intraday DAX strategy 5min mini1€ spread 1

• 01/19/2017 at 1:16 PM #21674
  Raul Vg

  I share the robot here, so we can improve it and show me the backtest with 200000 bars

  // Definición de los parámetros del código DEFPARAM CumulateOrders = false // Acumulación de posiciones desactivada DEFPARAM FlatAfter =173000 once ordersize=1 HoraEntradaLimite = 090600 HoraInicio = 090500 if Ordersize>15 then Ordersize=15 endif if Time >= HoraInicio and time <= HoraEntradaLimite then c1 = open < close-2 if not onmarket then IF c1 THEN IF PositionPerf(1) < 0 THEN OrderSize = OrderSize/2 if ordersize<1 then ordersize=1 ENDIF ELSIF PositionPerf(1) > 0 THEN OrderSize =OrderSize+2 if ordersize<1 then ordersize=1 ENDIF endif buy ordersize*6 contract AT close+2 stop endif c2= open > close-1 IF c2 THEN iF PositionPerf(1) < 0 THEN OrderSize = OrderSize/2 if ordersize<1 then ordersize=1 ENDIF ELSIF PositionPerf(1) > 0 THEN OrderSize =OrderSize+2 if ordersize<1 then ordersize=1 endif ENDIF sellshort ordersize*6 contract AT close-1 stop endif endif endif SET STOP ptrailing 5 Q

  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53

  // Definición de los parámetros del código DEFPARAM CumulateOrders = false // Acumulación de posiciones desactivada   DEFPARAM FlatAfter =173000   once ordersize=1   HoraEntradaLimite = 090600   HoraInicio = 090500   if Ordersize>15 then Ordersize=15 endif   if Time >= HoraInicio and time <= HoraEntradaLimite then   c1 = open < close-2 if not onmarket then IF c1 THEN IF PositionPerf(1) < 0 THEN OrderSize = OrderSize/2 if ordersize<1 then ordersize=1 ENDIF ELSIF PositionPerf(1) > 0 THEN OrderSize =OrderSize+2 if ordersize<1 then ordersize=1 ENDIF endif buy ordersize*6 contract AT close+2 stop endif   c2= open > close-1   IF c2 THEN iF PositionPerf(1) < 0 THEN OrderSize = OrderSize/2 if ordersize<1 then ordersize=1 ENDIF ELSIF PositionPerf(1) > 0 THEN OrderSize =OrderSize+2 if ordersize<1 then ordersize=1 endif ENDIF sellshort ordersize*6 contract AT close-1 stop endif endif endif SET STOP ptrailing 5 Q

   

   

   

  

  Raul Vg

  Participant

  Topics: 4

  Replies: 177

  Been thanked: 83 times
  01/19/2017 at 2:01 PM #21677
  jonjon

  Raul

  I tested the code above from 2013. You lose all of your equity by Q2 2013 even with starting equity of EUR20k.

  I then tested it with a stop and target as per previous versions. With Stop 5, Target 10. Unfortunately you again lose all of your equity by mid 2013 (even with EUR20k starting equity).

  I (hopefully) attach the results from backtest with Stop 3, Target 12.  You again need starting equity of EUR20k. Note the drawdown of EUR23k (I’m not sure I could live with that) and the large number of contracts you are sometimes trading (not sure how realistic). It performs really well 2014+

  

  jonjon

  Participant

  Topics: 5

  Replies: 64

  Been thanked: 4 times
  01/19/2017 at 2:35 PM #21680
  CN

  Hey,

  My broker has 4 as min. SL in dax.
  What broker are you with?

  Could you post the results? There’s no picutre

   

  

  CN

  Participant

  Topics: 1

  Replies: 156

  Been thanked: 17 times
  01/19/2017 at 2:45 PM #21683
  jonjon

  For some reason the files won’t upload. I didn’t receive any error messages this time though. If I figure it out I’ll post the results. Sorry guys.

  I use Marex for most of my trading. I haven’t signed up to a broker yet to trade through PRT. Will do when happy with the demo results. Sounds like I’ll be looking at a stop of 4 like you guys. As I mentioned on the previous thread a SL of 4 and Target 10 also worked well.

  

  jonjon

  Participant

  Topics: 5

  Replies: 64

  Been thanked: 4 times
  01/19/2017 at 2:47 PM #21686
  jonjon

  Looks like it worked that time.

  I’ll post updated results after Draghi finishes his press conferene

  

  jonjon

  Participant

  Topics: 5

  Replies: 64

  Been thanked: 4 times
  01/19/2017 at 2:47 PM #21687
  Nicolas

  I made upgrade of forums during your last post, that’s why I believe the upload didn’t work at this time.

  About the strategy result, is anybody using the tick/tick backtest already?

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20863

  Been thanked: 4774 times
  01/19/2017 at 2:57 PM #21691
  jonjon

  Thanks Nicolas. I’ve got it ticked when I run the backtest. I know, the equity curve looks unbelievable. I’ll post results of previous test (with Raul’s older version of the code) later. equity curve looks more like you’d expect

  

  jonjon

  Participant

  Topics: 5

  Replies: 64

  Been thanked: 4 times
  01/19/2017 at 2:59 PM #21692
  Joachim Nilsson

  Hey guys!

  Thanks for sharing! Tested it on PRT 10.3 with IG and this is the result. Really promising if you ask me!

  I will look into the code a bit more and see if I can make it better 🙂

  

  Joachim Nilsson

  Participant

  Topics: 4

  Replies: 62

  Been thanked: 18 times
  01/19/2017 at 3:19 PM #21695
  jonjon

  All

  In the previous discussion I was asked to provide the results from Raul’s earlier code. The code has changed somewhat into the version above however I’ll paste the previous version here and attach the results. NOTE: the code below is an older version (I don’t want to cause confusion).

  I’ve made it quite conservative and have run it with SL of 4 (instead of 3 that I previously used) and Target 10. Max order size 11. I’m happy to forfeit % return to minimise drawdowns and err on side of caution. I’ve run backtest with tick-by-tick and the equity curve is more that you’d expect. In particular you can see “steps” up in the curve that we discussed, with slow gradual tail-offs through losses before the next “step”. 2013 clearly isn’t a good year (although not too bad). 2014+ good.

   

  // Definition of parameters DEFPARAM CumulateOrders = false // No accumulation of orders // The position is closed at 4:30 p.m., local market time (London). DEFPARAM FlatAfter =162900 once ordersize=1 // No new positions after the candle closes at 08:06 EndTime = 080600 // Market analysis starts in the candle 5 minutes closes at 08:05 StartTime = 080500 // Max orders if Ordersize>11 then Ordersize=11 endif // Risk & multiplier contracts n=1 // Conditions for analysis if Time >= StartTime and time <= EndTime then // Conditions to enter long positions c1 = open < close-1 IF c1 THEN IF PositionPerf(1) < 0 THEN OrderSize = OrderSize/2//+1 if ordersize<1 then ordersize=1 ENDIF ELSIF PositionPerf(1) > 0 THEN OrderSize =OrderSize+2 endif // If the first day of 5 bar min is positive, we buy buy ordersize*n shares at market endif // Conditions to enter short positions c2= open > close-1 IF c2 THEN iF PositionPerf(1) < 0 THEN OrderSize = OrderSize/2//+1 if ordersize<1 then ordersize=1 ENDIF ELSIF PositionPerf(1) > 0 THEN OrderSize =OrderSize+2 ENDIF // If the first day of 5 bar min is positive, we sell. sellshort ordersize*n shares at market endif endif SET STOP ploss 4 //5 3 SET TARGET pPROFIT 10 //10 12

  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

  // Definition of parameters DEFPARAM CumulateOrders = false // No accumulation of orders   // The position is closed at 4:30 p.m., local market time (London). DEFPARAM FlatAfter =162900   once ordersize=1 // No new positions after the candle closes at 08:06 EndTime = 080600 // Market analysis starts in the candle 5 minutes closes at 08:05 StartTime = 080500   // Max orders if Ordersize>11 then Ordersize=11 endif   // Risk & multiplier contracts   n=1   // Conditions for analysis   if Time >= StartTime and time <= EndTime then   // Conditions to enter long positions   c1 = open < close-1 IF c1 THEN IF PositionPerf(1) < 0 THEN OrderSize = OrderSize/2//+1 if ordersize<1 then ordersize=1 ENDIF ELSIF PositionPerf(1) > 0 THEN OrderSize =OrderSize+2 endif // If the first day of 5 bar min is positive, we buy buy ordersize*n shares at market endif   // Conditions to enter short positions   c2= open > close-1 IF c2 THEN iF PositionPerf(1) < 0 THEN OrderSize = OrderSize/2//+1 if ordersize<1 then ordersize=1 ENDIF ELSIF PositionPerf(1) > 0 THEN OrderSize =OrderSize+2 ENDIF // If the first day of 5 bar min is positive, we sell. sellshort ordersize*n shares at market endif endif   SET STOP ploss 4 //5 3 SET TARGET pPROFIT 10 //10 12

   

   

  

  jonjon

  Participant

  Topics: 5

  Replies: 64

  Been thanked: 4 times
  01/19/2017 at 3:26 PM #21699
  Alco

  Great work jonjon.

  But why do you use take profit instead of trailing? Are the results with tp better?

  

  Alco

  Participant

  Topics: 9

  Replies: 133

  Been thanked: 38 times
  01/19/2017 at 3:38 PM #21700
  jonjon

  No, it’s great work Raul. I just ran a few backtests and learned how to post the results 😉

  With regards trailing vs take profit: I haven’t had time to look into it. I ran the test on the latest code with trailing stop of 5. That’s it (see posts above). If I have time I’ll look into it. For now I’m more interested in the entry prices that I’m getting paper trading.

  

  jonjon

  Participant

  Topics: 5

  Replies: 64

  Been thanked: 4 times
  01/19/2017 at 4:33 PM #21707
  Nicolas

  Beware of stop trailing at 4 points, I don’t think that’s possible with IG. I encourage using the “now well known” trailing stop code snippet of the blog instead of the built-in trailing function, much easier to debug into real trading, IMO 🙂

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20863

  Been thanked: 4774 times
  01/19/2017 at 4:34 PM #21708
  CN

  Where do we find this code Nicolas?

  

  CN

  Participant

  Topics: 1

  Replies: 156

  Been thanked: 17 times
  01/19/2017 at 5:16 PM #21715
  Nicolas

  These are the 2 different trailing stop blog articles:

  MFE trailing stop

  complete trailing stop function

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20863

  Been thanked: 4774 times
  01/19/2017 at 5:28 PM #21717
  CN

  How would you add that/those to the code that Raul posted? And where?

  

  CN

  Participant

  Topics: 1

  Replies: 156

  Been thanked: 17 times
• Author
  Posts