Adaptive lookback indicators - page 10
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Good idea
Simba i'll answer the last 2 first, my settings are for me default, except since i like black background charts, i have been changing the channel colors to something easier for me to see, also deviation at 2,4,6, and future center line=true,Filter is smoothing in HT remember reading from an earlier post that Ehlers likes it set to 7 but really don't see much difference in 7 or 1 using it in this indicator, now about it taking awhile before showing itself would like to say something catchy like its gathering all data to make its decision, but looking in the code really can't say, why!
Bill,
Sorry to say so,but the indicator appears,disappears-like ND drawing
-points up then down-repaints,redraws,and probably has loose personal habits when nobody is looking 
.
I think that your idea is exceptionally good,use HT to project the future,and I think so because,even using Fourier gives good results(see pic),if the base indicator is ok,so,I presume that using ITSD(the improved version of EMD and Hilbert Huang transform-Intrinsic Time Scale Decomposition) as a base extrapolator,instead of the usual Fourier,will be even better.
In the pic you can see a Fourier extrapolator,last bar at 68 bars ago,that has predicted,very fairly,the past 3 turns for EURUSD...Next one is in a couple of hours
....I believe that using ITSD will add some gin to the tonic....what do you think?Can we have a party? 
S
Bill,
Sorry to say so,but the indicator appears,disappears-like ND drawing
-points up then down-repaints,redraws,and probably has loose personal habits when nobody is looking 
.
I think that your idea is exceptionally good,use HT to project the future,and I think so because,even using Fourier gives good results(see pic),if the base indicator is ok,so,I presume that using ITSD(the improved version of EMD and Hilbert Huang transform-Intrinsic Time Scale Decomposition) as a base extrapolator,instead of the usual Fourier,will be even better.
In the pic you can see a Fourier extrapolator,last bar at 68 bars ago,that has predicted,very fairly,the past 3 turns for EURUSD...Next one is in a couple of hours
....I believe that using ITSD will add some gin to the tonic....what do you think?Can we have a party? 
SI agree ITSD (interesting name) as a base extrapolator would be interesting, have been trying to apply HT to FFT of color as base extrapolator instead of the existing base but no cigar so far!
Itd
Stlm can be interesting if based on the output of ITD. If I upload a Matlab file for Intrinsic Timescale Decomposition, would someone be willing to write the mql4?
And maybe we should try to do something with the Teager-Kaiser operator also..
I'm saying this because EMD clearly suffers from edge effects (because it uses a spline as basis expansion this is inevitable).
Some more interesting mods to me, posting the Stlm here my thinking is it is a digital adaptation., and also an non lag version of phase accumulation bbands!
hi mrtools
pa noglag bbmacd
mtf version + alerts
thanks
Look at the attachment - zipped ppt file with different kinds of improvement of EMD...
Look at the attachment - zipped ppt file with different kinds of improvement of EMD...
Interesting presentation, even if the author says under Future works:
"To improve boundary effect and preserve complete physical meanings of decomposed signals"
Stlm can be interesting if based on the output of ITD. If I upload a Matlab file for Intrinsic Timescale Decomposition, would someone be willing to write the mql4?
And maybe we should try to do something with the Teager-Kaiser operator also..
I'm saying this because EMD clearly suffers from edge effects (because it uses a spline as basis expansion this is inevitable).Do you have a matlab code for ITD or for the standard EMD?
If you do have the ITD, I'd be happy to see it and perhaps give it a try (maybe with some support from Mladen?
)
- Trendik
If you do have the ITD, I'd be happy to see it and perhaps give it a try (maybe with some support from Mladen?
I'll post Matlab code for ITD in 1 week.
I'll post Matlab code for ITD in 1 week.
Hi MrM, great. Looking forward to seeing this. Are you planning to write it yourself?
What bothers me about the ITD is the lag. Looking at the algorithm you see that the first intrinsic wave is extracted whenever you observe the next local extrema. Then you extract the fastest intrinsic wave and you extract the next one recuresively when you see the next local extrama in the reminder of the signal (after subtracting that wave), etc. So essentially you incur, even for the first wave, a lag that is similar zigzag. So I wonder how useful this can be in real time. Does this make sense? What do you think?
Regards,
- trendick
Here is something for someone, who wants to work with Teager-Kaiser Operator
Look into attachement - zipped files (pdf and matlab) .
Maybe someone can modified and implement them for mql???