Machine learning in trading: theory, models, practice and algo-trading - page 2445
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
covariation?
What with what?
More organic work with potentially infinite-dimensional objects.
Did you do something like this, or is it a theory?
Did you do something like this, or a theory?
Nonlinear regression of the vector-function type in the context of studying the distribution of TC returns with trailing and modification of the exit rule. I'm studying the possibility of applying the function-function type regression to avoid manual feature vector extraction from prices, but so far I haven't found anything interesting.
Nonlinear regression of the vector-function type in the context of studying the distribution of TC returns with trailing and modification of the exit rule. I am studying the possibility of application of regression of the function-function type in order to avoid manual extraction of a feature vector from prices, but so far I have not found anything interesting.
I counted 5 familiar prepositions and realized that the author was also taken aback by what he had written:-)
I counted 5 familiar prepositions, and realized that the creator of the phrase was also taken aback by what he had written:-)
brevity is our brother's sister)
Nonlinear regression of the vector-function type in the context of studying the distribution of TC returns with trailing and modification of the exit rule. I am studying the possibility of application of regression of the function-function type in order to avoid manual extraction of a feature vector from prices, but so far I have not found anything interesting.
I don't really understand it, it's not my topic. I don't think there's anything to be found with functional data in tabular form, I've already tested so many things, it's horrible...
I think this is the best tool for the market, but I need large computational power.
I don't really understand it anyway, it's not my topic... I don't think there's anything to do with functional data in tabular form, I've already tested so many things, it's horrible...
I think this is the best tool for the market, but it requires a lot of computing power.
We've all got our own thing going on here. Therefore I do not see much sense in presenting their approaches in detail, nor in trying to join forces of some kind. Unless some short reports about some relatively new methods.
Nonlinear regression of the vector-function type in the context of studying the distribution of TC returns with trailing and modification of the exit rule. I am exploring the possibility of using function-function type regression to avoid manual extraction of a feature vector from prices, but so far I have not found anything interesting.
How many modifications of input rules are there and are they specified or generated?
How many input rule modifications and are they set or generated?
We assume that we have some simple system (the main thing is that the entries are frequent enough, but not too frequent). With standard trailing, we enter and sit until it triggers. The modification lies in the fact that we see if at some points we cannot exit the trade or even reverse it. For example, when entering after the breakdown of some levels, it may make sense to go in the direction of the breakdown not very much and then reverse.
The rules of entry of the initial system are not touched, but some of them may disappear during modification. In principle, this is quite a standard approach.
We proceed from the presence of some simple system (the main thing is that the inputs are frequent enough, but not too frequent). With standard trailing, we enter and sit until it triggers. The modification lies in the fact that we see if at some points we cannot exit the trade or even reverse it. For example, when entering after the breakdown of some levels, it may make sense to go in the direction of the breakdown not very much and then reverse.
The rules of entry of the initial system are not touched, but some of them may disappear during modification. In principle, this is a fairly standard approach.
That is, reasonably logical input rules and their modification. There are thoughts floating around about generating rules. The rules are determined by logic and mathematics and the idea of processing some variant of the range. I am still thinking about exiting a trade. The trawl is just too simple. It does not take into account velocity, momentum... I'm still thinking manually)))