Adaptive Moving Averag

• 04/13/2023 at 9:03 PM #213253
  Markscragg

  I am trying to code a variation of Perry Kaufmans Adaptive Moving Average

  The Equation is AMA= (ER*Close) + (1-ER)*AMA[1]

  It Returns ‘Blank’

  Can anyone help?

   

  // parameter // n = 10 Change = ABS(close - close[n]) calc = ABS(close-close[1]) volat = summation[n](calc) ER = Change / volat once AMA=Average[n] AMA=(ER*Close)+(1-ER)*AMA[1] //or //V1=(1-ER) //V2=AMA[1] //V3=(V1*V2) //AMA=(ER*Close)+V3 Return AMA

  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23

  // parameter // n = 10   Change = ABS(close - close[n])   calc = ABS(close-close[1])   volat = summation[n](calc)   ER = Change / volat   once AMA=Average[n]   AMA=(ER*Close)+(1-ER)*AMA[1]   //or   //V1=(1-ER) //V2=AMA[1] //V3=(V1*V2) //AMA=(ER*Close)+V3   Return AMA

   

   

  

  Markscragg

  Participant

  Topics: 5

  Replies: 4

  Been thanked: 0 times
  04/13/2023 at 10:00 PM #213254
  JS

  Hi,

  This is how I know the Kaufman filter…

  KAMA

  Period = MAX(10, 4) FastPeriod = 7 SlowPeriod = 28 Fastest = 2 / (FastPeriod + 1) Slowest = 2 / (SlowPeriod + 1) IF BarIndex < Period THEN KAMA = Close ELSE Num = ABS(Close - Close[Period]) Den = summation[Period](ABS(Close - Close[1])) ER = Num / Den Alpha = SQUARE(ER * (Fastest - Slowest) + Slowest) KAMA = (Alpha * Close) + ((1 - Alpha) * KAMA[1]) ENDIF Return KAMA

  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

  Period = MAX(10, 4)   FastPeriod = 7 SlowPeriod = 28   Fastest = 2 / (FastPeriod + 1) Slowest = 2 / (SlowPeriod + 1)   IF BarIndex < Period THEN KAMA = Close ELSE Num = ABS(Close - Close[Period]) Den = summation[Period](ABS(Close - Close[1])) ER = Num / Den Alpha = SQUARE(ER * (Fastest - Slowest) + Slowest) KAMA = (Alpha * Close) + ((1 - Alpha) * KAMA[1]) ENDIF Return KAMA

  

  JS

  Participant

  Topics: 22

  Replies: 1423

  Been thanked: 557 times
  04/14/2023 at 9:40 AM #213279
  Markscragg

  JS

  Thanks for the quick reply.  However I was trying to code the following equation

  The Equation is AMA= (ER*Close) + (1-ER)*AMA[1]

  Where ER is the smoothing constantent dirived from  Perry Kaufman

  I think the problem lies in the second part of the equation!

  Regards

   

  

  Markscragg

  Participant

  Topics: 5

  Replies: 4

  Been thanked: 0 times
  04/14/2023 at 1:55 PM #213293
  Nicolas

  Needs at least n data in order to calculate, so embed your code into a barindex count condition:

  // parameter n = 10 if barindex>n then Change = ABS(close - close[n]) calc = ABS(close-close[1]) volat = summation[n](calc) ER = Change / volat once AMA=Average[n] AMA=(ER*Close)+(1-ER)*AMA[1] endif Return AMA

  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

  // parameter n = 10   if barindex>n then Change = ABS(close - close[n])   calc = ABS(close-close[1])   volat = summation[n](calc)   ER = Change / volat   once AMA=Average[n]   AMA=(ER*Close)+(1-ER)*AMA[1] endif   Return AMA

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20828

  Been thanked: 4742 times
  04/17/2023 at 9:59 AM #213394
  Markscragg

  Thank You Nicolas

  That seems to work

  I have been learning prorealcode for sometime now.  Could you recommend any text that may help me

  I have studied the online manuels plus the Vivian Schmitt ebook but still need more help.

  Regards

  Mark

   

   

  

  Markscragg

  Participant

  Topics: 5

  Replies: 4

  Been thanked: 0 times
• Author
  Posts