John Ehlers Instantaneous Trendline

John Ehlers Instantaneous Trendline

This is an implementation of John Ehlers’ Instantaneous Trendline, as described in Market Mode Strategies (1999-10-19).

The code has been slightly enhanced to return a Zero Lag exponential moving average, like it was done here.

A more recent formula of Hilbert Transform, described in Squelch those Whipsaws (2000-24-03), can be used, by replacing the beginning of the code with:

The attached screenshot shows the Instantaneous Trendline using original description (top) of and the alternative Hilbert Transform formula (bottom).

Share this

Risk disclosure:

No information on this site is investment advice or a solicitation to buy or sell any financial instrument. Past performance is not indicative of future results. Trading may expose you to risk of loss greater than your deposits and is only suitable for experienced investors who have sufficient financial means to bear such risk.

ProRealTime ITF files and other attachments : How to import ITF files into ProRealTime platform?

New! PRC is also now on YouTube, subscribe to our channel for exclusive content and tutorials

  1. Nicolas • 08/27/2018 #

    Thanks a lot for this new valuable addition to our library. I changed line 76 with “barindex” instead of “currentbar”.

  2. Horance • 08/27/2018 #

    Thanks. I’ve noticed this leftover after submission, but it was too late for editing. BTW, I think it would be nice to add the ability to edit our own posts.

  3. Etienne • 08/27/2018 #

    Thanks Horance for porting this code to PRT.
    I would like to bring some speedup to your code. I’ve been willing to use it some strategies, but the computation time is disastrous when backtesting.
    I optimized some parts of the code. The resulting figures are unchanged, but it the code is 4 time quicker to compute.

    I did 2 optimizations :
    1/ sum of deltaPhases can be computed more quickly if you observe that the current InstPeriod is close to the previous one. (which by the way is functionaly quite intutive).
    Therefore, we reuse the previous InstPeriod to compute the summation[InstPeriod](DeltaPhase) .Then we check if we went above or below 360, then we adjust the current InstPeriod with a few loops…
    On my machine, the code is twice faster on this part…

    2/ averaging using the calculated period
    In PRT, loops are slow, very slow. We are much better off, if we use native function like average…
    The gain in speedup is approx a ratio of 4 on this part…

    I’m providing the full code below.
    Thanks again

    // Instantaneous Trendline
    // Market Mode Strategies
    // 1999-10-19 John F. Ehlers

    Price = (high+low)/2
    Imult = .635
    Qmult = .338

    If BarIndex > 6 then

    // Detrend Price
    Value3 = Price – Price[7]

    // Compute InPhase and Quadrature components
    Inphase = 1.25*(Value3[4] – Imult*Value3[2]) + Imult*InPhase[3]
    Quadrature = Value3[2] – Qmult*Value3 + Qmult*Quadrature[2]

    // Use ArcTangent to compute the current phase
    If ABS(InPhase + InPhase[1]) > 0 then
    a = ABS((Quadrature+Quadrature[1]) / (InPhase+InPhase[1]))
    Phase = ATAN(a)

    // Resolve the ArcTangent ambiguity
    If InPhase 0 then
    Phase = 180 – Phase
    If InPhase < 0 and Quadrature 0 and Quadrature < 0 then
    Phase = 360 – Phase

    // Compute a differential phase, resolve phase wraparound, and limit delta phase errors
    DeltaPhase = Phase[1] – Phase
    If Phase[1] 270 then
    DeltaPhase = 360 + Phase[1] – Phase
    If DeltaPhase 60 then
    DeltaPhase = 60

    // Sum DeltaPhases to reach 360 degrees. The sum is the instantaneous period.
    InstPeriod = InstPeriod[1]+1
    Value4 = summation[InstPeriod](DeltaPhase)
    if Value4 360 then
    if Value4 > 360 then
    while Value4 > 360 do
    InstPeriod = InstPeriod -1
    Value4 = Value4 – DeltaPhase[InstPeriod]
    while Value4 < 360 do
    Value4 = Value4 + DeltaPhase[InstPeriod]
    InstPeriod = InstPeriod + 1
    InstPeriod = InstPeriod -1

    // Resolve Instantaneous Period errors and smooth
    If InstPeriod = 0 then
    InstPeriod = InstPeriod[1]
    SmoothPeriod = .25*InstPeriod + .75*SmoothPeriod[1]

    // Compute Trendline as simple average over the measured dominant cycle period
    Period = ROUND(SmoothPeriod)
    Trendline= average [Period + 2](Price)

    Value11 = .33*(Price + .5*(Price – Price[3])) + .67*Value11[1]

    If barindex < 26 then
    Trendline = Price
    Value11 = Price


    Return Trendline as "TR", Value11 as "ZL"

Register or


avatar avatar avatar avatar avatar avatar avatar avatar
Related users ' posts
juanj I have just the place for this.. love John Ehlers's work. Thanks Nicolas!
Sapo Thank you for your indicators. As always it's perfect, I thank you for your help it's preci...
eivaremir have u got the trading view version?
Santi Salut, les amis. L'indicateur avec la nouvelle version de PRT 11 fonctionne très mal, ne fon...
oeil62 Bonjour quelles sont les valeurs des variables pour une utilisation journalière ? merci pour...
ALE As soon as the final version of ProRealtimeV11 is released, I will check for any changes to ...
jandehaas Hello Vonasi, I tried it but it does not seem to work,any idea why ? Thanks,Jan defp...
matthieuatd Hello is it possible to add the m15? Thank you
Vonasi It is possible but I will need to reread the code and remind myself how it works as I coded ...
dario ti Thank you!
Razz Hello Vonasi super indicator! is it possible to extend the indicator by the time scales M10 ...
Razz Hello Vonasi I found out myself how I can change it. Thanks again for the indicator
Vonasi There is another version of this indicator that can plot the same lines but onto an indicato...
Martin35 Hello Vonasi, do you think it is possible with PRT to code with the same principle as on thi...
teddy58 Ich habe es mit N= 1,5 und N=2,5 getestet mit 200000 Einheiten. Es funktioniert leider erst ...
Matriciel I use this indicator to help me make a decision when there is a divergence. The divergences ...
Jan Wind Hi, I personnaly am not a big fan of ZIGZAG indicator, as it repaints itself during time. ...
Nicolas Thanks again for all your valuable posts... and well formatted code! :)
Abdelkrim Maksour hi sir is this indicator for mt4 and how i can get one .
Horance This is a good question. John Ehlers explains it in Rocket Science for Traders. You have to ...
cbeukes Much appreciated, I will give that a try. Also, is there any criteria that one could use to ...
firemyst Thanks for the code! I'm trying to understand how it works, and have a few questions... line...
Bebbo Grazie del prezioso contributo Nicolas, scusa la mia inesperienza, sono alle prime armi e s...
Nicolas To avoid recalculation, you should use the one from this post.
Bebbo Ok Grazie Nicolas.
Appsoluxions Hi Nicolas, do you have the MT4 version of the indicator? I am not sure if these kind of que...
Nicolas Hi, don't have one sorry. But you can still ask for private coding through our programming s...
Appsoluxions Cool. Thanks for the response.
Etienne Hi, thanks for proving this code. I would like to contribute by adding a computation speed...
Etienne //Compute Super Smoother coefficients once if barindex = 1 then a1 = exp( -1.414 * 3.14159...
Nicolas Thanks a lot!
Bard From the url link #post-65420 above but without the chart image: Stocks & Commodities...
Bard Note: lter = filter
Bard Apologies, there's been an issue with copying "f" and "I" from the article and pasting.. whi...