Predicting the future with Fourier transforms - page 50
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
Pardon me, but this is not an explanation of Fourier, but a demonstration of his total lack of understanding.
I don't agree with that. Take a semi-period of a sine (raw data) and do its decomposition in a harmonic series. I think you will be pleasantly surprised.
Well, you've got it all figured out, don't you?
I got it, of course. Still don't understand, unlike the drunken hedgehog, that after reaching a previously reached high the price will go to a previously reached low.
The point is that in financial markets the high on history can easily turn out to be the low in the future ))))
I don't agree. Take a semi-period of sine (raw data) and make its expansion in a harmonic series. I think you will be pleasantly surprised.
Disagree:)
Here you go:
It's a bit of a mess, it's a bit of a mess, but still...
The red one is the raw data. The yellow one is the result of summing up the terms.
Here's an extension for four periods:
Don't agree:)
Here you go:
It's a bit of a mess, it's a bit crooked, but still...
The red one is the raw data. The yellow one is the result of summing up the row terms.
Dimitri, you are simply superb! (I'm not kidding).
Allow the yellow one to extend to the right.
You beat me to it. The result is a nice periodic function that doesn't have much in common with the original series.
The original function was a sine wave. Trade on the yellow one... I'm going to bed.
Have fun
I get it, of course. I still do not understand, unlike the drunken hedgehog, that after reaching a previously reached high, the price will go to a previously reached low.
For the particularly gifted, one more time: not the price will go away, but the amplitude of the first harmonic.
Don't agree:)
Here you go:
It's a bit of a mess, it's a bit crooked, but still...
The red one is the raw data. The yellow one is the result of summing up the row terms.
What do you mean by that? That if you decompose half a period of a sine wave and add it back up, you get the same half sine wave? It's not like we're completely stupid and we know that. Show not the result of summing the terms of the series, but the individual terms of the series themselves. And explain why you need an accordion of frequencies that were not in the original signal. And if you show us what good can be done on the basis of Fourier (preferably extrapolator, because the thread is about it), it will be very good.
Here's mine for comparison. I just added it.
Green line - input signal s(i)=sin(PI/24*i)+sin(PI/3*i). When the tests and adjustments are finished, the prices will be here.
The white is the result of extrapolation, plotted from the data to the left of the vertical line inclusive.
Everything else is the result of signal decomposition by digital filters. The dotted lines are for the extrapolated signal, the solid lines are for the actual signal.
You can certainly do better on a Fourier basis, after all I don't understand anything about it.