new John Ehlers indicator Forward Reverse EMA

• 06/27/2017 at 11:14 AM #39171
  Francesco78

  Hello, from the last video of better system trader I discovered this new indicator from John Ehlers

  He gave the code in Easylanguage that you can find here

  http://downloads.bettersystemtrader.com/files/ForwardReverseEMA.txt

  {
  Forward / Reverse EMA
  (c) 2017 John F. Ehlers
  }

  Inputs:
  AA(.1);

  Vars:
  CC(.9),
  RE1(0),
  RE2(0),
  RE3(0),
  RE4(0),
  RE5(0),
  RE6(0),
  RE7(0),
  RE8(0),
  EMA(0),
  Signal(0);

  CC = 1 – AA;
  EMA = AA*Close + CC*EMA[1];

  RE1 = CC*EMA + EMA[1];
  RE2 = Power(CC, 2)*RE1 + RE1[1];
  RE3 = Power(CC, 4)*RE2 + RE2[1];
  RE4 = Power(CC, 8)*RE3 + RE3[1];
  RE5 = Power(CC, 16)*RE4 + RE4[1];
  RE6 = Power(CC, 32)*RE5 + RE5[1];
  RE7 = Power(CC, 64)*RE6 + RE6[1];
  RE8 = Power(CC, 128)*RE7 + RE7[1];

  Signal = EMA – AA*RE8;

  Plot1(Signal);
  Plot2(0);

   

   

  I translated like this for prt, can you please tell me if it is ok?

   

  AA= 0.1 CC = 1 - AA if barindex > 1 then EMA = AA*Close + CC*EMA[1] RE1 = CC*EMA + EMA[1] RE2 = EXP(2*LOG(CC))*RE1 + RE1[1] RE3 = EXP(4*LOG(CC))*RE2 + RE2[1] RE4 = EXP(8*LOG(CC))*RE3 + RE3[1] RE5 = EXP(16*LOG(CC))*RE4 + RE4[1] RE6 = EXP(32*LOG(CC))*RE5 + RE5[1] RE7 = EXP(64*LOG(CC))*RE6 + RE6[1] RE8 = EXP(128*LOG(CC))*RE7 + RE7[1] Signal = EMA - AA*RE8 endif return Signal

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

  AA= 0.1 CC = 1 - AA   if barindex > 1 then EMA = AA*Close + CC*EMA[1]       RE1 = CC*EMA + EMA[1] RE2 = EXP(2*LOG(CC))*RE1 + RE1[1] RE3 = EXP(4*LOG(CC))*RE2 + RE2[1] RE4 = EXP(8*LOG(CC))*RE3 + RE3[1] RE5 = EXP(16*LOG(CC))*RE4 + RE4[1] RE6 = EXP(32*LOG(CC))*RE5 + RE5[1] RE7 = EXP(64*LOG(CC))*RE6 + RE6[1] RE8 = EXP(128*LOG(CC))*RE7 + RE7[1] Signal = EMA - AA*RE8 endif return Signal

   

  

  Francesco78

  Participant

  Topics: 43

  Replies: 242

  Been thanked: 77 times
  06/27/2017 at 12:20 PM #39182
  Nicolas

  Your code is correct to me. You should post it into the library. Haven’t heard of this new John Ehlers’ formula before your post, thank you.

  1 user thanked author for this post.

  Francesco78

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20863

  Been thanked: 4773 times
  06/27/2017 at 12:22 PM #39184
  Francesco78

  Thank you Nicolas, only concern is that is not really an oscilator as is not bounded between 2 values. I am a bit puzzled..

  

  Francesco78

  Participant

  Topics: 43

  Replies: 242

  Been thanked: 77 times
  06/27/2017 at 12:28 PM #39187
  Nicolas

  Just like an MACD, who cares? 😉  There is a “zero” level, but I understand your concern .. how to use it effectively? Maybe the answers are in the Ehlers’ interview?

  1 user thanked author for this post.

  Francesco78

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20863

  Been thanked: 4773 times
  06/27/2017 at 12:33 PM #39188
  Francesco78

  in the interview he just said to use it as a mean reverting tool, without further specification. He is postponing the explanation to the people attending hismaster class understandably 🙂

  

  Francesco78

  Participant

  Topics: 43

  Replies: 242

  Been thanked: 77 times
  06/27/2017 at 12:50 PM #39195
  Francesco78

  Submitted to the library

  1 user thanked author for this post.

  Nicolas

  

  Francesco78

  Participant

  Topics: 43

  Replies: 242

  Been thanked: 77 times
  06/27/2017 at 1:19 PM #39208
  Nicolas

  Ok, so let’s use STANDARD DEVIATION like in these indicators, it would complete the analysis because as you stated this indicator is not bounded and doesn’t have any overbought/oversold areas:

  https://www.prorealcode.com/prorealtime-indicators/dynamic-zone-stoch-rsi/

  https://www.prorealcode.com/prorealtime-indicators/tti-atr-extreme/

  etc. Whatever examples using deviation of a normal distribution around a dynamic mean.

  1 user thanked author for this post.

  Francesco78

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20863

  Been thanked: 4773 times
  06/27/2017 at 1:26 PM #39210
  Francesco78

  Great idea, will do.

  

  Francesco78

  Participant

  Topics: 43

  Replies: 242

  Been thanked: 77 times
  06/27/2017 at 3:44 PM #39229
  ALE

  AA= 0.1 CC = 1 - AA if barindex > 1 then EMA = AA*Close + CC*EMA[1] RE1 = CC*EMA + EMA[1] RE2 = EXP(2*LOG(CC))*RE1 + RE1[1] RE3 = EXP(4*LOG(CC))*RE2 + RE2[1] RE4 = EXP(8*LOG(CC))*RE3 + RE3[1] RE5 = EXP(16*LOG(CC))*RE4 + RE4[1] RE6 = EXP(32*LOG(CC))*RE5 + RE5[1] RE7 = EXP(64*LOG(CC))*RE6 + RE6[1] RE8 = EXP(128*LOG(CC))*RE7 + RE7[1] result = EMA - AA*RE8 endif r = 50+(100-(result*10000))*2 g = 50+(100+(result*10000))*2 RETURN result COLOURED(r,g,50)style(histogram) as "result", 0 as "0"

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

  AA= 0.1 CC = 1 - AA if barindex > 1 then EMA = AA*Close + CC*EMA[1] RE1 = CC*EMA + EMA[1] RE2 = EXP(2*LOG(CC))*RE1 + RE1[1] RE3 = EXP(4*LOG(CC))*RE2 + RE2[1] RE4 = EXP(8*LOG(CC))*RE3 + RE3[1] RE5 = EXP(16*LOG(CC))*RE4 + RE4[1] RE6 = EXP(32*LOG(CC))*RE5 + RE5[1] RE7 = EXP(64*LOG(CC))*RE6 + RE6[1] RE8 = EXP(128*LOG(CC))*RE7 + RE7[1] result = EMA - AA*RE8 endif r = 50+(100-(result*10000))*2 g = 50+(100+(result*10000))*2   RETURN result COLOURED(r,g,50)style(histogram) as "result", 0 as "0"

  Nice colours!

  

  ALE

  Moderator

  Topics: 17

  Replies: 569

  Been thanked: 223 times
  06/28/2017 at 3:49 PM #39328
  zilliq

  Hi, thanks, I didn’t know this new indicator from Ehlers

  For fun, we see it’s a “suite” and I tried to replace the re2…re8 by an instruction for /next who will me more adjustable

  I don’t know why this suite doesn’t work, if someone as an idea

  // variable val=5 AA= 0.1 CC = 1 - AA if barindex > 11 then EMA = AA*close + CC*EMA[1] RE1 = CC*EMA + EMA[1] once result=0 for n=1 to val m=1 for nn=1 to val m=2*m next coef=EXP(2*m*LOG(CC)) RE1 = coef*RE1[n] + RE1[n+1] next result = EMA - AA*RE1 endif RETURN result

  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

  // variable val=5   AA= 0.1 CC = 1 - AA   if barindex > 11 then   EMA = AA*close + CC*EMA[1] RE1 = CC*EMA + EMA[1]   once result=0   for n=1 to val   m=1 for nn=1 to val m=2*m next   coef=EXP(2*m*LOG(CC))   RE1 = coef*RE1[n] + RE1[n+1]   next   result = EMA - AA*RE1   endif     RETURN result

   

  1 user thanked author for this post.

  Francesco78

  

  zilliq

  Participant

  Topics: 58

  Replies: 683

  Been thanked: 152 times
  06/29/2017 at 8:16 AM #39345
  Nicolas

  Seems to work fine for me Zilliq, what’s the problem please?

  
  @Francesco78

  Did you manage to get something new with the OB/OS dynamic areas?

  
  @ALE

  I don’t see any “nice colours”? 🙂

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20863

  Been thanked: 4773 times
  06/29/2017 at 8:26 AM #39347
  zilliq

  Hi Nicolas,

  On my graph if you compare this code and the one of reverse EMA it isn’t the same thing with a variable=4

  I don’t see where is the problem (Just for fun..)

  Zilliq

   

  

  zilliq

  Participant

  Topics: 58

  Replies: 683

  Been thanked: 152 times
  06/29/2017 at 8:41 AM #39352
  Nicolas

  I think the problem is that RE1 to RE8 are variables that must be set because they compute each one separately and you can’t create them dynamically like the way you did. It would be possible with arrays support though.. But I could be wrong, this is quick thought.

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20863

  Been thanked: 4773 times
  06/29/2017 at 8:48 AM #39353
  Francesco78

  hello everyone,

  @Nicoas, yes I did try some combination, I just wanted to test if it is usable for at least 1 strategy.

  Give me a couple of days and hopefully I will have something.

  cheers

   

  

  Francesco78

  Participant

  Topics: 43

  Replies: 242

  Been thanked: 77 times
  06/29/2017 at 9:03 AM #39354
  Nicolas

  Nice! I asked you because your post is still in the pending list for the library, we will change it accordingly to your new version 🙂

  1 user thanked author for this post.

  Francesco78

  

  Nicolas

  Keymaster

  Topics: 81

  Replies: 20863

  Been thanked: 4773 times
• Author
  Posts