Machine learning in trading: theory, models, practice and algo-trading - page 1596

[Aleksey Nikolayev]

Boris:

Can you answer questions without asking counter questions?

Learn the basics of the matstat. Then thoroughly understand which hypothesis in the Dickey-Fuller test is null and which is alternative.

[Boris]

Aleksey Nikolayev:

Learn the basics of the matstat. Then thoroughly understand which hypothesis in the Dickey-Fuller test is null and which is alternative.

So you can't.

thanks

[Aleksey Nikolayev]

dictum sapienti sat est

[Boris]

You remind of a character from the beginning of the thread, who is distinguished by the fact that he starts insulting his interlocutor if he can't answer a straightforward question

and, of course, thinks he's smarter than everyone else

Do you want to try on your certificates/diplomas?

[Aleksey Mavrin]

Maxim Dmitrievsky:

I prefer to consider stationarity from the significant lag of the increments, as in the last article. For example, lag ~24 for hourly increments is robust. Then there is no uncertainty in the choice of window.

Isn't that the ratio/difference of the price to what it was 24 hours ago? In fact it is almost the same as the analysis of the days?

[Maxim Dmitrievsky]

Aleksey Mavrin:

Isn't that the ratio/difference in price to what it was 24 hours ago? In fact, is it almost the same as the analysis of the days?

The article describes it in detail. There are net dependencies of specific hours on their lagged values that last for years. For the daily rates, you can look for others. In short, the quotes are full of dependencies. The only obstacle is changing the sign of the average increments.

[Boris]

Maxim Dmitrievsky:
The article is detailed. There are pure dependencies of specific clocks on their lagged values, which last for years. For the days you can look for other ones. In short, the quotes are full of dependencies.

So you have the Grail in your hands.

[Maxim Dmitrievsky]

Boris:
So you have the Grail in your hands.

Almost there, just need to finish the modes

[Boris]

Maxim Dmitrievsky:
Almost there, just have to finish the modes

success!!!

[Boris]

Maxim Dmitrievsky:
The article is detailed. There are pure dependencies of specific clocks on their lagged values, which last for years. For the days, you can look for others. In short, the quotes are full of dependencies. The only obstacle is changing the sign of the average increments.

If we didn't need to guess the sign of increments, we would have the Grail in our pockets a long time ago.

The Grail has long been known on the modulus of gradients.