Machine learning in trading: theory, models, practice and algo-trading - page 1359

 
Gianni:

Normal 100-500 trades a day

What's the profitability? Because spreads/slippage/commission, you know, want a lot of money.

 
Aleksey Vyazmikin:

And what is the profitability? Because spreads/slippage/commission, you know, they want a lot of money.

I would say that the profitability is very high, but it's kind of hard to improvise with DTs.

They have high profitability, but it's kinda hard to improvise with DCs.

 
Aleksey Vyazmikin:

I watched the video (Saveliev S.V.). All is normal, in the plan globally nothing new, but denial of compensation by the brain of the damaged fields (zones) has caused bewilderment, this phenomenon was studied even by Luria A.R. on concrete people, and besides he in the beginning confirms that neural connections are formed constantly, as well as all accompanying tissues. In general, a man with his own position, which was formed long ago and is now monetized. Talking about psychology out of his mouth seemed to be a bit sweeping. But I would read the propagandized book "Morphology of Consciousness" for a small fee, I just wonder if there is something new there...

If you are interested in neuropsychology, read the textbook"Neuropsychology: Textbook for High Schools. Chomskaya E.

you're always looking at something you don't understand and making conclusions you don't understand.

about the brain compensating for deficiencies, hemorrhages and injuries to the brain itself he said with the example of vision

About the rest in the same way, I will not answer, I am not interested in any non-existent neuropsychology and everything with the prefix "psycho".

 
Gianni:

channelers mainly on limiters, market makers, "grids".

Profitability is very high, but it's kinda hard to improvise on DT

I thought you were bragging about yourself and there are specifics.... There is a limit on limiters on the exchange (Moex) and then there is an additional commission if anything...

 
Everyone here is basically predicting direction, which is almost unrealistic, but in the case of channels, you can predict less noisy time series properties.
 
Maxim Dmitrievsky:

You're always looking at something you don't understand and jumping to conclusions you don't understand.

About brain compensation of defects, hemorrhages and traumas of the brain itself he said on the example of eyesight

About everything else exactly the same, I will not answer, I am not interested in any non-existent neuropsychology, as well as everything with the prefix "psycho"

That's the thing, he's confused about the statement - you don't have to look with your mouth open ;)

If even neuropsychology isn't a science to you... It's like learning the structure of microcircuits without understanding the logic of their work.

The example of vision is compensation for physiology, not brain function due to trauma.
 
Zhenya:
Everyone here is mainly predicting direction, which is almost unrealistic, but in the case of channels, you can predict less noisy properties of time series.

What do you mean by less noisy time series properties? I'm ready to predict anything, the main thing here is the result :)

 
Aleksey Vyazmikin:

I thought you were bragging about yourself and there are specifics.... There is a limit on limiters on the exchange (Moex), and then there is an additional commission, if anything...

As far as I remember, you can pay $600 per month and make a hit every nanosecond.

 
Aleksey Vyazmikin:

That's the thing, he's confused - you don't have to watch with your mouth open ;)

If even neuropsychology isn't a science to you... it's like studying the structure of microcircuits without understanding the logic of how they work.

I think you're confused about something, but not Mr. Saveliev, who devoted his whole life to studying the brain.

i downloaded it for general education, not for discussion with those who know nothing

 
Aleksey Vyazmikin:

And how do you measure the confirmation of the movement, and especially the development potential? Well let's say the potential can be determined by the presence of some kind of conditional resistance levels / breakpoints - even if the same strike next to it. I don't understand what your model is responsible for - at me it's just a x-point movement forecast with CatBoost, almost a regression.

The model I showed you the other day is just the first experiment. What will it add up to? - I have no idea.

Okay, I also have just a movement forecast. The old model just shows that the transaction may be profitable, without specifying what and how much - classification. Showed the pictures a year and a half ago.

Potential - well, yes, there is a lot to strive for). Confirmation - look at the speed and volumes - the process must evolve. The dynamics is different from the usual fluctuations.

Reason: