Machine learning in trading: theory, models, practice and algo-trading - page 2680

New comment
 
mytarmailS #:

Well, from that position, of course it's stupid.

Well, we've been dumb for two days without reading the definitions.
 
Maxim Dmitrievsky #:
h ttps://ru.m.wik ipedia.org/wiki/Сигнал
And I'm from the position of signal definition

It turns out there's some guy sitting on the other side sending you messages via stock quotes, which you filter to decipher it. Stupid? Yeah, it is.

Two blokes!) The second one adds noise to the messages from the first one)

 
Aleksey Nikolayev #:

Two blokes!) The second one adds noise to the messages from the first one)

Exactly )))
 
Maxim Dmitrievsky #:
We've been at this for two days without reading the definitions.

That's bullshit. I read the definitions in the book.

 
mytarmailS #:

That's bullshit, I've read the definitions in the book.

The added irony is that the information in a signal, even if there is any, does not necessarily tell you the future of that signal.

And you're confusing a signal with waves, which have amplitude, etc. They are the carrier of the signal and may carry nothing at all. If you decompose a graph into waves, you can't read a signal about the future of the system. Because you don't have a clue, even if the quotes hypothetically carried the signal.
 
Maxim Dmitrievsky #:
The additional irony is that the information in a signal, even if there is any, does not have to say anything about the future of that very signal

And you're confusing a signal with waves, which have amplitude, etc. They are the carrier of the signal and may carry nothing at all. If you decompose a graph into waves, you can't read a signal about the future of the system. Because you don't have a clue, even if the quotes were hypothetically carrying a signal.
I gave the definition of a signal, I certainly never confused it with waves
 
mytarmailS #:
I gave the definition of a signal, I certainly never confused it with waves
No, you're just extrapolating waves (aka components of a time series), judging by the pictures. It doesn't work if there is no seasonality and other cycles. There's no signal at all.
 
Maxim Dmitrievsky #:
No, you are just extrapolating waves (aka components of the time series), judging by the pictures. It doesn't work if there is no seasonality and other cycles. There's no signal at all.

depending on which pictures you're talking about.

 
Interesting thought...
What if we create a sample data generator that will come up with new data rows and add them to the sample. Add such samples that will cause the error to fall off
 

Posted documentation on the new features of the built-in matrices and vectors:

Please go through the descriptions of the functionality.

If you can give us your comments and ideas, we will be very grateful.

1...267326742675267626772678267926802681268226832684268526862687...3529
New comment
Reason: