Fourier connoisseurs... - page 2
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
Actually, the prediction is not to continue the periodic function, but to continue the
sequence which is behind the initial point of transformation, i.e. after transformation and removal of high frequencies simply to continue the curve...
And the task of what we will observe beyond 10 points is not so sharp.
It is enough to have no more than 10 points to be able to say with certainty what we are going to see...
I'm going to attach a picture of what I've got here a little bit later...
'Predicting the future with Fourier Transforms'.
I'll attach a picture of what I've got here a bit later...
You can't!
And there are five reasons for this:
The first reason is the non-stationarity of the harmonics on the price VR.
The second is ... >> go on, as the song goes.
the curve in red in the bottom picture is the Fourier transform and a couple of other functions...
green is the raw data...
The Fourier transform process requires a period selection to get a stable process at the starting point time[0]...
Fourier transform has no further effect on this process...
the curve in red in the bottom picture is the Fourier transform and a couple of other functions...
The green one is the raw data...
I don't believe it!
It's a pretty good picture - no lag, no ironing... Something must be wrong! It must be overdrawing?
What else could it be? - Otherwise, it's just a scam.
I don't believe it!
The picture is so good - no lag, and the ironing is good... Something must be wrong! Must be overdrawing?
What else could it be? - Otherwise, it's just a scam.
No. This is an elementary approximation of the period by the OPT + its error by 2*PI (the 0th bar). Because if the values at 0 and 2*PI are not equal, the OPT will produce an error on them by equating the values to the 0th harmonic, i.e. the arithmetic average of the analysed period. You can take a simple moving average and giving it as an input value the number of bars to be analyzed, we will get at the 0-th bar the value of this very moving average equal to the value of the RTO by 2*PI.
Actually, the prediction is not to continue the periodic function, but to continue the
sequence which is behind the initial point of transformation, i.e. after transformation and removal of high frequencies simply to continue the curve...
And the task of what we will observe beyond 10 points is not so sharp.
It is enough to have no more than 10 points to be able to say with certainty what we are going to see...
I'm going to attach a picture of what I've got here a little bit later.
You see, the presence of high frequencies, low frequencies, shifts are all properties of the mathematical model that describes the process. You can change and fine tune the properties and characteristics of the model infinitely, but if the mathematical model itself doesn't describe the process, you will never get a sensible result - it's like a horse model: the ideal horse is a ball in a vacuum.
Although there is another principle (this is more of an applied science) - "a driven in screw holds tighter than a driven in nail". Based on that you can try and pick up a hammer instead of a screwdriver ;).
>> Good luck.
I mean, if there is an understanding of what is being done and why, then there will be an understanding of what can be expected.
You see, the presence of highs, lows and shifts are all properties of the mathematical model that describes the process. You can endlessly change and modify properties and characteristics of the model, but if the mathematical model itself doesn't describe the process, you will never get a sensible result - it's like a horse model: the ideal horse is a ball in a vacuum.
Although there is another principle (this is more of an applied science) - "a driven in screw holds tighter than a driven in nail". Based on that you can try and pick up a hammer instead of a screwdriver ;).
Good luck.
SZZ I mean that if there is an understanding of what is being done and why, then there will also be an understanding of what can be expected.
The change in period is systematic...(-6...-4)...(+4...+6) depending on how the change
and the change either increases, up to a certain period as you like... or decreases in the same way
and then there is a sharp return to a lower period (when increasing) or higher (when decreasing)...
Neutron 24.04.2009 14:11
forte928 wrote :>>
the bottom figure is the red curve obtained by the Fourier transform and a couple of other functions...
green is the raw data...
I don't believe it!
It's a pretty good picture - no lag, and the smoothing is good... Something must be wrong! It must be overdrawing?
What else could it be? - Otherwise, it's just a scam.
Reshetov 24.04.2009 14:54
Neutron wrote (a) >>
I don't believe it!
It's a pretty good picture - no lag, and ironing is good... Something must be bad! Must be overdrawing?
What else could it be? - Otherwise, it's just a scam.
No. It's an elementary approximation of the period by OPT + its error by 2*PI (0th bar). Because if the values at 0 and 2*PI are not equal, the OPF will produce an error on them by equating the values to the 0th harmonic, i.e. the arithmetic mean of the analysed period. It is possible to take a simple moving average and having set to it as an input value the number of bars analyzed, we will receive at the 0-th bar the value of this same moving average equal to the value of the FSR by 2*PI.
If we have no more than 10 points, we can say with certainty what we are about to see...
I beg to differ, let's say we are at the end of the movement and after 10 points the trend will change,
so is it worth jumping on the train, especially since the validity of those 10 points is questionable.
I have often noticed that the first 10 points are not true, but the nearest real quotes are equal to the forecasted ones.
Here the question flows smoothly into "Fourier or last point effect", and on this question it seems to me that the effect
is caused by another effect. Try to set a straight line of the form y = k*x + c, and then extrapolate with Fourier,
and instead of an upward straight line we get a downward curve. I would call it the incomplete wave effect.
I.e. if wave doesn't fit in measurement section then correct prediction by Fourier method is not possible.
Both straight and long-period harmonics are subject to this effect.