Moving Average Slope 1Hr Strategy

• 10/26/2017 at 9:13 PM #50649
  juanj

  I coded the below indicator provided by @MAZ into a strategy for EURUSD 1Hr

  https://www.prorealcode.com/prorealtime-indicators/moving-average-slope/

  Defparam cumulateorders = False possize = 1 LMA = Average[200](close) MA = Average[26](Close) slope = MA - MA[1] If slope > 0 then once GS = slope AGS = (GS + slope)/2 ga = 0 gb = 0 For a = 0 to 29 Do If slope[a] > slope[a+1] Then ga = ga + 1 ElsIf slope[a] < slope[a+1] then gb = gb + 1 Endif Next GS = AGS ElsIf slope < 0 then once RS = slope ARS = (RS + slope)/2 ra = 0 rb = 0 For b = 0 to 29 Do If slope[b] < slope[b+1] Then ra = ra + 1 ElsIf slope[b] > slope[b+1] then rb = rb + 1 endif Next RS = ARS Endif If countofposition = 0 and close > LMA and slope > (AGS+(AGS*0.618)) Then Buy possize contract at market ElsIf countofposition = 0 and close < LMA and slope < (ARS+(ARS*0.618)) Then Sellshort possize contract at market EndIf If longonmarket and (slope < (AGS+(AGS*0.618)) or gb >= ga) Then Sell at market ElsIf shortonmarket and (slope > (ARS+(ARS*0.618)) or rb >= ra) 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

  Defparam cumulateorders = False   possize = 1   LMA = Average[200](close) MA = Average[26](Close) slope = MA - MA[1]   If slope > 0 then once GS = slope AGS = (GS + slope)/2 ga = 0 gb = 0 For a = 0 to 29 Do If slope[a] > slope[a+1] Then ga = ga + 1 ElsIf slope[a] < slope[a+1] then gb = gb + 1 Endif Next GS = AGS ElsIf slope < 0 then once RS = slope ARS = (RS + slope)/2 ra = 0 rb = 0 For b = 0 to 29 Do If slope[b] < slope[b+1] Then ra = ra + 1 ElsIf slope[b] > slope[b+1] then rb = rb + 1 endif Next RS = ARS Endif   If countofposition = 0 and close > LMA and slope > (AGS+(AGS*0.618)) Then Buy possize contract at market ElsIf countofposition = 0 and close < LMA and slope < (ARS+(ARS*0.618)) Then Sellshort possize contract at market EndIf   If longonmarket and (slope < (AGS+(AGS*0.618)) or gb >= ga) Then Sell at market ElsIf shortonmarket and (slope > (ARS+(ARS*0.618)) or rb >= ra) Then Exitshort at market EndIf

   

  4 users thanked author for this post.

  soad41, Nicolas, Berra, Henrik

  

  juanj

  Participant

  Topics: 51

  Replies: 505

  Been thanked: 170 times
  10/27/2017 at 12:56 AM #50659
  BC

  Thanks for share, will try it tonight.

  2 users thanked author for this post.

  soad41, Berra

  

  BC

  Participant

  Topics: 24

  Replies: 170

  Been thanked: 34 times
  11/06/2017 at 9:06 PM #51736
  Yannick

  Hello Juanj

  Thanks for sharing.

  Could you explain the concept of this ? ( I don’t understand the 0.618 value)

  slope > (AGS+(AGS*0.618))

  1	slope > (AGS+(AGS*0.618))

   

  Regards

  

  Yannick

  Participant

  Topics: 13

  Replies: 106

  Been thanked: 28 times
  11/07/2017 at 12:22 AM #51751
  BC

  0.618 is  a important number of Fibonnacci number sequence.

  https://en.m.wikipedia.org/wiki/Golden_ratio

   

  

  BC

  Participant

  Topics: 24

  Replies: 170

  Been thanked: 34 times
  11/07/2017 at 6:29 AM #51756
  Yannick

  Hi

  I Know that 0.618 is the golden number and the ratio of infinite convergence of Fibonnaci serie.

  it is used in trading for retracement of waves but why is it used in slopes of moving average ?

  regards

   

  

  Yannick

  Participant

  Topics: 13

  Replies: 106

  Been thanked: 28 times
  11/07/2017 at 8:21 AM #51761
  juanj

  @Yannick simple answer it works well. I use an optimization algorithm during backtesting that I graph to track optimal values during different points in the backtested period and found that ~0.6 is frequently the optimal value. I then just decided to make it 0.618 for the heck of it.

  

  juanj

  Participant

  Topics: 51

  Replies: 505

  Been thanked: 170 times
  11/08/2017 at 4:16 PM #51976
  JR1976

  HI JuanJ,

  good work for the code , seems interest

  Just a question ,  Did you test it on 200k  bar  ?

  Thanks

  

  JR1976

  Participant

  Topics: 16

  Replies: 88

  Been thanked: 8 times
  08/15/2019 at 2:52 PM #104800
  nfulcher

  Hi, very interesting post. Can somebody please help with my understanding as I’m new to the site.

  In the code above, it includes a For/Next Loop in which the ‘user variable’ slope is incremented through each pass of the loop.

  For a = 0 to 29 Do If slope[a] > slope[a+1] Then ga = ga + 1 ElsIf slope[a] < slope[a+1] then gb = gb + 1 Endif Next

  1 2 3 4 5 6 7

  For a = 0 to 29 Do If slope[a] > slope[a+1] Then ga = ga + 1 ElsIf slope[a] < slope[a+1] then gb = gb + 1 Endif Next

  Is this piece of code testing the change in the Moving Average (MA) of each of the last 30 (0 to 29) bars and incrementing ga and gb accordingly?

  If so, does it somehow pass the bar reference number (a) to the slope calculation at the start of the code further up (slope = MA – MA[1]) to calculate the step – or have I got this completely wrong?

  I appreciate any help to understand this better.

  Thanks.

  

  nfulcher

  Participant

  Topics: 1

  Replies: 17

  Been thanked: 0 times
  08/15/2019 at 3:03 PM #104801
  robertogozzi

  It is used to eventually exit LONG trades at line 43.

  1 user thanked author for this post.

  nfulcher

  

  robertogozzi

  Moderator

  Topics: 69

  Replies: 16517

  Been thanked: 3663 times
  08/15/2019 at 3:23 PM #104808
  nfulcher

  Thanks Roberto.

  I’m still a little confused with the coding – doesn’t the ‘slope = MA – MA[1]’ need to be nested within the For-Next loops – or does PRC know to look back up the code to find ‘slope’ calculation and recalculate it for each change in [a] and [b] (‘slope [a] > slope[a+1]’) ?

   

  

  nfulcher

  Participant

  Topics: 1

  Replies: 17

  Been thanked: 0 times
  08/15/2019 at 3:38 PM #104813
  robertogozzi

  Line 7 is the base to start with. It changes each new candle.

  The code later scans the past 30 (0 – 29) MA slopes to accomplish its goal.

  slope [a] > slope[a+1]

  1	slope [a] > slope[a+1]

  will be translated, at run time, with:

  (MA[a] - MA[a+1]) > (MA[a+1] - MA[a+2])

  1	(MA[a] - MA[a+1]) > (MA[a+1] - MA[a+2])

   

   

  

  robertogozzi

  Moderator

  Topics: 69

  Replies: 16517

  Been thanked: 3663 times
  08/15/2019 at 5:32 PM #104815
  nfulcher

  Thanks Roberto, makes perfect sense.

  

  nfulcher

  Participant

  Topics: 1

  Replies: 17

  Been thanked: 0 times
• Author
  Posts