Discussion of article "Metamodels in machine learning and trading: Original timing of trading orders" - page 3
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
We're not here either, according to Advaita. So I suggest you take a brandy and relax. These people are ignorant if they think they know anything.
They know how to take your money.
They know how to take your money.
A field is an abstract mathematical entity. That doesn't mean it exists.
The question of existence is already a philosophical question.
Let's get to the heart of the matter. What should I add to the code to make it better?
"Since in the process of developing trading systems (including those using machine learning) the researcher deals with uncertainty, it is impossible to strictly formalise what is being searched for in the end. These are some more or less stable dependencies in a multidimensional space..."
If you don't know what you are looking for, what is there to find?
Maybe you should first think about what to look for, what kind of dependencies, - on whom or on what, to understand how things work in general?
"Since in the process of developing trading systems (including those using machine learning) the researcher deals with uncertainty, it is impossible to strictly formalise what is sought in the end. These are some more or less stable dependencies in a multidimensional space..."
If you don't know what you are looking for, what can you find?
Maybe you should first think about what to look for, what kind of dependencies - on whom or on what, to understand how things work in general?
I guess it's something similar, only here it's a tandem of models
The model of the environment depends on its complexity. The complexity of the model depends on the number of processes that influence the result. The complexity of building/guessing/finding the right model depends directly on the ability to account for these processes.
In my opinion the metamodel should be more complex. So far it's too simple solutions that work in areas that are similar to the learning areas. But so far the power to build model logic is lacking in my opinion.
The model of the environment depends on its complexity. The complexity of the model depends on the number of processes that influence the result. The complexity of building/guessing/finding the right model depends directly on the ability to account for these processes.
In my opinion the metamodel should be more complex. So far it's too simple solutions that work in areas that are similar to the learning areas. But so far the power to build model logics is not enough in my opinion.
No, what is proposed here is a graphical mower that searches for something and even sometimes finds something. You can propose informative signs instead of increments - it will be constructive.
As I understood, the logic of your models is straightforward everywhere - if there was a rise yesterday, there will be a rise today, if the pattern worked last week, it will work this week.
In life it doesn't work like that."The time to be born and the time to die; the time to plant and the time to pluck out what has been planted". Yesterday the market makers drove the wave, "planted" the illusion of an upward move, and today they harvested the crop.
The logic is to look for impulses and reaction to impulses. "There is no point in blowing when no one is blowing back". If on the fifth wave of infusion - impulses do not get a response from the takers, what is the point of leading the trend? It gets reversed in the other direction.
It's just physics.
No, what is proposed here is a graphical mower that searches for something and even sometimes finds something. You can suggest informative signs instead of increments - that would be constructive.
Some recommendations for trading https://drive.google.com/file/d/1RFTdBDPVZA61fYBfMeB_oX9FDvULbA88/view?usp=sharing
The model of the environment depends on its complexity. The complexity of the model depends on the number of processes that influence the result. The complexity of building/guessing/finding the right model depends directly on the ability to account for these processes.
In my opinion the metamodel should be more complex. So far it's too simple solutions that work in areas that are similar to the learning areas. But so far the power to build model logic is lacking in my opinion.
Complexity is being dealt with by synergetics. There is very little applied research, precisely because of complexity.