How would you create a Mov Ave Distribution Histogram?

• 03/03/2019 at 4:09 PM #92759
  Bard

  Hi

  I was reading Shay Campbells’ E-Mini Swing Trading System: http://timingcharts.com/articles/2017/10/20/e-mini-sp-500-swing-trading-system/
  where he uses a technique to enter the market based on pull backs of the price based upon a Moving Average Distribution indicator: Pls see image.

  “Using a very short moving average of median price does a good job of defining a usable equilibrium point.  A distribution can then be calculated around that equilibrium by subtracting the equilibrium point from the market price.

  To normalize these readings for volatility, this result is then divided by the recent daily range of price.

  Buy Signal=  Trend Direction is Up (6 to 9 month Mov Ave) and the MA Distribution registers a significant Pullback, i.e, negative value.

  Sell Signal=  Trend Direction is Down and the MA Distribution registers a significant Pullback, i.e, positive value.”

  I was attempting to re-create that indicator but wondered how as I’m not a coder. My attempt based on other indicators is not working but I hope I’m thinking along the right lines.. unless I have over complicated it?

  Cheers

  Moving Average Distribution

  DEFPARAM CalculateOnLastBars = 100 Period = 20 MAPeriod = 20 ATRPeriod = 20 r = 0 g =0 Ave = Average[MAPeriod](MedianPrice) ATR = AverageTrueRange[ATRPeriod](Close) S = 0 FOR i = 1 TO Period TP = TotalPrice[i] Dist = (TP - Ave)/ATR > 0 // 1 = TP > Ave IF Dist = 0 THEN Dist = (TP - Ave)/ATR < 0 * -1 // -1 = TP < Ave ENDIF S = S + Dist //Sum up all Dist’s NEXT Dist = Dist*(s) //Histogram Ave = Average[MAPeriod](MedianPrice) //Signal Line Return Dist coloured(r,g,0) style(histogram) as "Moving Average Distribution", Ave AS "Signal Line", 0 AS "Zero"

  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

  DEFPARAM CalculateOnLastBars = 100 Period = 20 MAPeriod = 20 ATRPeriod = 20 r = 0 g =0   Ave = Average[MAPeriod](MedianPrice) ATR =  AverageTrueRange[ATRPeriod](Close)   S = 0 FOR i = 1 TO Period TP = TotalPrice[i] Dist = (TP - Ave)/ATR > 0      //  1 = TP > Ave   IF Dist = 0 THEN Dist = (TP - Ave)/ATR < 0 * -1 // -1 = TP < Ave ENDIF S = S + Dist //Sum up all Dist’s   NEXT Dist = Dist*(s) //Histogram Ave  = Average[MAPeriod](MedianPrice) //Signal Line     Return Dist coloured(r,g,0) style(histogram) as "Moving Average Distribution", Ave AS "Signal Line", 0 AS "Zero"

   

  

  Bard

  Participant

  Topics: 138

  Replies: 814

  Been thanked: 99 times
  03/04/2019 at 9:49 AM #92798
  Nicolas

  Bard wrote:

  this result is then divided by the recent daily range of price.

  daily range is not the ATR, it is called the ADR (an average of the last X days range). “Recent” would mean how many period?

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20863

  Been thanked: 4774 times
  03/04/2019 at 6:47 PM #92840
  Bard

  Oh, right, thanks. In the article Shay doesn’t specify exactly other than to say recent. It would make sense to keep the ADR in relation to the moving average for the distribution, no? He writes:

  The MA Distribution tells us what’s happening in the very short term using just a couple weeks of data.

  So I guess 20 days?

  

  Bard

  Participant

  Topics: 138

  Replies: 814

  Been thanked: 99 times
  03/05/2019 at 10:10 AM #92863
  Nicolas

  Ok, I took this value in the below code, you can change it of course. The indicator makes sense since it finds easily the pullbacks when you are in a well established trend.

  //PRC_MovingAverage Distribution | indicator //05.03.2019 //Nicolas @ www.prorealcode.com //Sharing ProRealTime knowledge // --- settings ShortMAPeriod = 20 ADRperiod = 20 // --- end of settings //compute ADR sum=0 for i = 0 to ADRperiod-1 do sum=sum+(dhigh(i)-dlow(i)) next adr=sum/ADRperiod shortMA=average[ShortMAPeriod](medianprice) diff=medianprice-shortMA MAD=(diff*100)/adr return MAD style(histogram)

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

  //PRC_MovingAverage Distribution | indicator //05.03.2019 //Nicolas @ www.prorealcode.com //Sharing ProRealTime knowledge   // --- settings ShortMAPeriod = 20 ADRperiod = 20 // --- end of settings   //compute ADR sum=0 for i = 0 to ADRperiod-1 do sum=sum+(dhigh(i)-dlow(i)) next adr=sum/ADRperiod   shortMA=average[ShortMAPeriod](medianprice) diff=medianprice-shortMA   MAD=(diff*100)/adr   return MAD style(histogram)

  If you find it correct according to the original description, I will put it into the library for everyone to benefit from this concept.

  1 user thanked author for this post.

  Bard

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20863

  Been thanked: 4774 times
  03/05/2019 at 2:37 PM #92887
  Bard

  https://www.youtube.com/watch?v=eZDiG2jynNs

  Yes, please do.

  

  Bard

  Participant

  Topics: 138

  Replies: 814

  Been thanked: 99 times
  03/05/2019 at 3:31 PM #92891
  Nicolas

  😆

  The indicator is now available in the library: Moving Average Distribution

  1 user thanked author for this post.

  Bard

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20863

  Been thanked: 4774 times
• Author
  Posts