Jump to content

clovelly

Members
  • Posts

    180
  • Joined

  • Last visited

Posts posted by clovelly

  1. Re: My custom trading system-STEALTH Based

     

    seems to do a little better than break-even. Has TONS of potential. Sharing it with everyone and hoping someone can make it better.

     

    Enjoy everyone.

    The template I am providing shows what the STEALTH is really doing in the ares that you would normally have no color or blank spaces. The EA will only trade on trend "flips".

    Tek, did you write or test this EA? The EA does not use any Stealth-based indicator, only uses 2 MAs to make trades (above both to buy, and vice versa). Backtesting does not show any profit.

  2. Re: Need help. Most EAs always stop working after some days.

     

    You had this problem already too?

    I have it very often and it really disturbs.

    Always it stops after 1-2 weeks and I don't know why.

    Then I lose trades. I wanted test EAs without losing some days...

    It's really bad.

    Ok. I will reinstall Metatrader. Maybe that helps. But I don't understand why others don't have the same problem like me.

    Most of these EAs will not trade often as they are scalpers, so maybe if you can switch off the PC after closing of FX markets on Friday midnight and turn it back on again late Sunday, ready for the new week. That also gave a chance for clearing up memory blocks due to bad allocation/release. If the EA is designed correctly, resetting a PC does not interfere with its operation.

  3. Re: Artemis Sitter 3.55 for YOU!!

     

    Inside manual it is mentioned that

     

    "you can change stoploss and profit by using mouse on the screen"

     

    Does it work for you?

     

    :?:

    It works, slowly. In fast market it is of no use. Sometimes it misses the keyboard shortcuts (Alt+n), you really have to hold the keys until it receives.

  4. Re: REQ: forex derivative EA

     

    This is from 4xproject.com:

     

    As we got informed there is an new version of derivative available. This new version has the possibility to lock a trade in profit. That feature reduces the number of possible loosing trades but will reduce the overall performance as well as the trades have less room to swing.

     

    We got an explanation as well about the backtest section in the code of derivative. The expert advisor has initially be designed before the new hedging rule was in place.

    To cover the new hedging rule even in backtest they added a code which some people claimed to be to fake the backtest. We took a look into the code and the activation date for that part of the code is 18.05.2009 which fits to the explanation.

     

     

    what do you guys think of this "explanation"?

     

    FD is alarmed by our diagnosis, as traders at other forum are now aware of that fact too. The date is correct (when the new ruling started to be in effect) but I doubt about the "explanation": the EA already has the optional parameter for "hedging", so the correct backtest still could be performed by the right setting if so wished. Their argument does not stand. The fact that the code passes both the pair AND the year to the DLL does not look "healthy".

     

    As for their andeavour to improve the EA performance, this is expected for a supplier after a bad begining. Backtest & Forward testing will see to it.

  5. Re: REQ: forex derivative EA

     

    yep ive tested it over 18+ months.

     

    the only change i could find was the money management.

     

    do u want to comment out the calls and i can try that one?

    Thanks Storming. I already tried commenting that routine out but the testing of 2009 did not go far as it printed out lots of lines about values of DeMarker (probably for their debugging). Commenting out the line as you did is probably sufficient for testing past years (rather than just last 18 months, as the curve-fitting is performed for each year.

     

    Now, with what you & fxknight just found about the hardcoded years confirms that view.

  6. Re: REQ: forex derivative EA

     

    that's SNEAKY!

     

    i commented that line out (dont know if that was enough).

     

    ran the tests again. . . same trades, but reduced money management risk.

     

    drawdown was about 9% instead of 18%.

     

    profits were reduced.

     

    but it seems the trades were the same.

    I would like to comment out the called routine DefineParameters() (by inserting a return at its beginning) and test again. This routine is called at every tick, and it in turn calls the DLL to get back about 100 control parameters. These parameters then are used across the EA to determine actions. The code is hard to read & follow, but I seem to see those parameters in things like Timeframe/periods for various indi calls (RSI, DeMarker, MA, WPR, etc.), ATR factors, waves,... There is a set of pre-defined default parameters that can be used. Maybe I am too suspicious but I cannot see any reason for the DLL calls at every tick to reload these 100 parameters.

     

    Can you retest with your full historical data? I do not have the full set to test properly. Thanks

  7. Re: REQ: forex derivative EA

     

    After seeing a number of 'strange' behaviours of the EA, I checked the code and found this interesting line

     

    if (IsTesting() && TimeCurrent() < gi_1876) l_year_4 = TimeYear(TimeCurrent());

     

    This actually checks if it is backtesting and the time is before May18, 2009 then it will ask the DLL to give it the working parameters (defining everything, including S&R, SL, TP, etc.), otherwise it will work on a different set, again coming from the DLL. I suspect that the DLL is heavily customised to make a good backtesting, hence making it attractive for selling. The size of the DLL is so huge (500K) seems to convince me that it contains lots of past data. So, for now, I do not trust any backtesting, and will forward test much longer to see how the EA performs before deciding whether to keep the EA or not, or asking for refund.

  8. Re: Forex Killer Formula claims

     

    Hi,

     

    I came across this product called Forex Killer Formula (http://www.forexkillerformula). On the website, it says that this system will give 100% profit and no loss. I never knew that this was even possible.

     

    Therefore, I would like to ask anyone who bought this product to let us know whether it's true that one will not get any losses with this system.

     

    I wouldn't mind buying this product if it is that good, but I want to hear experiences of others first.

     

     

    Ken

    The guy is probably a scammer. One of my friends sent in the order (of price $100 as advertised on the website). After Clickbank processed, the site came back and said it needs another $100 because the price is now $200. My friend refused to pay the extra and asked ClickBank for the refund.

  9. Re: REQ: forex derivative EA

     

    Hi Guys,

     

    This is the EA FD with Trailing stop.

     

    Try it and let us know about your results: http://www.mediafire.com/download.php?jvo1wymk24z

     

    Good luck,

     

    Best wishes,

     

    Scarface

    Scarface, thank you for educating the EA. Can you also post the 'uneducated' source, as decompiled from the original? There seem to be a big difference in results between the original EA and the 'educated' (with no trailing) one. For GU on one of my tests, for example:

     

    the original EA: +6800 (38 trades) vs +3473 (24 trades) from "educated" one, for the period April-May09.

     

    Thanks

  10. Re: REQ: forex derivative EA

     

    Check forward test using Alpari UK demo account here

    http://www.4xproject.com/statements/derivative/statement.htm

     

    I naively put this on my live account on GBPUSD and EURUSD.. hm.. so far not so good.. :(

    hm.. I think we should wait for a few more weeks to see how it performs in forward test :evil:

    I wonder why 4xProject gets so many trades executed and the results are totally different from mine (only 6 trades). The difference in datafeed between Alpari & FX Clearing would not make that big deal. Anyone has other forward test result?

  11. Re: REQ: forex derivative EA

     

    Collin, there is nothing special with my backtest process (I think) but I just realised that my data quality is only 25%, so I will get the correct data and test again.

     

    I believe 25% modelling is as far as we get, since I am getting this value as well. Thanks sgteo for the clarifications.

     

    Can I confirm with you on your databank? It seems that you are using FXClearing. I have downloaded a copy of it but FXClearing does not provide the historical data dating back to 2008. Other than this, I am not sure if I miss out anything else. The parameters used are same as the original educated ex4, except hedging which is changed to 'true'. I am currently using v224 MT4 for the test.

     

    Norman, are you also using Alpari data platform for the testing? Can you also shed some light on your testing procedures?

     

    forexample, I am using Alpari MT4 platform and the databank provided by Alpari. I just test using the dll and the educated ex4 provided. Additoinal steps include converting the M1 data to M5, M15, M30, M60, M240 and D1 (or 1440). Ensure the "Allow DLL imports" and "Allow external experts import" are ticked (under Options->Expert Advisors).

     

    I am attaching the 2008 backtest results. Strangely this is the 4th attempt I am doing but the results are still the same. Can any parties help to provide some guidance on this?

     

    http://www.4shared.com/file/109094543/2fda52f3/MyBT.html

     

    Thanks again for all the help.

     

    Regards,

    Collin

     

    I did not download the data from Alpari database but use the Tools/History Center feature of MT4 (that's why the data quality is only 25%. The Alpari data server is currently down so we cannot download extended M1 data). I think maybe you are using version 1 instead of v1.1 or 1.2 of the EA. They have updated the EA so the "educated" is only v1. The version 1.2 is somewhere in this thread.

     

    You also tested only till 28Nov but the gain since then till the end of 2008 is minimal. Otherwise, overall the results are much the same (i thought you meant yours is much worse, but as the data quality is only 25%, the results sometimes vary by a small margin.)

     

    BTW, the EA just had a small win with GU (9 pips) on a 3pip spread broker (FX Clearing), but nothing on a 4-pip GoTrader.

  12. Re: REQ: forex derivative EA

     

    Hi guys..

     

    i need your help.. i used this EA but am facing the following problems:

    1: i installed the EA as mentioned in the manual but now whenever i open the metatrader 4 it crahes down giving all the

    technical details for the crash.

     

    2: where to get the key. they mentioned in the manual to download the key from the members area. when u go to the

    forex derivative members area it asks u to put in ur id and password.. what is the id n passoword here.. n where to get it

    i could not find anything on the main page where i can sign up or become a member.

     

    thanks (in Advance)

     

    1. I do not know the answer to this, probably a technical issue with your PC?

    2. The id & pwd are given to you once you buy the EA. For testing, you can use the key as from the "educated" version posted earlier in the thread. I already bought an official copy to support the programmer.

  13. Re: REQ: forex derivative EA

     

    Hi clovely,

     

    Despite serveral attempt for the backtesting, I am still unable to produce the results that you provide. Fact is my results are a far cry from the offical backtest shown in the offical website. Would you be kind enough to shed some light on your backtest process? I am currently using Alpari databank.

     

    Has anyone able to produce a similar result for all the years 2006 to 2009 shown in the website? Appreicate any feedback.

     

    Thanks.

     

    Regards,

    Collin

     

    Collin, there is nothing special with my backtest process (I think) but I just realised that my data quality is only 25%, so I will get the correct data and test again.

×
×
  • Create New...