Random Flow Theory and FOREX - 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
I didn't want to get into a discussion, but the wikipedia definition of stationary noise is this:
White noise is stationary noise whose spectral components are evenly distributed throughout the range of frequencies involved.
I think that the predictability of the signal has not yet come out of it. Or what did you want to predict? Well on the first point (that we're dealing with white noise), I'm not so sure it is....
(This is not addressed to "chumazik", but in general).
What the fuck "frequencies"?! What "frequencies" are there? Who determined that there are "frequencies"? Who can even claim that there is a "sine wave" or a "group of sine waves"? There are CYCLES there, but cycles are not necessarily a SYNUSOID, in relation to which it is customary in modern science to speak of "frequency".
See? Parsing word by word you can see for yourself that completely inapplicable terms and methods are applied to NAOBUM trading.
But isn't that wonderful?! So a new and reliable forecasting system can be created.
Don't play dumb with me, Chumazik. I can "see the sine waves" even in the picture of a Negro on your avatar if I want to, using a Photoshop filter or something else. But that doesn't mean that a picture of a negro is COMPLETELY made up of some kind of sinusoids. You (and others) have a broken causal link between phenomena: You think that if some filter (e.g. Fourier filter) can interpolate a chunk of samples using sinusoids, then you think this means that the process in question is in fact generated by a group of sinusoidal oscillations and "consists" of sinusoids "and some noise add-on". Can you see where you have an error or explain in more detail. There was a thread here on the forum talking about the fallacy of applying Fourier to anything and everything.
As a nerd to a PTU student: I don't believe in Fourier analysis for quote flows for certain reasons. I do not believe that quotes are white noise. If I have not made myself clear, please accept a thousand apologies.
P.S.
If you make a negro out of sinusoids, then it's silly to claim that it doesn't consist of them :) But actually, we're not interpolating here, we're extrapolating.
I didn't want to get into a discussion, but the wikipedia definition of stationary noise is this:
Stationary noise is noise that is characterized by constancy in the mean parameters: intensity(power), intensity distribution over the spectrum(spectral density), and the autocorrelation function.
Stationary noise is noise that has spectral components evenly spaced throughout the range of frequencies involved.
I don't think the predictability of the signal has yet to result from it. Or what did you want to predict? And on the first thesis (that we are dealing with white noise), I'm not sure that it is so .....
In mathematics, a stationary process is a process with mean and covariance independent of time. That is, the two basic parameters are costants.
The simplest example: a process with normal distribution N(0,1). For such a process, if x(t)=2, then with a 97.5% probability x(t+1) will be less than 2. That is, the process will go down. It is not guaranteed, then in 97 cases out of 100 it will be.
A more complex example: the AR(1) process x(t)=x(t-1)*a + s(t), where a<1 and s(t) is a stationary process, noise with some finite parameters. This process will also be stationary and its parameters can be calculated from the parameters s(t) and a. Accordingly, if this process has deviated from the mean, it can always be calculated when it will return there with a given probability.
But if parameter a=1, then we get random walk, i.e. non-stationary process, and where it will end up cannot be predicted.
Naturally, we will never see white noise in the real date, just as we will never see a real stationary process, but with a certain degree of assumption we can assume that the noise is still white and the process is stationary.
First of all, you guy doesn't know what "stationary" is, because it's a concept introduced in modern science for completely different natural phenomena, and if you start going into detail about its definition you will come across every, I repeat at every word, CARDINAL discrepancy in the definition of "stationary (noise, process)" to such a HUMAN phenomenon as the flow of currency prices.
That's the news. "And the men don't know!". They give out Nobel Prizes to each other, they've invented a whole science - econometrics. If I meet someone, I'll be sure to tell him.
You ate a lot of French alcohol or something, you're too chipper.
linking into a place where you show your own ignorance...
You're making a fuss about multiples. They're all multiples in a discrete series, and whoever needs more multiples, they go deeper down to the tick chart.
In mathematics, a stationary process is one in which mean and covariance are independent of time. I.e. the two main parameters are costants.
The simplest example: a process with normal distribution N(0,1). For such a process, if x(t)=2, then with a 97.5% probability x(t+1) will be less than 2. That is, the process will go down. It is not guaranteed, then in 97 cases out of 100 it will be.
A more complex example: the AR(1) process x(t)=x(t-1)*a + s(t), where a<1 and s(t) is a stationary process, noise with some finite parameters. This process will also be stationary and its parameters can be calculated from the parameters s(t) and a. Accordingly, if this process has deviated from the mean, it can always be calculated when it will return there with a given probability.
But if parameter a=1, then we get random walk, i.e. non-stationary process, and where it will end up cannot be predicted.
Naturally, we will never see white noise in the real date, just as we will never see a real stationary process, but with a certain degree of assumption we can assume that the noise is still white and the process is stationary.
Not exactly, not every process is characterised by mean and covariance. Your first sentence describes... A covariance stationary process, which is also stationary :)
http://books.google.de/books?id=B8_1UBmqVUoC&pg=PA46&lpg=PA46&dq=process+mean+covariance&source=bl&ots=2nJH-s67AR&sig=J_QcD2llCaELbBgPt_THGGi8ZXM&hl=de&ei=ozNoSoaCOsOg_Aa__5yeCw&sa=X&oi=book_result&ct=result&resnum=6
Not exactly, not every process is characterised by mean and covariance. Your first sentence describes... A covariance stationary process, which is also stationary :)
It's weakly stationary, it's wide stationary, it's just stationary. >> so?
In mathematics, a stationary process is one in which mean and covariance are independent of time. I.e. the two main parameters are costants.
The simplest example: a process with normal distribution N(0,1). For such a process, if x(t)=2, then with a 97.5% probability x(t+1) will be less than 2. That is, the process will go down. It is not guaranteed, then in 97 cases out of 100 it will be.
A more complex example: the AR(1) process x(t)=x(t-1)*a + s(t), where a<1 and s(t) is a stationary process, noise with some finite parameters. This process will also be stationary and its parameters can be calculated from the parameters s(t) and a. Accordingly, if this process has deviated from the mean, it can always be calculated when it will return there with a given probability.
But if parameter a=1, then we get random walk, i.e. non-stationary process, and where it will end up cannot be predicted.
Of course, we will never see white noise in the real date and we will never see a real stationary process, but with some assumptions we can assume that the noise is still white and the process is stationary.
I just don't know of any 'stationary' processes in mathematics. There are stationary RUNNING processes. What does this have to do with a tick price series that is generated by the deliberate activity of a group of disparate people looking at a graph and making decisions based on a target function communicated to them by their superiors? What does this have to do with randomness or stationary randomness? What does this random stationarity have to do with our well-known cycles, in which cyclicality does not and cannot exist, because it will no longer be a random process? What does one have to do with the other?
I just don't know of any 'stationary' processes in mathematics. There are stationary RUNNING processes. What does this have to do with a tick price series that is generated by the deliberate activity of a group of disparate people looking at a graph and making decisions based on a target function communicated to them by their superiors? What does this have to do with randomness or stationary randomness? What does this random stationarity have to do with our well-known cycles, in which cyclicality does not and cannot exist, because it will no longer be a random process? What does one have to do with the other?
Do you mean the full period i.e. 2*pi as cyclic?
She's weakly stationary, she's wide stationary, she's just stationary. >> so?
No, it's not "the same": https://en.wikipedia.org/wiki/Stationary_process. And for AlexEro, why don't you look it up if you're too lazy to google it?