ERUUSD spectra - is this proof of non-stationarity? - page 10

 
Urain >> :

>> And for the period not to be longer than the window.

>> Can you tell us more about it? Or a link to a forum thread.
 
viknau >> :
Would you mind telling us how you calculated the spectral density from the price chart? Is there a special mathematical tool in MQL or did you convert a quote file and then calculate it in MathCad.


There is no dedicated apparatus in MQL4. Which spectral analysis method are you interested in?

I do not recommend using Matcad or Matlab. There is a lot of mess in digital signal processing packages.

Until you understand how and what they count you will not be able to use them properly.

And when you do - Matcad or Matlab will not be needed, you may write everything perfectly well in MQL.

It will be faster, more reliable and, most importantly, it will calculate what you need.

Matcad or Matlab can be useful for double-checking and solving arguments.

 
viknau писал(а) >>
Could you please tell us how the spectral density from the price chart was calculated? Is there a special mathematical apparatus in MQL or converted the quotation file and then calculated it in MathCad.

Taken from the Digital Method Generator, an analogue of the more famous Kravchuk system

 
Urain писал(а) >>

So take it from the 2nd extremum to 0. (The second extremum will not be redrawn for sure).

Only the ZZ parameters should be adjusted so that the window does not change every couple of bars, so that there is at least some statistics.

Looked at the Internet - the problem of windows is common. Dynamic process because the value of cyclic fluctuates.

 
begemot61 >> :
Can you elaborate on that? Or a link to a forum thread.

Read it here.

 
faa1947 >> :

Looked on the internet - the window problem is common. The reason a dynamic process is dynamic is because the cyclicality value floats.

Yes,a discussion broke out on a forum several times but it is still there.

I think that the theory is so indisputable that it must be taken into account and tried to be circumvented.

You have to find the condition that the parameters that go beyond the theory are not important and can be neglected.

Here is in brief the general direction of the search.

 
begemot61 >> :

It's funny...

You shouldn't like cats. You just don't know how to cook them!

You don't even have to cook them, they're not there and that's it, otherwise everyone would be Soros, Buffets.

 
registred >> :

You don't even have to cook, they're not there, otherwise everyone would be Soros, Buffett.

Naive... Do you think that having a "grail" will make you rich quickly and immeasurably?

In the best-case scenario, you'll be locked up for being an insider. No one will care that there was no inside information. At worst, they won't find your body.

You got to be careful with this kind of thing. And get rich slowly.

I'm not talking about domestic kitchens. Everything is simpler here. It's all about the rules.)

 
Zhunko >> :

Naive... Do you think that having a "grail" will make you rich quickly and immeasurably?

In the best-case scenario, you'll be locked up for being an insider. No one will care that there was no inside information. At worst, they won't find your body.

You got to be careful with this kind of thing. And get rich slowly.

This is not about domestic kitchens. Everything is simpler here. It's all by the book :-))

Getting rich beyond measure is like robbing a bank, I tell you. It's as utopian as the grail of stable frequencies.)

 
registred >> :

Getting immensely rich is like robbing a bank, I tell you. It's as utopian as the grail with sustained frequencies:)

There are no sustained frequencies, of course. But does it matter?