Strat – Bol break with vol filtr DAX.H1 [Seemore]

• 06/05/2017 at 8:56 AM #37419
  Maz

  RE: https://www.prorealcode.com/prorealtime-trading-strategies/dax-bollinger-break-v

  “Dax Bollinger break with volume. 1 Hour”

  Just opening a forum topic for this breakout strategy idea posted by Seemore Profit. I’m a bit short on time but wanted to post a quick optimization. Also I have a few more points about this.

  • Not really a fan of hard coded vars – volume threshold should be dynamic
  • Bollenger periods also hard coded; could be dynamic
  • Needs some proper money-management code. Hard coded basis point stop and limit are not allowing for shifts in volatility. So add ADX / FDI filters and such like

  Anyway point being I think the overall idea is cool and there’s much further to go on this 🙂

  Have a great week!

  Best,

  Maz

   

  Super quick clean up

  // System Settings DEFPARAM CumulateOrders = False DEFPARAM FLATBEFORE = 070000 DEFPARAM FLATAFTER = 181500 // Old Variables //positionSize = 1 //volumeThreshold = 8000 //pointTarget = 60 //pointStop = 30 //bolUpPeriod = 20 //bolDnPeriod = 20 // New variables positionSize = 1 volumeThreshold = 12500 pointTarget = 53 // must make this dynamic pointStop = 30 // must make this dynamic bolUpPeriod = 20 // not optimized yet bolDnPeriod = 20 // not optimized yet // Indicators bolUp = bollingerUp[bolUpPeriod](close) bolDn = bollingerDown[bolDnPeriod](close) volumeFilter = ( Volume >= volumeThreshold ) tradingDay = not (OpenDayOfWeek = 6 OR OpenDayOfWeek = 0) // Entry logic // long side bc1 = (close > bolUp) and volumeFilter bc1 = bc1 and tradingDay // short side sc1 = (close < bolDn) and volumeFilter sc1 = sc1 and tradingDay // Execution IF bc1 then BUY positionSize PERPOINT AT MARKET elsif sc1 THEN SELLSHORT positionSize PERPOINT AT MARKET ENDIF // Stops and targets SET STOP pLOSS pointStop SET TARGET pPROFIT pointTarget

  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

  // System Settings DEFPARAM CumulateOrders = False DEFPARAM FLATBEFORE = 070000 DEFPARAM FLATAFTER = 181500     // Old Variables //positionSize = 1 //volumeThreshold = 8000 //pointTarget = 60 //pointStop = 30 //bolUpPeriod = 20 //bolDnPeriod = 20   // New variables positionSize = 1 volumeThreshold = 12500 pointTarget = 53 // must make this dynamic pointStop = 30   // must make this dynamic bolUpPeriod = 20 // not optimized yet bolDnPeriod = 20 // not optimized yet     // Indicators bolUp = bollingerUp[bolUpPeriod](close) bolDn = bollingerDown[bolDnPeriod](close) volumeFilter = ( Volume >= volumeThreshold ) tradingDay = not (OpenDayOfWeek = 6 OR OpenDayOfWeek = 0)       // Entry logic   // long side bc1 = (close > bolUp) and volumeFilter bc1 = bc1 and tradingDay   // short side sc1 = (close < bolDn) and volumeFilter sc1 = sc1 and tradingDay       // Execution IF bc1 then BUY positionSize PERPOINT AT MARKET elsif sc1 THEN SELLSHORT positionSize PERPOINT AT MARKET ENDIF   // Stops and targets SET STOP pLOSS pointStop SET TARGET pPROFIT pointTarget

  Let’s see some further development 🙂

  4 users thanked author for this post.

  Daniel Rönnqvist, Francesco78, BC, Nicolas

  

  Maz

  Participant

  Topics: 14

  Replies: 121

  Been thanked: 139 times
  06/05/2017 at 10:53 AM #37463
  Francesco78

  Here is what I get
  // New variables

  atrperiod = 10
  atr = averagetruerange[atrperiod]
  positionSize = 1
  volumeThreshold = 12500
  pointTarget = 3*atr
  pointStop = 7*atr
  bolUpPeriod = 20
  bolDnPeriod = 32

  

  Francesco78

  Participant

  Topics: 43

  Replies: 242

  Been thanked: 77 times
  06/05/2017 at 2:38 PM #37485
  Seemore Profit

  Hi all thanks for the feedback.

  Maz i like to use hard stops as i feel it lowers the risk of blow outs and gives a nice balance of risk to reward. I agree that something could be used to improve this is someway whilst keeping the balance in place. Maybe a reversal signal on a shorter time frame might work?

  Francesco I added your variables, but didnt seem to improve the results in a back test. What were your results?

  

  Seemore Profit

  Participant

  Topics: 5

  Replies: 21

  Been thanked: 7 times
  06/05/2017 at 4:26 PM #37487
  Francesco78

  HI Seemore, yes pls find it in the attachment.

   

  1 user thanked author for this post.

  Seemore Profit

  

  Francesco78

  Participant

  Topics: 43

  Replies: 242

  Been thanked: 77 times
  06/05/2017 at 6:42 PM #37499
  JR1976

  Dear All,

  nice work , but why  the backtest start only since August 2016  and not before ??

   

  Regards

  

  JR1976

  Participant

  Topics: 16

  Replies: 88

  Been thanked: 8 times
  06/05/2017 at 6:51 PM #37503
  Francesco78

  Volume data are lacking before then.

  I wonder if with the premium account they could be retrieved.

   

  2 users thanked author for this post.

  Seemore Profit, JR1976

  

  Francesco78

  Participant

  Topics: 43

  Replies: 242

  Been thanked: 77 times
  06/06/2017 at 8:41 PM #37660
  2Cent

  Hi Francesco,

  with premium account I get the same limitation for volume data. Volume data starts at monday, Aug 29, 2016.

  So I see no possibility to backtest this volume based strategy for a longer period than aprox 9 months.

  BR Cent

  

  2Cent

  Participant

  Topics: 1

  Replies: 2

  Been thanked: 0 times
  06/20/2017 at 2:39 PM #38654
  David

  Due to the lack of volume data I looked at building this in MT4.

  The problem is that the volume data is different.

  I spoke to IG and a prop trader about the different data and I was informed that indices volume data varies and in the case of IG it represents their client pool volume.

  1 user thanked author for this post.

  Seemore Profit

  

  David

  Participant

  Topics: 24

  Replies: 170

  Been thanked: 31 times
  06/20/2017 at 2:51 PM #38656
  Nicolas

  David wrote:

  in the case of IG it represents their client pool volume.

  and this is exactly the same with any other MT4 brokers for retail traders.

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20872

  Been thanked: 4797 times
  07/06/2017 at 11:48 AM #39948
  Seemore Profit

  David wrote:

  Due to the lack of volume data I looked at building this in MT4. The problem is that the volume data is different. I spoke to IG and a prop trader about the different data and I was informed that indices volume data varies and in the case of IG it represents their client pool volume.

  Thanks For finding this out, was wondering about this.

  My final strat from the original saw the best vol set at 7000 and the target price changed to 58.

  Gave these results, anyone find anything else worth adding/changing?

  

  Seemore Profit

  Participant

  Topics: 5

  Replies: 21

  Been thanked: 7 times
  07/06/2017 at 1:51 PM #39954
  JR1976

  Dear Seemore profit,.

  In your screenshot the backtest start from 2013 , is it by IG/prorealtime platform ? Because I understood that the volume for IG is only since August 2016

  Thanks

  

  JR1976

  Participant

  Topics: 16

  Replies: 88

  Been thanked: 8 times
  07/07/2017 at 3:41 PM #40034
  Seemore Profit

  That is correct JR the results are only from trades starting from August 16.

  I guess it has no trades before then because of the volume data.

  1 user thanked author for this post.

  JR1976

  

  Seemore Profit

  Participant

  Topics: 5

  Replies: 21

  Been thanked: 7 times
  07/23/2017 at 2:19 PM #41412
  Yannick

  Hello

  For volumeThreshold, it is possible to calculate an average of last volume data , in case of evolution of volume of trading of the intrument.

  There is no volume during night, so volume could be filled with last volume value to prevent a zero in the average.

   

  Volume filter

  If volume>2000 then myvolume= volume volumeaverage=average(myvolume)[50] Endif volumeThreshold = x*volumeaverage//12500 x=2 gives good results in backtest

  1 2 3 4 5 6

  If volume>2000 then myvolume= volume volumeaverage=average(myvolume)[50] Endif   volumeThreshold = x*volumeaverage//12500 x=2 gives good results in backtest

  Backtest with x= 2 gives good results

  Real robot get 3 consecutives winning trades

   

   

  2 users thanked author for this post.

  JR1976, Seemore Profit

  

  Yannick

  Participant

  Topics: 13

  Replies: 106

  Been thanked: 28 times
  11/23/2017 at 8:55 PM #53769
  Yannick

  Hi I have found this article for similar strategy adapted to FX , maybe on other commodities for diversification

  https://blog.quandl.com/bollinger-band-breakout-augmented-volume  

  The strategy is making some money according to backtest.

  Please find live results for people who would be interested in this strategy.

  Even if the DAX is in a bull market, the short side wins more than long side due to the better ratio Earning/loss. This could be due to the fact that indices grows slowly but falls sharply in fear mode. I will try to find “safe heavens money flow” related to fear mode , maybe  Gold, Yen, BTC

  1 user thanked author for this post.

  Despair

  

  Yannick

  Participant

  Topics: 13

  Replies: 106

  Been thanked: 28 times
• Author
  Posts