Dominant Cycle calculation by band pass crossing over

Hasardeur · Tue, 03 Apr 2018 21:14:59 +0000

Dear Traders, Dear Nicolas, the attached code made by John Ehlers calculates the dominant cycle by counting the zero crossing over of a band pass filter. The original filter is using a fix DCPeriod of 20. I have made the code adaptive by iterative using the result of DC instead of fix DCPeriod at the beginning of the code. So far it is working well. By changing the actual shown time frame for example from 100 days to 1000 days the DC of the actual bar (today) is sometimes fluctuating. That means that for example at shown 100 days the DC of the actual bar is 14, by changing the shown timeframe 1000 days DC of the actual bar is 35. Does someone has an idea to modify the code into a correct and stable iterative Dominant Cycle Period? Many thanks for your reply in advance. Hasardeur REM --------------------------------------------- REM Modified Dominant Cycle Measured by Zero Crossing of Band Pass filter from Cycle Analytics for Traders by John F.Ehlers 2013 REM --------------------------------------------- //PeriodDC= 20 // original by John Ehlers PeriodDC= MAX(8, MIN(DC, 48)) IF barindex = 1 THEN BP= 0.01 DC= 20 ELSIF Barindex >= 2 THEN REM --------------------------------------------- REM Highpass Filter and BP or SuperSmoother Filter together form a Roofing Filter REM --------------------------------------------- HPPrice= close HPPeriod= PeriodDC HPBandwidth= 0.3 // 0.7 original Dominant Cycle Measured, 0.3 (0.10-0.50) from Roofing Filter HPalpha= (COS(0.25*HPBandwidth*360/HPPeriod) + SIN(0.25*HPBandwidth*360/HPPeriod)-1) / COS(0.25*HPBandwidth*360/HPPeriod) HP= (1+HPalpha/2)*(HPPrice-HPPrice[1]) + (1-HPalpha)*HP[1] BPPrice= HP BPPeriod= PeriodDC BPBandwidth= HPBandwidth beta= COS(360/BPPeriod) gamma= 1/COS(360*BPBandwidth/BPPeriod) BPalpha= gamma - SQRT(gamma*gamma-1) BP= 0.5*(1-BPalpha)*(BPPrice-BPPrice[2]) + beta*(1+BPalpha)*BP[1] - BPalpha*BP[2] REM --------------------------------------------- REM Calculation of Dominant Cycle by counting Zero crossing of BP REM --------------------------------------------- CountPrice= BP DC= DC[1] DC= MAX(6, DC) counter= counter + 1 IF CountPrice CROSSES OVER 0 OR CountPrice CROSSES UNDER 0 THEN DC= 2*counter ThrashH= 1.11 // original 1.25 IF 2*counter > ThrashH*DC[1] THEN DC= ThrashH*DC[1] ENDIF ThrashL= 0.90 // original 0.8 IF 2*counter < ThrashL*DC[1] THEN DC= ThrashL*DC[1] ENDIF counter= 0 ENDIF ENDIF DC= MAX(8, MIN(DC, 48)) RETURN DC coloured(0,0,255) as "DomCyclePeriod"

This topic has 52 replies, 3 voices, and was last updated 8 months ago by LucasBest.

Viewing 15 posts - 31 through 45 (of 53 total)

← 1 2 3 4 →

08/09/2019 at 2:53 PM #104477

Hasardeur

Hi Bard,

i have no idea, the code is working well at my system.
I am using the beta version 11.1.
Did you check the availability of the Stochasticd…. at your workstation.

1 user thanked author for this post.

Bard

Hasardeur
Participant

Topics: 7
Replies: 38
Been thanked: 12 times

08/09/2019 at 3:38 PM #104479

Vonasi

The syntax for STOCHASTIC in v10.3 can be found here:

https://www.prorealcode.com/documentation/stochastic/

There are only two variables which is why you get the error. Perhaps in v11 STOCHASTIC has been modified to allow for the third value without an error but as Nicolas explains in the comments the third value is not really required as it is just the smoothing period for the signal line and so is not required to calculate it.

1 user thanked author for this post.

Bard

Vonasi
Moderator

Topics: 192
Replies: 5088
Been thanked: 1810 times

09/05/2019 at 4:49 PM #106623

Bard

Hi @Vonasi,

I looked at the documentation and see the format is:

Stochastic[N,K](price)

1

Stochastic[N,K](price)

So I removed the extra third value but still get a warning using the following code? :

DCSTO= Stochasticd[Period,3](close) mySTO= Stochasticd[14,3](close)

1
2

DCSTO= Stochasticd[Period,3](close)
mySTO= Stochasticd[14,3](close)

Any ideas?

Cheers!

Bard
Participant

Topics: 131
Replies: 809
Been thanked: 99 times

09/05/2019 at 11:19 PM #106661

Vonasi

I’m guessing that the extra ‘d’ on the end of STOCHASTIC is a typo and not actually in your code?

With this code I get two lines drawn in a separate indicator window – do you?

period = 10 DCSTO= Stochastic[Period,3](close) mySTO= Stochastic[14,3](close) return dcsto, mysto

1
2
3
4
5

period = 10
DCSTO= Stochastic[Period,3](close)
mySTO= Stochastic[14,3](close)

return dcsto, mysto

Vonasi
Moderator

Topics: 192
Replies: 5088
Been thanked: 1810 times

09/06/2019 at 2:47 PM #106714

Bard

Ah, yes thanks, it works taking out the “d” and using your code but now it says “the variable is not used in the code:bp”?

Pls see https://www.prorealcode.com/topic/dominat-cycle-calculation-by-band-pass-corssing-over/page/2/#post-104226

Cheers

Bard
Participant

Topics: 131
Replies: 809
Been thanked: 99 times

09/06/2019 at 3:28 PM #106723

Vonasi

Change BP for IGNORED in the CALL line.

1 user thanked author for this post.

Bard

Vonasi
Moderator

Topics: 192
Replies: 5088
Been thanked: 1810 times

09/16/2019 at 7:01 PM #107773

Bard

Hi @Vonasi,

I just noticed something which I don’t understand, I can import BG DC STO (I had deleted it last week) and even with the extra values of “3”:

DCSTO= Stochasticd[Period,3,3](close) mySTO= Stochasticd[14,3,3](close)

1
2

DCSTO= Stochasticd[Period,3,3](close)
mySTO= Stochasticd[14,3,3](close)

it does load okay onto the bottom of a chart (whereas I am sure weeks ago I couldn’t get it to because of the extra “3” and hence my writing on this thread) but now.. as soon as I replace BG with Ignored and add it, it now gives me this warning — Pls see image:

Your code from post above (#106661) loads fine but then it doesn’t use the dominant cycle period, right? (#104226)
Any ideas!?

Cheers.

Bard
Participant

Topics: 131
Replies: 809
Been thanked: 99 times

09/16/2019 at 8:25 PM #107779

robertogozzi

Remove the extra “d” from Stochasticd, since it thinks it’s a variable and you are wrongly using two parameters, instead of one, to reference previous bars!

Moreover, use only 2 parameters with STOCHASTIC.

1 user thanked author for this post.

Bard

robertogozzi
Moderator

Topics: 69
Replies: 16477
Been thanked: 3643 times

09/16/2019 at 8:58 PM #107781

Bard

Sorry, I just realised I had downloaded the original code with the “d” in the code. Removed it. All good now, thanks.

Bard
Participant

Topics: 131
Replies: 809
Been thanked: 99 times

09/16/2019 at 9:33 PM #107784

Bard

Dominant Cycle Stochastic and regular Stochastic Comparison — £/Y:

Bard
Participant

Topics: 131
Replies: 809
Been thanked: 99 times

09/16/2019 at 9:42 PM #107788

Bard

Brent Crude – including Dominant Cycle RSI:

Bard
Participant

Topics: 131
Replies: 809
Been thanked: 99 times

10/03/2019 at 9:03 AM #109149

Hasardeur

Hi Bard,

niche to see it is working!

Do you have an impression about the advantage?

Best regards

Hasardeur

Hasardeur
Participant

Topics: 7
Replies: 38
Been thanked: 12 times

10/03/2019 at 6:34 PM #109204

Bard

Hi Hasardeur,

Yes, all good now! I am not sure if you can use it as a stand alone indicator because these type indicators can stay overbought and oversold for long periods.

I think it’s good to use it (with other indicators) and also when the RSI or Oscillator crosses above the 30 line and below the 70 line. Do you use it on it’s own or do you prefer to use it with another indicator(s)?

I’ve been reading a lot about Hurst cycles lately. Did you ever find a good indicator that can calculate the average cycle period? John Ehlers MESA Period or his Dominant Cycle Finder I believe will only tell you what the cycle period is right now but Hurst recommends calculating the time period between the lows on daily and weekly charts and then totalling the time up and then averaging these time spans by the number of points identified to determine when the next low. He identified stock market cycles with these time spans – pls see image. It would be great if there was s/ware that could determine those averages. Do you know of anything like this @Nicolas? Like an average cycle finder.

Cheers!

1 user thanked author for this post.

Meta Signals Pro

Bard
Participant

Topics: 131
Replies: 809
Been thanked: 99 times

10/03/2019 at 9:03 PM #109211

Hasardeur

Hi Bard,

mainly I am using the following indicators for ma favorite trading system
Trend: Aroon, Ravi, Momentum: RMI, Trend following: CCI, Volume: Volume MFI

In addition I am looking for Bollinger Bands, Stochastic MADC and volume as second trading system.

The dominant cycle period indicator I am using to tune the above indicators (to the 1*period or 0.5*Period by using DC) in order to be as near as possible to the actual real cycle period.

In order to get an easy overview I have sum up the above indicators in a genesis matrix design.

Best regards
Hasardeur

2 users thanked author for this post.

Meta Signals Pro, Bard

Hasardeur
Participant

Topics: 7
Replies: 38
Been thanked: 12 times

08/24/2023 at 4:22 PM #219750

Meta Signals Pro

This is the RAVI indicator developed by @Nicolas (thanks btw)

https://www.prorealcode.com/prorealtime-indicators/range-action-verification-index-ravi-extended/

This is the RMI indicator developed by @Gabri (thanks btw)

https://www.prorealcode.com/prorealtime-indicators/relative-momentum-index-rmi/

Meta Signals Pro
Participant

Topics: 40
Replies: 310
Been thanked: 37 times
Author

Posts