A small quote from a man almost everybody knows (John Ehlers) :
The problem with virtually all indicators is that they are causal. That means they directly depend on data for their computation. As a result, the computation cannot be accomplished until after the data arrives, and - as a result - all causal indicators have lag. Lag is perhaps the traders' worst enemy, particularly when trading the cycle mode when relatively short term entries and exits are expected.
So "to causal or to non-causal" or maybe "to non-causal or to causal" ?
mladen: Funny that you mention Goertzel (being a sub-set of FFT) and Mesa Appart from that, I guess that crodzilla will be very pleased with you reading his posts over at FF - namely this post :
But than, after all, knowledge does not know borders ...
the question is How do you combine waves with different frequencies and amplitudes into 1 composite waveform?this is the key i thinks
Let me answer indirectly : Brian Millard, which continued some work of Hurst, recommended in his work only 1 (one) indicator - centered Triangular Moving Average. Now snake is Centered Triangular Moving Average + some extrapolation. And since I like Millards work, the conclusion is ...
PS: attaching you one version of centered triangular MA (it was already posted on some of the public threads, but right now can not remember where, and I figure it wont hurt to post it again). I recomend it not because I wrote this version but because the code in this one is simpler and when it comes to code : the simpler it is the smaller the chance that there will be an error (and much easier to use in any further development)
regards
mladen
Thank you Mladen...I have used your T3 version too...
mrtools: Like you Leledc been experimenting with macd (and some other things) using some of Mladen's and also Igorad's indicators so far so good, main thing for me is computer hasn't blown up!!
Allow the use of cookies to log in to the MQL5.com website.
Please enable the necessary setting in your browser, otherwise you will not be able to log in.
You are signed in. Two-factor authentication has been disabled. Set it up again and scan a new QR code, or continue with two-factor authentication disabled.
...
A small quote from a man almost everybody knows (John Ehlers) :
So "to causal or to non-causal" or maybe "to non-causal or to causal"
?
non-causal indicators
anecdotal evidence
personal vendetta's
alternative for composite wave?
So "to causal or to non-causal" or maybe "to non-causal or to causal"
That is not the question
because the answer is known. Instead of wasting energy on this, why not get back to the original goal of this thread: digital filters?
Does anyone have any suggestions for an alternative method to recreate the composite waveform after doing the spectral analysis
(with Goertzel / Mesa / ...):
Cycle is bars per cycle
u = (2*PI)/Cycle
w = Cycle
x = Amplitude
y = Phase in Radians
z = (y * (180/PI)) / 360 : to convert radians into % of wave
h = reference (drives waveform creation)
wave = x * (sin(u * (h + (z * w))))
and then simply add up all the waves to create the composite.
...
Funny that you mention Goertzel (being a sub-set of FFT) and Mesa
Appart from that, I guess that crodzilla will be very pleased with you reading his posts over at FF - namely this post :But than, after all, knowledge does not know borders ...That is not the question
because the answer is known. Instead of wasting energy on this, why not get back to the original goal of this thread: digital filters?
Does anyone have any suggestions for an alternative method to recreate the composite waveform after doing the spectral analysis
(with Goertzel / Mesa / ...):
Cycle is bars per cycle
u = (2*PI)/Cycle
w = Cycle
x = Amplitude
y = Phase in Radians
z = (y * (180/PI)) / 360 : to convert radians into % of wave
h = reference (drives waveform creation)
wave = x * (sin(u * (h + (z * w))))
and then simply add up all the waves to create the composite.hello mladen
Funny that you mention Goertzel (being a sub-set of FFT) and Mesa Appart from that, I guess that crodzilla will be very pleased with you reading his posts over at FF - namely this post :But than, after all, knowledge does not know borders ...
the question is How do you combine waves with different frequencies and amplitudes into 1 composite waveform?this is the key i thinks
leledc,
Let me answer indirectly : Brian Millard, which continued some work of Hurst, recommended in his work only 1 (one) indicator - centered Triangular Moving Average. Now snake is Centered Triangular Moving Average + some extrapolation. And since I like Millards work, the conclusion is ...
PS: attaching you one version of centered triangular MA (it was already posted on some of the public threads, but right now can not remember where, and I figure it wont hurt to post it again). I recomend it not because I wrote this version but because the code in this one is simpler and when it comes to code : the simpler it is the smaller the chance that there will be an error (and much easier to use in any further development)
regards
mladenThank you Mladen...I have used your T3 version too...
Like you Leledc been experimenting
Good work !
Good work !
Thanks man, same with you your Fourier mods are making a big difference with my decision making!
Thanks man, same with you your Fourier mods are making a big difference with my decision making!
Yes the same with me, Leledc has made a brilliant work
A week we have it, and me too it changes one of my sub-system for the best
Thanks !