Predicting the future with Fourier transforms - page 40
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
The MA smoothes (averages), not predicts. In essence, so does Fourier. So, is smoothing (averaging) a prediction?
Well, I would personally look at examples here.
Trololo, I take it you are not happy with Vadim's methods. But then how do you pose the question? Are you opposing Fourier? Sorry, I really don't understand, I have no time to reread everything.
And I'll be able to havemy arguments in private. :)
Clearly, smoothing is ex post facto and cannot be predictive for non-stationary data. For stationary BPs, the result of averaging is a prediction because MO = Const, i.e. if BP deviates from it, it is bound to return to it.
I totally agree with it. But maybe the harmonic decomposition is the wave function of the signal...
mt4trade:
Well, I would personally look at the examples here.
Trololo, I take it you are not happy with Vadim's methods. But then how do you pose the question? Are you opposing Fourier? Sorry, I really don't understand, I have no time to reread everything.
And I can havemy arguments in private. :)
I don't want to discuss this man, even though I have every reason to stick with my opinion of him (not the best opinion). I was warned that a ban will follow, so please don't. I don't want to lose my nickname yet.
I'm not interested in the volume of such an unnecessary reading in a branch, not me.
I'm sorry, it's also hard to retell it all personally from beginning to end.
There are people here who took part in this thread a long time ago and they've already decided what to expect from them. Those who are for-silent, those who are against- why repeat yourself?
Look at how many pages of leftist talk it's been and you'll see that it's not over yet.
They are not that interested, they just want to piss off. Fucking hell, make a page, no one has more fun.
How, then, can Fourier be used to predict the future?
... I, on the contrary, am in favour of Fourier, and they rub it in my face that it doesn't work.
Well, we should be discussing not the person, but the methods - whether they're effective or not. And here, as a rule, someone works fine or more or less well, and someone else does not.
But by Fourier I sketched it out: in the "window" it's a fit, when it comes out it's again, unsteady. But harmonics can work for a while. And here we just need statistics. If someone has researched and knows - then why step on the rake again. If, for example, you have a different point of view, so give the info-justification, pls.
Then, how can Fourier be used to predict the future?
And with MA you can, and with non-stationary data you can even use the direction of the MA.
It depends on how )))) The MA is the most honest indicator.
How, then, can Fourier be used to predict the future?
In fact, Fourier is not necessary for prediction, i.e. it is like a fifth leg for a dog. If we found that BP is periodic, then prediction task is trivial: take some point in the future, step back from it by x * 2 * PI into already known past, where x is any integer and get prediction value.
And if BP is aperiodic, Fourier will not help here either, because Fourier theorem makes sense only for periodic functions.
What we need to discuss is not The discussion should be about a person, and about techniques - whether they are effective or not. And here, as a rule, someone has everything fine or more or less works, and someone else does not.
But by Fourier I sketched it out: in the "window" it's a fit, when it comes out it's again, unsteady. But harmonics can work for a while. And here we just need statistics. If someone has researched and knows - then why step on the rake again. If, for example, you have a different point of view, then give reasons, pls.
How to discuss them if there is no methodology, except for beautiful colorful pictures and poetry about seven years of writing code. i understand the essence, but i do not want to pick the thoughts of someone who does not want to discuss them, and i do not want you to. If you want to move in the same direction, read the branch of the method of tendential planimetry, there I wrote a little bit about it, the beginning is exactly the same. Further a little more complicated, the volume of calculations is very large, that is the complexity, then sometime I will add that idea on the index fans, while another is interesting.
Why am I repeating myself? If you are not too lazy to read my posts and my profile, read my postings and trollolo's nickname. https://forum.mql4.com/ru/45987/page11
There is a description from different sides.