Collaboration Dolly + Isakas + Nina System - page 16
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Testing DIN_EA_v2
I forgot to mention. If you run backtest on this ea you will get more accurate results running the test on 5 or 1 minute timeframes. The signal timeframe is set to 30 and the trend timeframes are set to 240 and 60. The reason for the accuracy problem has to do with exit more than entry. With 15 pip take profit you do not know exactly when the exit occurs when testing on higher timeframes. But with 1 minute is it closer to real trading. Not exact but closer.
Robert
I have modified KUSKUS_EA to this one.
I think I have all the rules correctly now but I have added a method to determine what works best.
There are a list of integer inputs starting with usexxx that can be set to 0 for do not use or 1 for use. That will allow backtest to use Start 0, Step 1 and Stop 1 to determine what really helps the strategy.
BTW the function in DIN_EA_v1 for BBandsStop does not work correctly.
The default settings match KUSKUS_EA.
Hopefully the input names are clear enough to change the settings to match what DIN is attempting.
Please note that Trading Hours are for FXDD (GMT+2), yours might be different. One problem that might need to be addressed is if your Asian session starts before 0. I think IBFX starts at 22:00.
I am running a backtest for all the usexxx from Jan 1, 2007. Then will run from Aug 2006.
This should determine what is adding to the strategy.
Also any new ideas can be implemented in the same way and tested with 0 and 1 to see if it really helps.
I do not remember if I already sent the HMA3 indicator. It is included here just in case. It allows the EA to modify the settings. There is a thread discussing the HMA and the settings of 0,0 (SMA, PRICE_CLOSE) seem to work better for forex than the default settings of 3,1 (LWMA, PRICE_OPEN).
RobertThank you and wow, alot of settings. Can you post one of your .set files that has been working great. I don't seem to be that good at my own settings. I think I try and make it to conservative and it doesn't work out for me. Ii would probably help if I new what all the options meant also
Thanks
I forgot to mention. If you run backtest on this ea you will get more accurate results running the test on 5 or 1 minute timeframes. The signal timeframe is set to 30 and the trend timeframes are set to 240 and 60. The reason for the accuracy problem has to do with exit more than entry. With 15 pip take profit you do not know exactly when the exit occurs when testing on higher timeframes. But with 1 minute is it closer to real trading. Not exact but closer. Robert
How tineframe for forward test ? M30?
One question, can you add fuction BreakEven after x pips to entry or entry +- relative pips?
hi all i'm new in this forum i'm interesed this strategy for me is very good.
how do you trade this day ?? show me your trade guys.
all greats for you
DIN_EA_v2
Sorry for all the inputs but it will allow for better testing to optimize the EA for each currency pair.
I do not trade this system, I only wrote the EA based on documentation. So I do not have the indicators on the chart.
The default settings match my current understanding of the strategy.
The only setting that might be changed is set UseTradingHours to false. I do not remember where but someone asked for the EA to have 3 trading sessions so that is in the EA. There was a post on the Yahoo Metatrader group that a live trade was placed correctly at 0:00 FXDD time but the entry should have been much earlier. Otherwise the trade was correct. The entry at 0:00 was because the trading hours were set to 0 to 3 for the asian session. and trade NY session was set to false. On my backtest that is when most of the losses occured. That might be different with the current version so needs to be tested.
I am currently running a backtest on the GBPUSD to check what indicators work best for entry and exit. It has run for 8 hours so far, only 4 more hours to go.
When finished I will use the results for default settings of the useXxx inputs.
I will also do a doc file to explain the different inputs.
Hopefully others on this thread can run backtest for other currency pairs are report the results. Then I can add code to the EA to determine settings based on Currency pair.
The doc file will explain further what I mean about initial backtesting.
BTW current high return on backtest is over $1300 for January. I think I am testing with either $3000 or $5000 initial balance.
Robert
I dont understand, why EA still buy gbpusd...no init first signal... no H4 and D1 trend long direct?
I think, open trade only first correct signal (first bar or second) and no anytime.
To frantacech,
have you tried to hedge not by opening setting ReverseSignal = 1, but by setting in MT options for one window ONLY SHORT positions, and for another window ONLY LONG positions? EA choose the direction which can increase the probability of right enter, so when there will be absolute signal for buying, there will be one position in the market, and if the trend is up, EA will just buy without opening sell position for hedge by second window's EA. But when trend reverses, the first EA will be opening 0.2 0.3 0.5 etc., while the second EA will be trading by down trend. I belive, these manipulations will decrease the number of bad entries in a strong trend, as positions will be opening in accordance with the trend. However, in that case we should restrict opening too many positions in block.
ooops! I missed the thread! sorry!
DIN_EA_v2
Thank you for testing the EA.
First, what are your times to trade settings?
As I mentioned in an earlier post the EA is set for FXDD times which are GMT + 2. I do not know what Velocity uses for their server times.
Also it looks like your trade is during the NY session and that is set to false by default. When did you start the EA?
Second, I am not using GMACD for trend as the default. I use HMA on 4 hr only by default.
Third, when both trends are used the default is 4 HR HMA and 60 minutes HMA. You would need to change the timeframe settings to 1440 and 240 for the 4 hr and 1 hr timeframe inputs.
To use GMACD and DigiStoch they need to be selected for the two filter methods. That is why I included string prompts. Otherwise it would be difficult to know which method to use by number.
As for limiting trades to first signal only that might miss some good trades if ther is a move up with a retrece closing the trade followed by another muve up that would reach profit.
That is why I am running the backtest to determine what combination of indicators might work best for entry and exit.
Robert
I dont understand, why EA still buy gbpusd...no init first signal... no H4 and D1 trend long direct? I think, open trade only first correct signal (first bar or second) and no anytime.
Adding Scalp trades
I have a method already in place to add scalping to the DIN EA. That will happen once the EA is at a better place as far as indicators to use and what to use for determining trend.
It is based on the TriplePlayTesting EA I developed for the yahoo group.
When a trend following trade is opened it opens scalp trades at 10 and 20 pips above/below the entry with take profit of 20 for the first and 10 for the second. It has worked well added to other EAs I have developed. I hesitate to make the takeprofit smaller as many brokers will not allow scalping. With 20 and 10 pip targets it usually takes longer fot the trade to close so there is no alert to the broker that you are scalping.
Also the orders are determined internally instead of using stop orders.
I also have code that will close trades at any profit level by using internal take profit.
Robert