I've developed MTF MAMA with Signals, Alerts & Warnings.
It's correct version of this indicator. All previous versions have bugs including Godzilla's version.
thanks 4 mama, Igor
btw. what was the bug in other versions?
i hope this work...
Godzilla's version bugs:
1. Q2 = Q1 - jI; should be Q2 = Q1 + jI;
2. if (period < 6.00*period) period = 6.00;
if (period > 50.0*period) period = 50.0;
if (period < 6.00) period = 6.00;
if (period > 50.0) period = 50.0;
3. Function CountVelue doesn't work properly(try to test without it).
Other versions are repainting the past(ex. #MAMA).
sinwave and others
Here is a screenshoot of test of different Ehlers indis with noised signal with
known content (2 cyclical components 20 and 100 bars+ gauss noise).
The results were not astonishing.
Cycle measurement was done with three methods.
Hilbert transform original Ehler code (white and blue)
Hilbert transform new Ehler code (red and yellow)
Corona chart methods
only corona shown proper period of 20 bars. Hilbert tranform method had an error measurement sometimes around 50% !!! even after smoothing with JMA.
Sinwave indicator was showing some cycles but timing with main signal was terrible
Swing position (corona charts) in some cases was showing proper swings and in other cases failing regardless of smoothing with JMA. It was showing the proper swings in area where cycle was measured properly i.e. 20 bars.
well I thought it will be better. Comments how to improve it are welcome.
thank you again, and big thanks 4 saring your works and thoughts
Scalping system #11 (The Center of Gravity) | Forex strategies revealed
trailing (tracking) sticked behind COG (polynormal, not JE); put on tester, enjoy the show
I've developed Instantaneous Trendline according to Ehlers article(in attachment), but instead of MESA I use Hilbert Transform.
Yes, That's very good instant trend line but I think Hilbert transform is not accurate enough to drive it. I use this one but use Goertzel for Dominat Cycle measurement.
See screenshots whats happen with period measured with Hilbert in case of noise. I also found Ehler document about it which I believe he removed from his site....
Regarding screenshots. In all cases there is a signal 20 and 100 bars. Only in case of clean signal DC is recognised properly.
I think the best combination will be R-MESA from Richcap from digital filters thread to measure DC plus this instant trendline.
Here is how three instant trendlines with elliptic filter perform.
Green - driven by Goertzel
Black - driven by Hilbert period indicator
Blue - driven by Corona Period
Seems to be quite big differences between them. The most accurate theoretically is green one as cycle period is measured with the lowest error.
Sinewave Support and Resistance
Hey guys does anyone know how to code a sinewave indicator for MT4 to plot support and resistance lines when the indicator crosses. Something similar to the Better Sinewave indicator available for Tradestation? If you google better sinewave e-mini I believe there is a video explaining what I mean. The indi looks rather useful so hopefully this could work on MT4 aswell. I have a couple of different sinewave indicators that I can post.
Please enable the necessary setting in your browser, otherwise you will not be able to log in.