Machine learning in trading: theory, models, practice and algo-trading - page 998
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 strange opinion.) Why should it be so?
These are the graphs obtained after exponential thinning of the tick BP. As we see, the dispersion is nearly a constant both day and night.
These are the graphs obtained after exponential thinning of the tick BP. As you can see, the variance is practically a constant both day and night.
Well, I don't know. Imho, there is no need to make such a mess. f(t)-R(f(t)) is already quite a stationary series.
What functions are these?
There is an opinion that with some "thinning" of the initial BP it will be possible to obtain a process that is as close to stationary as possible.
It seems to me that a simple counter-example would be a sharp change of trend by a trend in another direction (a top or a trough). How would thinning help here?
what are these functions?
As I understand it - BP itself minus its Fourier-image
It seems to me that a simple counter-example would be a sharp change in trend with a trend in the other direction (top or trough). How would thinning help here?
I'm working on that. The problem is not an easy one, but it is not the right solution to work with the initial BP.
I understand - BP itself minus its Fourier-image
I can't get that into my head...
like two camels were flying...
I'm going for a smoke break...
It won't do any harm, either.) If we exclude the trend component, the market is quite stationary. It is easy to check it in Excel.
What happened to the ARCH effects? Of which there are over a hundred? And of which there is no end in sight?
As I understand it, BP itself minus its Fourier image
I am working on it. The task is not easy, but just to work with the initial BP I think the wrong solution.
Not to notice non-stationarity is also impossible. A possible standard option is to represent the initial series as a superposition of fast stationary and slow non-stationary processes.