|
|||||||
| Free Forex Trading Systems Got the "Holy Grail" system? Want to share it for free and become everyone's hero? This is the place to do it. (No advertisers please!) Also, follow along as our very own Pip Surfer posts daily updates from his Cowabunga System in the Pip My System Forex Blog. |
 |
|
|
LinkBack | Thread Tools | Display Modes |
06-19-2008, 07:00 AM
|
|||
|
|||
Hi MT,
I was looking through your code, and it's relatively easy to read (especially considering this is my 1st crack at dealing with this platform), so you definitely get points for that...  That said though, I think you may have misinterpreted the OP's system a bit, as I believe that his entry rules, divided into "1" and "2" for both long and short, are 2 separate sets of entry rules. In other words: Quote:
Or if MACD is positive, and your EMA cross each other upwards. Assuming I interpreted the OP correctly, I'm thinking your entry conditions should look something like: Code:
if((MacdPrevious<0 && MacdCurrent>0 && Rsi>50) OR (ShortEma>LongEma)) // or split it up into 2 separate if statements, which I think would be preferable... Edit: I'm trying to get a grasp on the platform as we speak, so I'll try to see what I can do with it when I do... Hope you don't mind me working off of yours? ^_^ Last edited by yoni45; 06-19-2008 at 07:17 AM. 
|
|
06-19-2008, 03:36 PM
|
|||
|
|||
|
Hello Everybody,
I have been out for a bit and havent really done the testing w/ the EA, it was a quick project. I agree that I misunderstood the entry rules, I will make the corrections tonight and post them. Feel free to make edits to the EA and just use it as a starting point, just post what you come up with so we can get this working correctly. When I get home tonight I will post the new EA w/ my Skype Info, if anyone would like to work on the EA together that would be nice. I am very new to coding and forex in general so if we can format the EA and System to produce profits I say "lets get it done" I am usually online around 11pm/midnight-ish till Mountain Time (US).
|
|
06-19-2008, 06:34 PM
|
|||
|
|||
|
Alright, here's what I have so far, it seems to follow the rules to the dot, but also seems to need a bit of tweaking to weed out fake-outs...
Over 2 months of backtesting, it registered 36 entries, and it seems to have lost approx. 600 pips... That said though, after weeding through the charts, it seems much of that is based on sideway movements where EMA's just happen to barely cross each other or MACD's just happen to barely cross over... Here's the what I have so far if anyone wants to tinker with it - and I'll see if I can tighten the rules... (Er... this is a tad awkward, but my file attachment options aren't available... do I need a certain post count? =/) Edit: Nvm, found it, was a browser thing... Last edited by yoni45; 06-19-2008 at 06:37 PM.
|
|
06-19-2008, 08:19 PM
|
|||
|
|||
|
Ok, so I've done a bit more tweaking to the entry conditions...
First off, what I did was set buffers for what constituted a "crossover", whether MACD or EMA. In other words, in order to be considered a crossover, the values not only needed to cross each other, but cross each other *and* get a certain value over the cross. So, for example an EMA crossover only becomes a 'true' crossover, if it goes up by more than 0.0002 (or whatever number) over the cross... This should weed out cross overs that barely touch each other on a horizontal pattern... I also added the ability for similar buffers to the RSI, so you can choose whether you want >50 to actually be >50, or perhaps only if its >55... after some backtesting using the last 2 months... a. the MACD cross over entries seem to lose overall, regardless of any buffers I put between them... b. the EMA entries, using the buffers, substantially improved. In fact, when taking the MACD entries out of the picture, and using an EMA buffer of 0.0007, what was previously a loss of ~650 pips over those 2 months, turned into a gain of ~300 pips. 15 trades made, 12 won, 3 lost. c. RSI buffer seems to help, but only barely when using a buffer of 5 (anything more loses more than it gains)... I've attached the most recent version of the EA - this one has a whole bunch of external parameters that you can change and tweak, so you can play with it yourself if you wish and see what gives you the best results... Lots = 0.1 - Number of lots to bet TakeProfitNum = 0.0025; - This is the take-profit number of pips as per #2 rule set - default to 25. StopLoss = 30; - This is stop-loss, default as per rule set. EmaBuffer = 0.0007; - Buffer used for EMA crossover entries MacdBuffer = 0.1111; - Buffer used for MACD entries (should be set to ~0.0001-0.0005, default to 0.1111 to disable MACD entries). MacdBufferB = 0; - Buffer used for the MACD variable for EMA entries. RsiBuffer = 5; - Buffer used for the RSI check. TakeProfit = 25; - Not used. TrailingStop = 0; - Not used. Last edited by yoni45; 06-19-2008 at 08:24 PM.
|
|
06-19-2008, 09:18 PM
|
|||
|
|||
|
Very Nice Yoni.
I like the new ideas and how you worked them into the EA. Because you took the time to do that, I was able to spend my time doing some backtesting. I would suggest moving your Take Profit to 0.0040 and you will see the best results (Testing from Jan 1st 2008 - Current.)
|
|
06-19-2008, 09:42 PM
|
|||
|
|||
|
Quote:
It doesn't seem to be as good on all the currencies though (I've been testing on EUR/USD, I'm assuming you have been too) - which begs the question as to whether this algorithm is simply tweaked to our test data, or just to this pair (if it's the pair, then it's not so bad)... Any chance you have more data going further back than Jan. 1st? I'd set it up myself, but I'm low on disk space and don't have much time left to try it at the moment. If you could give that a shot to see if it still gives positive returns on (say) Nov-Jan, or Sept.-Nov. etc., it'd give us an idea of how useful this may be... ^_^
|
|
06-19-2008, 10:05 PM
|
||||
|
||||
|
@MT and Yoni,
WOW. Great work guys! I cannot help you at all as far as MT4 and coding, etc but will definitely use this when it is ready and share my experiences. Can you also make a version (or tell me what to do) to just make the EA give a signal but not do all the trading? Again, I don't know what I am asking because I am not familiar with MT4. I am coming over from DealBook 360. It is sweet and a DREAM to use but not enough coders for it (everybody does everything for MT4)and having to keep outfoxing GFT to get another 30 day demo is too aggravating. lol I have to say, MT4 is really clunky! Why is there no feature to just copy a chart and change the instrument? GEEEEZ! Apparently, MT5 will not have that feature either. UGHHH! Anyway, thanks for all your hard work and for helping keep this system (and thread) alive.
|
|
06-19-2008, 10:08 PM
|
||||
|
||||
|
Quote:
|
|
06-19-2008, 10:39 PM
|
||||
|
||||
|
One other thing
I don't know if this helps or matters but I found this and thought it was cool. Expert Advisor Builder for MetaTrader 4 It is a kind of auto EA builder for MT4. Believe me, I am not sure if it is good or not, but thought it was interesting and maybe either one of you could use it.
|
|
06-20-2008, 06:38 AM
|
||||
|
||||
|
Sorry to jump in without reading this thread. I just wanted to offer my 2 cents on the EA builder above. It's pretty good; however, what I did was create a dummy EA with it so see how it does it's coding, then copy/pasted the template (from the first page) into MetaEditor and save it as a template. I code what I want myself and find it almost as easy, but with more flexibility than with the editor... If you want more info, feel free to PM me or email me at forex@renocomputerservices.com. Please not that this is not a commercial email address, just an email address I designated through one of my domains for Forex related emails...
Happy PIPing 
|
|
| Thread Tools | |
| Display Modes | |
|
|
Linear Mode
