Heikin Ashi Differential

• 04/04/2021 at 4:07 PM #166167
  carlvan

  Bonjour Nicolas,

  Je ne parviens pas à convertir cette formule AFL Amibroker (au demeurant très simple) en code AFL: je bute sur la formule AMA (adaptive MA) et je n’obtiens jamais le même résultat sur PRT.

  Merci d’avance pour votre aide – j’ai joint un chart Amibroker.

  Carl

  Voici le code AFL:

  HaClose=(O+H+L+C)/4;
  HaOpen=AMA(Ref(HaClose,-1),0.5); //Adaptive Moving Average du HaClose précédent, avec smoothing factor de 0.5
  HaDiffCO = (HaClose-HaOpen);

  HaClose2=(H+L+C)/3;
  HaOpen2=AMA(Ref(HaClose2,-1),0.1); //Adaptive Moving Average du HaClose2 précédent, avec smoothing factor de 0.1 cette fois
  //HaHigh2=Max(H,Max(HaClose2,HaOpen2));
  //HaLow2=Min(L,Min(HaClose2,HaOpen2));
  HaDiffCO2 = HaClose2-HaOpen2;

  HDCPeriods = 15;
  Width = Param(“Width”, 2, 0, 10, 0.05 );

  Plot( BBandTop( Hadiffco, HDCPeriods, Width ), “BBTop” + _PARAM_VALUES(), Colorblack, Styleline);
  Plot( BBandBot( Hadiffco, HDCPeriods, Width ), “BBBot” + _PARAM_VALUES(), Colorblack, Styleline);
  Plot(HaDiffCO,”HaDiffCO”,colorDarkRed,styleThick|styleLine);
  Plot(HaDiffCO2,”HaDiffCO2″,colorGreen,styleThick|styleLine);

  

  carlvan

  Participant

  Topics: 11

  Replies: 34

  Been thanked: 0 times
  04/05/2021 at 2:47 PM #166208
  Nicolas

  https://www.amibroker.com/guide/afl/ama.html

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20828

  Been thanked: 4738 times
  04/05/2021 at 2:55 PM #166209
  carlvan

  Oui, c’est bien cette formule AFL- là pour l’AMA, mais je ne parviens pas à la convertir en code PRT.

  Merci.

   

  

  carlvan

  Participant

  Topics: 11

  Replies: 34

  Been thanked: 0 times
  04/06/2021 at 9:03 AM #166227
  Nicolas

  I know I’ve been there before years ago, but I can’t find the code I made for at that time..

  Did you test those in this topic? https://www.prorealcode.com/topic/adaptive-moving-average-ama/

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20828

  Been thanked: 4738 times
  04/06/2021 at 12:00 PM #166265
  carlvan

  Yes I tested those but it does not give the same results as with the AFL code unfortunately. Also, there are those 3 variables in the current PRT formula (period, fast and slow) that has no equivalence to the smoothing factor used in the AFL code (0.5 for the HaDiffCo and 0.1 for HaDiffCo2).

  So what I did was trying to reproduce the same AMA based on the AFL explanation for the AMA function within Amibroker doc:

  output = AMA( input, factor )

  for( i = 1; i < BarCount; i++ ) { output[ i ] = factor[ i ] * input[ i ] + ( 1 – factor[ i ] ) * output[ i – 1 ]; }

  But I have never been good with looping formulas 🙂

   

   

   

   

  

  carlvan

  Participant

  Topics: 11

  Replies: 34

  Been thanked: 0 times
  04/06/2021 at 12:03 PM #166266
  carlvan

  Here is the closest I could get with my current PRT formula from the AFL code but again, not the same results on the chart:

  //Period = 9
  //FastPeriod = 1
  //SlowPeriod = 30
  //Period2=18
  //Fast2= 2
  //Slow2= 30

  haclose=(open+close+low+high)/4
  haopen= AdaptiveAverage[period,fast,slow](haclose)[1]

  haclose2=(close+low+high)/3
  haopen2= AdaptiveAverage[period2,fast2,slow2](haclose2)[1]

  HaDiffCO = (HaClose-HaOpen)
  HaDiffCO2 = HaClose2-HaOpen2
  //HaLow2=Min(L,Min(HaClose2,HaOpen2));
  sDev = std[15](hadiffco)
  avg = average[15](hadiffco)
  UpperBand = avg+(2*sDev)
  LowerBand = avg-(2*sDev)

  RETURN hadiffco as “HaDiffCo”, hadiffco2 as “HaDiffCo2”, 0 as “Zero”, upperband as “UpBB”, lowerband as “LoBB”

  

  carlvan

  Participant

  Topics: 11

  Replies: 34

  Been thanked: 0 times
• Author
  Posts