Again, this model is the closest to BP. Specifically designed for non-stationary, as opposed to FFT, processes.
Will it save? don't know. Although I think the waylets are much closer NS to BP.
The thing that appeals to me most about wavelets is that the AFC has a duration in time, rather than being defined on an infinite axis.
Wavelets are not a model, just a way of representing (or decomposing, as you like) the very historical series you dislike so much. For all its merits, this method has the same problem as PF - an edge problem that forces you to make assumptions about the behaviour of the function beyond the interval. What assumptions, such are the predictions to which the method leads.
It is naive to assume that there is some method that by itself, without any meaningful models, will lead to a non-trivial result.
Do you have any meaningful market model ? OK, a simpler question - do you have a non-trivial way of solving the edge problem with wavelets ?
Wavelets are not a model, but just a way of representing (or decomposing, as you like) the very historical series that you dislike so much. For all its merits, this method has the same problem as PF - an edge problem that forces you to make assumptions about the behaviour of the function beyond the interval. What assumptions, such are the predictions to which the method leads.
It is naive to assume that there is some method that by itself, without any meaningful models, will lead to a non-trivial result.
Do you have any meaningful market model ? OK, a simpler question - do you have a non-trivial way of solving the edge problem with wavelets ?
A model is a representation of the original series by something else, which has a property of the original series (with some certainty), but which has another property that makes it possible to apply some known and tested methods. Example, the stubborn desire to replace BP by a stationary process.
The question about the boundary problem is IMHO far-fetched, since we don't need anything beyond the edge of the wavelet, not the FFT. The wavelet has ended and a new wavelet or group of wavelets has started, which should be let into the thrasholding. Roughly, we are interested in confirmation of the trend start and end - we don't care about the rest.
You see :) You published an article on this site for some reason, but when asked to make a short summary, a 20-word summary, you replied that - fuck you... idiots.
I can't add anything new to this thread in principle, but at least I can make fun of it...
What's new to add? Reshetov read a bit about game theory. He wrote a whole article entitled "How to make money on unsteady markets? He added "(Article)" on the side to make it clear to everyone. By the way, he still hasn't answered the question. Next, he will mutter that he will not cast beads in front of swine and will not reveal the secret. And to explain everything properly to nerds - he will write ten more articles and lectures.
But even with the rudeness and porkiness of our home-grown man of letters - I respect him for his tireless energy and curiosity!
The question about the boundary problem is farfetched IMHO since we don't need anything beyond the edge of the waylet, it's not the FFT. Weylet has run out and a new weylet or group of weyletons has started, which should be let into the thrasholding.
It's weird, I thought the profit lies right there, behind the edge of the wavelet.
Allow the use of cookies to log in to the MQL5.com website.
Please enable the necessary setting in your browser, otherwise you will not be able to log in.
You are signed in. Two-factor authentication has been disabled. Set it up again and scan a new QR code, or continue with two-factor authentication disabled.
faa1947, why do you think it is wavelets that will save the father of Russian democracy?
faa1947, why do you think that wavelets will save the father of Russian democracy?
Again, this model is the closest to BP. It is specially designed for non-stationary, as opposed to FFT, processes.
Will they save me? I do not know. Although I think the waylets are much closer NS to BP.
The thing that appeals to me most about wavelets is that the ACF has a duration in time, rather than being defined on an infinite axis.
Again, this model is the closest to BP. Specifically designed for non-stationary, as opposed to FFT, processes.
Will it save? don't know. Although I think the waylets are much closer NS to BP.
The thing that appeals to me most about wavelets is that the AFC has a duration in time, rather than being defined on an infinite axis.
Wavelets are not a model, just a way of representing (or decomposing, as you like) the very historical series you dislike so much. For all its merits, this method has the same problem as PF - an edge problem that forces you to make assumptions about the behaviour of the function beyond the interval. What assumptions, such are the predictions to which the method leads.
It is naive to assume that there is some method that by itself, without any meaningful models, will lead to a non-trivial result.
Do you have any meaningful market model ? OK, a simpler question - do you have a non-trivial way of solving the edge problem with wavelets ?
That too to the annals.
P.S. Yuri, nothing provocatively personal. I can't bring anything principally new into this branch, but at least I'll make fun of it...
Wavelets are not a model, but just a way of representing (or decomposing, as you like) the very historical series that you dislike so much. For all its merits, this method has the same problem as PF - an edge problem that forces you to make assumptions about the behaviour of the function beyond the interval. What assumptions, such are the predictions to which the method leads.
It is naive to assume that there is some method that by itself, without any meaningful models, will lead to a non-trivial result.
Do you have any meaningful market model ? OK, a simpler question - do you have a non-trivial way of solving the edge problem with wavelets ?
A model is a representation of the original series by something else, which has a property of the original series (with some certainty), but which has another property that makes it possible to apply some known and tested methods. Example, the stubborn desire to replace BP by a stationary process.
The question about the boundary problem is IMHO far-fetched, since we don't need anything beyond the edge of the wavelet, not the FFT. The wavelet has ended and a new wavelet or group of wavelets has started, which should be let into the thrasholding. Roughly, we are interested in confirmation of the trend start and end - we don't care about the rest.
faa1947 писал(а) >>
The game of two players - the trader and the market - is bullshit!
...
Holy shit! I did not have time to move away, but again nerds - litterbugs polluted the whole branch.
Gentlemen assholes, who forbids you to create your own threads and discuss there all sorts of aliens, wavelets, and other trousers with codpiece?
You see :) You published an article on this site for some reason, but when asked to make a short summary, a 20-word summary, you replied that - fuck you... idiots.
And now all of a sudden :) you wonder.
Why did you write here? So you can be rude? :)
Think about it.
Mathemat писал(а) >>
I can't add anything new to this thread in principle, but at least I can make fun of it...
What's new to add? Reshetov read a bit about game theory. He wrote a whole article entitled "How to make money on unsteady markets? He added "(Article)" on the side to make it clear to everyone. By the way, he still hasn't answered the question. Next, he will mutter that he will not cast beads in front of swine and will not reveal the secret. And to explain everything properly to nerds - he will write ten more articles and lectures.
But even with the rudeness and porkiness of our home-grown man of letters - I respect him for his tireless energy and curiosity!
P.S. Yuri, nothing too defiantly personal. I can't really contribute anything new to this thread, but at least I'm making fun of it...
That's what I'm saying. :-))
The question about the boundary problem is farfetched IMHO since we don't need anything beyond the edge of the waylet, it's not the FFT. Weylet has run out and a new weylet or group of weyletons has started, which should be let into the thrasholding.
It's weird, I thought the profit lies right there, behind the edge of the wavelet.