Fast Fourier Transform - Cycle Extraction - page 107
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
And this is the final version (one case when the calculating function was falling through is identified and solved). Now it is ready for mt5
And, just for the sake of fooling around, here is a cpo of cci too :
And this is the final version (one case when the calculating function was falling through is identified and solved). Now it is ready for mt5
And, just for the sake of fooling around, here is a cpo of cci too :
Dearest MLADEN
Thanks for new tools,interesting.
may i know "cpo" stands for what ,if cycle predictor oscillator ?
and if simple cpo based on rsi ?
regards
Dearest MLADEN
Thanks for new tools,interesting.
may i know "cpo" stands for what ,if cycle predictor oscillator ?
and if simple cpo based on rsi ?
regards
mntiwana
I don't know who gave that name and what does it mean , and I did not want to change it :)
It was using a part of qpwr's "extrapolator" - but since it is not the "extrapolator", left the name as is.
And this is the final version (one case when the calculating function was falling through is identified and solved). Now it is ready for mt5
And, just for the sake of fooling around, here is a cpo of cci too :
Does the past channel of the CPO repaints(fitting) itself ?
Yes. That is fitting algorithm : the whole channel back can be changed
And the last updates (cleaned up). Next will be posted mt5 version
As an addition : extrapolated ema (using the Quinn and Fernandes algorithm for fitting as well as for extrapolating). Setting the ema period to <= is equivalent to using price (so it can be an extrapolation of price too)
Does the past channel of the CPO repaints(fitting) itself ?
A couple of additional words of explanation : setting the last bar of pas data to > 0 makes it recalculate only when a new bar is form. If you keep the last bar at 0. then any value change will case new fitting calculation and new values (it will recalculate/repaint)
Hi Mladen,
Thank you so much again for your prompt responses and for taking the initiative to update the codes. Also, great idea to make the EMA extrapolated for the main chart window. I was using your TriangularMA centered indis for the main chart for some time now, however, I like the prognostic part of the EMA extrapolated. With the right settings, I made the two indis look very similar... Unfortunately, after having placed 2 EMA extrapolated indis (1 longer and 1 shorter time periods) plus the CPO (RSI), the MT4 is considerably/visibly slower. It takes seconds to change a timeframe or a chart. Also, the fan runs much more often. I was wondering if you have any advice on how to reduce the CPU load… Or alternatively, is it possible to make the TriangularMA centered indi with forecasting lines, too? They are quite similar in my eyes, however, the TMA seems to be more MT4-friendly, if I may say so...
Hi Mladen,
Thank you so much again for your prompt responses and for taking the initiative to update the codes. Also, great idea to make the EMA extrapolated for the main chart window. I was using your TriangularMA centered indis for the main chart for some time now, however, I like the prognostic part of the EMA extrapolated. With the right settings, I made the two indis look very similar... Unfortunately, after having placed 2 EMA extrapolated indis (1 longer and 1 shorter time periods) plus the CPO (RSI), the MT4 is considerably/visibly slower. It takes seconds to change a timeframe or a chart. Also, the fan runs much more often. I was wondering if you have any advice on how to reduce the CPU load… Or alternatively, is it possible to make the TriangularMA centered indi with forecasting lines, too? They are quite similar in my eyes, however, the TMA seems to be more MT4-friendly, if I may say so...
They are similar (almost all fitting type of calculations tend to produce similar results - which is logical given that the data is the same and they should fit to that data)
As of the CPU : if you use large number of harmonics and a big value for past data, it will be slow (it is calculating number of harmonics loops x past bars calculations (x some internal loos for each calculation - you get the picture :))). Attaching versions that can be set to calculate on each tick or just on a new bar (but on a new bar, all changes of the current price will be ignored if you use the last bar in history set to 0). It will lessen the CPU lad and added one more code change that should help when changing time frames or symbol (even though I did not have problems with 4 instances added (like the example) of the old version(s), but try it out now (set the calculate on every tick to false in all and see what it does then)