Machine learning in trading: theory, models, practice and algo-trading - page 1349
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
This is a theorem about the spectral representation of the correlation functions of stationary processes.
It does not give answers to the questions posed.
Get into the essence of the questions asked. And try again to answer them.
This is the basis for constructing the spectral representation of a stationary process (p. 248).
Point out, if you can, other places in this book related to the concept of the spectrum.
A spectral representation of a stationary process is based on this (p. 248).
Point out, if you can, other places in this book related to the concept of the spectrum.
Exactly!
And you act Jesuit-like, asking"You do not agree with Korolyuk?"But it has nothing to do with Korolyuk. It's just your speculation, and an attempt to hide behind someone else's authority. (Cicero's speech, to use your own words)
Korolyuk nowhere states thatthe concept of spectrum is defined only for a stationary process.
Nowhere does Korolyuk state that for a non-stationary process the concept of spectrum does not exist.
I recommend that you master spectral analysis, you will learn many interesting things about spectra that you are not aware of.
Exactly!
And you act in the Jesuit way, asking, "Do you disagree with Korolyuk?"But it has nothing to do with Korolyuk. This is just your speculation, and an attempt to hide behind someone else's authority.
Nowhere does Korolyuk state thatthe concept of spectrum is defined only for a stationary process.
Nowhere does Korolyuk state that for non-stationary process the concept of spectrum does not exist.
I recommend that you master spectral analysis, because you will learn a lot of interesting things about spectra that you do not know at the moment.
Let me try to explain in simple terms. A spectrum can be defined only if the ACF depends on one variable and not on two variables as in the general case. This is only true for stationary processes in the broad sense (in fact, this is part of their definition).
With radio amateurs, the ACF always depends only on one variable, because they do not distinguish between the concept of ACF and its sampling version. That is, as I wrote earlier, they mix the concepts of a random process and its realization. Perhaps this is not critical in their field, but it is wrong to do so for prices.
Let me try to explain in simple terms. A spectrum can only be defined if the ACF depends on one variable, not two as in the general case. This is true only for stationary processes in the broad sense (in fact, this is part of their definition).
With radio amateurs, the ACF always depends only on one variable, because they do not distinguish between the concept of ACF and its sampling version. That is, as I wrote earlier, they mix the concepts of a random process and its realization. This may not be critical in their field, but it is wrong to do so for prices.
To repeat :
Forum on trading, automated trading systems and trading strategy testing
Machine Learning in Trading: Theory and Practice (Trading and Beyond)
Oleg avtomat, 2019.02.19 06:53
Why are you so hung up on the theorist... Probably because your knowledge is limited to that.
I see that you are not familiar with the concepts of:
1) generalized spectra;
2) instantaneous spectra;
3) spectral structure of an unsteady process.
You do not know how nonstationary processes are analyzed. You don't even know how to approach it.
Look it up on the Internet, read it thoroughly. And don't say stupid things.
Take interest in pp.1, 2, 3, and you'll understand what your mistake is.
Alexei, you need to bring the situation to the point of absurdity - to open and maintain a "Theory" branch. Just "Theory," that's all. That would be a masterpiece.
If you want practice - do what Koldun asked to do and I ask to do: take a moving time period, make a multiplication - how the probability density of tick (or second) increments behaves within this period, calculate the kurtosis and asymmetry.
You will learn a lot and it will be a real case.
I repeat :
Take an interest in p.1, p.2, p.3 -- you'll see where your mistake lies.
he won't understand...
don't waste your energy
I repeat :
Take an interest in the listed p.1, p.2, p.3 -- you'll see where your mistake lies.
Ideas of quasi-stationarity, piecewise stationarity and the like can be useful by themselves, but related to spectral analysis they will only be harmful.
Ideas of quasi-stationarity, piecewise stationarity, and the like may be useful on their own, but related to spectral analysis they will only be harmful.
???
Harmful? Harmful to whom? or to what? Harmful to you? or harmful to spectral analysis? or harmful to science? or harmful to humanity?
Apparently, all the same, harmful for you, and only for you, because they put you in a stupor.
he won't understand...
Don't waste your energy.
Apparently he does...
.
Here's a simple, but very useful problem. Can you handle it?