Machine learning in trading: theory, models, practice and algo-trading - page 3719

[[Deleted]]

Renat Akhtyamov #:

kind of classic

formulas are the same

correlation, cointegration, etc.

I don't see the difference

Maxim, can you point me in the right direction?

or did he multiply the tails by a parabola?

Another physical principle is described, not on the constancy of correlation, but on its sharp deviation from the norm. To find structural changes in delayed series.

[Aleksey Nikolayev]

Maxim Dmitrievsky #:

Stumbled across an interesting paper on determining autocausality in time series, seems to be better than hirst and entropy and Granger.

https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4841224

This Alejandro Rodriguez is frequent - in the Quantopian mailing list they promote his monograph about portfolios and also on the topic of causality.

A new star of quanta to replace Prado? )

[[Deleted]]

Aleksey Nikolayev #:

This Alejandro Rodriguez is frequent - in the Quantopian newsletter they promote his monograph about portfolios and also on the topic of causality.

A new quantum star to replace Prado? )

And they, after Prado's contest on the topic of cauzal over graphs, got in touch and now they PR each other :) Maybe it's one of the winners, can't remember.

[Aleksey Nikolayev]

While I'm just thinking about probabilistic MO, there's already an article coming out on the forum on the topic.

[[Deleted]]

Aleksey Nikolayev #:
While I'm just thinking about probabilistic MO, there's already an article coming out on the forum on the topic.

It remains to figure out how to make millions on it :)

[Aleksey Nikolayev]

Maxim Dmitrievsky #:
Just need to figure out how to make millions from it :)

Combine it with causal? )

[[Deleted]]

Aleksey Nikolayev #:
Combine with causal? )

I don't know if it will help when tests don't show persistence of patterns on long history. For the short term, it can.

[Mihail Marchukajtes]

I don't know about you, but I'm trying to make predictions using chatGPT. I am still at the very beginning of my work. But VERY interesting!!!!!

[[Deleted]]

Mihail Marchukajtes #:
I don't know about you, but I'm trying to make predictions using chatGPT. I am still at the very beginning of my work. But VERY interesting!!!!!

Then it is necessary to have web search enabled to pull up-to-date information. Imho, we need teamwork of specialists to configure all the features/queries.

[Mihail Marchukajtes]

Maxim Dmitrievsky #:
Then it is necessary to have web search enabled to pull up-to-date information. Imho, you need teamwork of specialists to configure all the chips/queries.

Not really. It has already pulled all the information I specified, proavda could not take everything I need from moex as it is not freely available. He wrote me scripts in python, too. But everything doesn't work because there is no access to the actual data. I'll send you the forecast for today, which he made.....