strategy BarHunter DAX v1p

• 01/21/2020 at 10:41 PM #117459
  GraHal

  We’ve had 0 / zero real trades up to now so it’s still early days.

  There’s been problems related to the variable minimum distance stop settings used by IG!

  1 user thanked author for this post.

  Francesco

  

  GraHal

  Participant

  Topics: 204

  Replies: 8401

  Been thanked: 2042 times
  01/22/2020 at 2:56 PM #117519
  fifi743

  Hello, Why not use a virtual stop

  

  fifi743

  Participant

  Topics: 9

  Replies: 851

  Been thanked: 401 times
  01/22/2020 at 3:16 PM #117520
  GraHal

  Go on … tell us what a virtual stop is? 🙂

  

  GraHal

  Participant

  Topics: 204

  Replies: 8401

  Been thanked: 2042 times
  01/22/2020 at 4:42 PM #117534
  Paul

  @fifi743 you got me curious too!

  @GraHal maybe we should do more testing, meaning more different barnumbers for entry1 to 4 so we can pick a entry-type which is reliable regardless of the bar-number and results.

  

  Paul

  Participant

  Topics: 40

  Replies: 812

  Been thanked: 660 times
  01/22/2020 at 5:13 PM #117539
  GraHal

  Paul wrote:

  so we can pick a entry-type which is reliable regardless of the bar-number and results.

  Yeah but as backtest does not perform same as Live (which is a nonsense in itself!?) then how can we prove what is reliable other than Live Forward Test … which would take for ever as so few trades?

  I guess if we don’t use stop entry then backtest results should be closer to Live results and so more reliable?

  If I recall correctly, my 5 min version is an at market entry  … hopefully there should be would be no Issues re minimum distance

  

  GraHal

  Participant

  Topics: 204

  Replies: 8401

  Been thanked: 2042 times
  01/22/2020 at 5:39 PM #117543
  fifi743

  // ================trailing atr stop VIRTUAL================== if enabletsvir then // once stepsvir=0 once minatrdistvir=0 once atrtrailingperiodvir = 2 // atr parameter once minstopvir = 10 // minimum distance if barindex=tradeindex then trailingstoplongvir = 5 // trailing stop atr distance trailingstopshortvir = 5 // trailing stop atr distance else if longonmarket then if newslvir>0 then if trailingstoplongvir>minatrdistvir then if newslvir>newslvir[1] then trailingstoplongvir=trailingstoplongvir else trailingstoplongvir=trailingstoplongvir-stepsvir endif else trailingstoplongvir=minatrdistvir endif endif endif if shortonmarket then if newslvir>0 then if trailingstopshortvir>minatrdistvir then if newslvir<newslvir[1] then trailingstopshortvir=trailingstopshortvir else trailingstopshortvir=trailingstopshortvir-stepsvir endif else trailingstopshortvir=minatrdistvir endif endif endif endif // atrtrailvir=averagetruerange[atrtrailingperiodvir]((close/10)*pipsize)/1000 trailingstartlvir=round(atrtrailvir*trailingstoplongvir) trailingstartsvir=round(atrtrailvir*trailingstopshortvir) tglvir=trailingstartlvir tgsvir=trailingstartsvir // if not onmarket or ((longonmarket and shortonmarket[1]) or (longonmarket[1] and shortonmarket)) then maxpricevir=0 minpricevir=close newslvir=0 endif // if longonmarket then maxpricevir=max(maxpricevir,close) if maxpricevir-tradeprice(1)>=tglvir*pointsize then if maxpricevir-tradeprice(1)>=minstopvir then newslvir=maxpricevir-tglvir*pointsize else newslvir=maxpricevir-minstopvir*pointsize endif endif endif // if shortonmarket then minpricevir=min(minpricevir,close) if tradeprice(1)-minpricevir>=tgsvir*pointsize then if tradeprice(1)-minpricevir>=minstopvir then newslvir=minpricevir+tgsvir*pointsize else newslvir=minpricevir+minstopvir*pointsize endif endif endif // if longonmarket and close <newslvir and newslvir>0 then sell at market endif if shortonmarket and close>newslvir and newslvir>0 then exitshort at market endif

  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 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85

  // ================trailing atr stop VIRTUAL==================   if enabletsvir then // once stepsvir=0 once minatrdistvir=0   once atrtrailingperiodvir    = 2  // atr parameter once minstopvir              = 10   // minimum  distance   if barindex=tradeindex then trailingstoplongvir    = 5  // trailing stop atr distance trailingstopshortvir    = 5   // trailing stop atr distance else if longonmarket then if newslvir>0 then if trailingstoplongvir>minatrdistvir then if newslvir>newslvir[1] then trailingstoplongvir=trailingstoplongvir else trailingstoplongvir=trailingstoplongvir-stepsvir endif else trailingstoplongvir=minatrdistvir endif endif endif   if shortonmarket then if newslvir>0 then if trailingstopshortvir>minatrdistvir then if newslvir<newslvir[1] then trailingstopshortvir=trailingstopshortvir else trailingstopshortvir=trailingstopshortvir-stepsvir endif else trailingstopshortvir=minatrdistvir endif endif endif endif // atrtrailvir=averagetruerange[atrtrailingperiodvir]((close/10)*pipsize)/1000 trailingstartlvir=round(atrtrailvir*trailingstoplongvir) trailingstartsvir=round(atrtrailvir*trailingstopshortvir) tglvir=trailingstartlvir tgsvir=trailingstartsvir // if not onmarket or ((longonmarket and shortonmarket[1]) or (longonmarket[1] and shortonmarket)) then maxpricevir=0 minpricevir=close newslvir=0 endif // if longonmarket then maxpricevir=max(maxpricevir,close) if maxpricevir-tradeprice(1)>=tglvir*pointsize then if maxpricevir-tradeprice(1)>=minstopvir then newslvir=maxpricevir-tglvir*pointsize else newslvir=maxpricevir-minstopvir*pointsize endif endif endif // if shortonmarket then minpricevir=min(minpricevir,close) if tradeprice(1)-minpricevir>=tgsvir*pointsize then if tradeprice(1)-minpricevir>=minstopvir then newslvir=minpricevir+tgsvir*pointsize else newslvir=minpricevir+minstopvir*pointsize endif endif endif // if longonmarket and close <newslvir and newslvir>0 then sell at market endif if shortonmarket and close>newslvir and newslvir>0 then   exitshort at market   endif

  I have two trailing stops Whoever moves the stop And the other if it is closed is above or below the newslvir

  3 users thanked author for this post.

  Francesco, GraHal, Paul

  

  fifi743

  Participant

  Topics: 9

  Replies: 851

  Been thanked: 401 times
  01/22/2020 at 5:55 PM #117546
  GraHal

  Above added as Log 196 here …

  Snippet Link Library

  1 user thanked author for this post.

  fifi743

  

  GraHal

  Participant

  Topics: 204

  Replies: 8401

  Been thanked: 2042 times
  01/22/2020 at 8:46 PM #117565
  fifi743

  Backtest file edit

  3 users thanked author for this post.

  winnie37, Paul, GraHal

  

  fifi743

  Participant

  Topics: 9

  Replies: 851

  Been thanked: 401 times
  01/22/2020 at 9:07 PM #117569
  Paul

  Thanks fifi! The virtual stop is implemented as I thought. The original ts still running with atrtrailingperiod=14 and the virtual with value 2. Also the test de nombre de bar negative ajouter makes a difference. Total results is a bit the same, but the equitycurve has a nicer shape!

  I’am working on something to drop the stop for entry. But it’s not there yet.

  2 users thanked author for this post.

  winnie37, GraHal

  

  Paul

  Participant

  Topics: 40

  Replies: 812

  Been thanked: 660 times
  01/23/2020 at 4:47 PM #117627
  Paul

  @fifi, with importing I noticed tick by tick wasn’t enabled, so I did the test again and with target profit disabled.

  your average loss dropped quite a bit. Will test to to see what difference it brings to the market version.

  Unfortunately I don’t understand the benefit of the virtual trailing stop and see very little difference. Maybe it shows more with live trading.

   

   

  

  Paul

  Participant

  Topics: 40

  Replies: 812

  Been thanked: 660 times
  01/23/2020 at 7:09 PM #117637
  fifi743

  Good-morning Paul,
  To avoid errors of stops too close.
  The value remains in the variable
  There may be more difference in lower time frame.

  1 user thanked author for this post.

  Paul

  

  fifi743

  Participant

  Topics: 9

  Replies: 851

  Been thanked: 401 times
  01/25/2020 at 4:23 PM #117810
  fifi743

  Hello,
  Entry type = 1 I reduced the hour to 21
  I have placed two different stops depending on the trading hours.
  To be checked so as not to have an error
  Today I’m at 271 points for Stop

  2 users thanked author for this post.

  winnie37, Paul

  

  fifi743

  Participant

  Topics: 9

  Replies: 851

  Been thanked: 401 times
  01/26/2020 at 12:26 AM #117845
  Paul

  Hi Fifi43

  Thanks I will look into this. I found irregularities when using barindex numbers. Because when there is no data, everything shifts a bar! Doesn’t happen too much on 1 hour timeframe, but it could be more often on lower timeframes and I want it to be universal.

  So I’m testing a way too quickly and efficiently optimise time using hour & any minutes. Also looking in your nicely coded FILTRE_Prise_Position if it could be of use for profittaking.

   

  

  Paul

  Participant

  Topics: 40

  Replies: 812

  Been thanked: 660 times
  01/26/2020 at 7:10 AM #117852
  jmf125

  @Grahal

  Did you figure out the exact reason of the difference ? As I am also getting a different equity curve and number of trades. I believe it is the timezone. I only 100 K bars to test, so I can’t compare exactly to the latest from fifi. But if I change the timezone to be inline with the others, I get exactly the same drawdown (869.80) and similar profit profile per month.

  Optimizing barnumber might not give you the same result.

  

  jmf125

  Participant

  Topics: 8

  Replies: 111

  Been thanked: 14 times
  01/26/2020 at 7:18 AM #117853
  jmf125

  @Grahal : Disregard my previous comment. It will give you exactly the same result.

  

  jmf125

  Participant

  Topics: 8

  Replies: 111

  Been thanked: 14 times
• Author
  Posts