Machine learning in trading: theory, models, practice and algo-trading - page 2326
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
In my opinion, the main problem lies in the choice of approach to describing the behavior of the largest players in the market - states. They (1) strongly influence the market, (2) their behavior changes significantly over time, (3) their goals for action in the market often lie outside the market itself and are poorly known to us, (4) there are many states and they can interact with each other in very different (for the market) ways. Mathematically, the result is a complex, unsteady, and unclosed system.
The problem is not that you can't come up with a model for such a system, but that you can come up with too many different ones and probably even contrary to each other in conclusions)
Fractal behavior of players on different scales) may help)
this is the other side of the approach from the beginning. RCTs in medicine for anything and everything by the way killed medical techniques not reproducible placebos)))
The task of behavior or condition is to find causation)
which side of the approach originally
Well, the obvious thing is that two or three sine waves can't even approximate the price properly over a long period of time. Okay, don't you agree?
No I don't, you're missing the point again...
I see the transducer as a complex nonlinear structure, most likely a neural network cascade
input price --- converter --- sine wave.
For illustration, I showed an example with one sinusoid, but there may be more. At input a price, at output a sinusoid with the same parameters except phase
If the network learns how to do this, then there is no need to do anything else....
I have an idea - a market model through a network of players.
It roughly looks like this (I will try to look at the MO):
.......
....
The idea of the model is well-known, complex, with unclear goals for modeling.
Enthusiasts have not yet reached, digest all sorts of GPT-3 and other breakthrough stuff, and maybe someone will reach some of the directions of development will indicate in this.
Well, no one forbids to believe, but it's better to dilute theory with practice.
4) Let's calculate the profit
which side of the approach initially
The causes (according to the treatment article) are known, but the effects, or effects of the various treatments are not known, and the challenge is to find the best treatment and confirm the effectiveness.
We don't have known causes because there are too many of them....
Modeling is sort of supposed to give some kind of relationship of causes and their effect on price. Or vice versa, according to the behavior of the price it will be possible to identify the causes affecting the players, and then their actions on the price.
4) Let's calculate the profit
my favorite oscilloscope))) And what beautiful pictures could be drawn)
The causes (according to the treatment article) are known, but the effects, or effects of the different treatments are not known, and the challenge is to find the best treatment and confirm the effectiveness.
We don't have known causes because there are too many of them....
Modeling is sort of supposed to give some kind of relationship of causes and their effect on price. Or vice versa, by the behavior of the price it would be possible to isolate the causes affecting the players, and further their actions on the price.
Ah, well, I guess. I will do it later at my leisure.
Oh, I guess so. I'll do it later at my leisure.
Not today's task. It will be a little early.)
I do not see any obstacles