Machine learning in trading: theory, models, practice and algo-trading - page 3736

 
Maxim Dmitrievsky #:
Don't confuse marginalised people with no education (no understanding of the subject) with researchers :)
And there was such a marginalised Darwin. We still can't get rid of him.
 
Inquiring #:
If anything, Galileo was a marginalist too.
Galileo relied on theories known since antiquity.
[Deleted]  
Inquiring #:
And then there was that marginalised Darwin. We're still trying to figure it out.
No, he was a follower of the theory of evolution.
You don't call scientists marginalised, it's a different term.
 
Inquiring #:
And then there was that marginalised Darwin. We still can't get rid of him.
It would be more correct to remember such a charismatic marginal as Lysenko, because of whom the country still lags behind in genetics and therefore is almost completely non-self-sufficient in seed production.
 
Maxim Dmitrievsky #:
No, he was a follower of the theory of evolution.
No need to call scientists marginalised, it's a different concept.
Darwin was the founder, not a follower of the theory of evolution. And he wasn't a scientist, he was a traveller. Do some reading somewhere, don't mislead the populace.
 
Yuriy Bykov #:

Imagine that there has long been a way to make it so that the models don't care about the magnitude of the number, and the only thing that matters is the fact that the numbers are different.

So we should start from that.

And develop the idea of graph formatting. That is, to build structure and feed it.

Why do neural networks need the last 20 values of prices, price increments, indicators and so on? When the power influence comes from different scales.

Well, this is "informativeness" and "uninformativeness", relevance and irrelevance.

Moreover, uninformativeness (the oldest prices of one price series) can be destructive and simply destructive for the TS, as a result of which the model will look for patterns that are obviously unworkable.

That's why I gave the example of a 10 years old microstructure, which has no effect on the current price impulses. But the scale relationship is obvious. The informativeness and relevance (highlighting recent areas of structures of different scales) is a subject to be investigated both with one's own brain and with the help of MO.

As a result, as soon as there is an adequate markup of the price chart, as soon as incomplete (active) structures are highlighted - then this "naked knowledge", like a rough diamond, can be transferred to the machine to be honed.



What is the difference between a real trader and a forum MOSHnik?

A trader's stop-loss is structural.

A MOSHnik's stop-loss is undesirable.
[Deleted]  
Inquiring #:
Darwin was the founder, not a follower of the theory of evolution. And he wasn't a scientist, he was a traveller. Read it somewhere, don't mislead people.
No, Lamarck was the founder, and the ideas had been in the air for a long time, in one form or another
 
Maxim Dmitrievsky #:
No, he was a follower of the theory of evolution.
No need to call scientists marginalised, it's a different concept.
Marginal is a general term meaning that a person is not on the mainstream (i.e. not running in a rut in a closed circle). Marginal scientists include, for example, David Bohm, Rupert Sheldrake. They are real scientists, with degrees, but they are marginalised.
 
Inquiring #:
Darwin was the founder, not a follower of the theory of evolution.

No, of course not. The theory of evolution existed safely long before Darwin. It had its beginnings in the ancient natural philosophers.

Darwin proposed and justified natural selection as the basic mechanism of evolution.

 
Inquiring #:
Marginal is a general term meaning that a person is not on the main path of movement (i.e. does not run in a rut in a closed circle). Marginalised scientists include, for example, David Bohm, Rupert Sheldrake. They are real scientists, with degrees, but they are marginalised.
No, they're not. Marginal scientists are Fomenko with his chronology in historical science (in geometry he is a normal scientist).