puzzlebird
-
Posts
35 -
Joined
-
Last visited
Content Type
Profiles
Forums
Articles
Posts posted by puzzlebird
-
-
Re: EXMQ1 Expert Advisor - 50% Profit Monthly
http://www.mediafire.com/?2zcajtnnwhi
As requested by a few guys, here is the 2009 back test using Dukascopy data in MT4, 99% modeling. Risk level is 4%, you can increase to 7% so that profit will be much higher...
-
Re: EXMQ1 Expert Advisor - 50% Profit Monthly
MetaCrazy,
this is what original version lot size:
lot1 = 0.1 (assuming it is 0.1)
lot2 = 0.1
lot3 = 0.3
mine is
lot1 = 0.1 (assuming it is 0.1)
lot2 = 0.2
lot3 = 0.2
This keeps the same risk level (almost), but high profit potential as to the back test.
In my forward test on demo account, it has three trade today. 2 win, one loss (close because it last more than two hours)
-
Re: EXMQ1 Expert Advisor - 50% Profit Monthly
I should clarify that what I mean for "Set SL to 43" is following expression
stoploss = MathMax(2*iATR(Symbol(),PERIOD_H1,14,0)/gPointValue, 43);
So if the market is relatively volatile and not filtered by market filter, the SL may be larger than 100pips, but it would rare, and lot size will be reduced accordingly to manage the corresponding risks.
-
Re: EXMQ1 Expert Advisor - 50% Profit Monthly
Thanks for the feedback. Because the revised EA is more for myself (in FXCM), it does work there in both backtest and live test. If you decompile my ex4 file, you may check on GetLots functions to see what might be wrong. You may also turn on "DebugMode" parameter to see more printout, that may help as well.
On the other hand, you may set "FixedLots" to non-zero value, so that the dynamic lots function will be overridden..
The revision on Turbo is to change the indicator itself with limited bars to consider. so you may consider using mine updated version to get much better performance in back test.
-
Re: EXMQ1 Expert Advisor - 50% Profit Monthly
Here you go. :geek:
It is my revised version of EXMQ1, totally re-written.
Changes are:
1. Set SL to 43
2. Support Duskascopy testing with DST support (birt's method) -- Set TestWithDukascopy to TRUE
3. Use OrderSendReliable library
4. Use MaxLots theory to enable dynamic lots
5. More reliable program logic and structure (my own framework)
6. Fixed Turbo-JRSX2 indicator so that it runs much much faster
7. Added market filter to avoid volatile market
8. Reduced the trading window to GMT19-GMT23
9. first trade will closed within 2 hours, two supporting trades will be close within one hours after opening (this is an important feature to control risk)
It is better to test with MT4 (Dukascopy) so that modeling accuracy is 99%, my test for 2009 fully year is 300% increment. Using MT4 its own backtesting, the accuracy is only 25% and test result will be misleading. <):)
http://depositfiles.com/files/fa6rqgdr9
:D
Kudo pls if you like it. :-bd
-
Re: EXMQ1 Expert Advisor - 50% Profit Monthly
Forget to mention, another change I applied is to set working hours from 19-23GMT, rather than 19-24 GMT, and use a filter system based market volatility in last one hour.
-
Re: EXMQ1 Expert Advisor - 50% Profit Monthly
I would like to share my tip on this:
the logic on this EA works. I re-program it into my scapler EA with SL=43, and MaxOpenHours for each order is 2 hours, it works well in for 2009/2010 test (Dukascopy version).
-
Re: BBW Scalper
Hi birt,
I am able to download your educated files, but not able to decompress the BBW_Scalper_v4_4_0_Pro.ex4 file. Interestingly, the rest of files are OK.
! C:\Download\b.rar: CRC failed in the encrypted file BBW_Scalper_v4_4_0_Pro.ex4 (wrong password ?)
Can you help verify? Thanks a lot.
-
Re: !!! MetaTrader 5 !!!
are we able to connect to any MT4 broker using MT5 client? big question is that if any existing EA/Indicators works without modification? Any insight will be helpful. anyway can not wait to test it.
EXMQ1 Expert Advisor - 50% Profit Monthly
in MetaTrader Expert Advisors
Posted
Re: EXMQ1 Expert Advisor - 50% Profit Monthly
1. When you specify "TestWithDukascopy" parameter, the GMToffset is set internally to override brokertime
2. When using dukascopy data, M5 includes all the ticks as M1 test, so it does not matter.
Forward testwith my own copy today has 4 trades, all wins.