please post!
- You must be logged in to reply to this topic.