Machine learning in trading: theory, models, practice and algo-trading - page 1338

 
Maxim Dmitrievsky:

yes

That's good, thanks for the guidance.

What can I say, if you digest these formulas, you have a clear advantage over me!

The words there are beautiful, partly understandable, but I need implementation, and for this you need to understand all these formulas, I do not even know how to read, not to mention the understanding.

Have you implemented everything described there that you found interesting? Just as an option, if you understand what we are talking about and can control the correctness of the calculations, then you can make some class to work with the described methods, including through freelancing. In general, I propose cooperation on mutually beneficial terms.

 
Hello, I can't get this system to work. What should I do?
 
Audrius Liudvinaitis:
Hello, I can't get this system to work. What do I have to do?

To go to work myself - for the system...

 
Maxim Dmitrievsky:

There is nothing complicated there, you just need to spend time to understand it.

I don't know how I can help, because I have a different approach and keep developing it, so I probably won't even have time to write something else.

there are examples of brute force implementation in kodobase, but i don't use it that way now

even if you get some theoretical aspects from it, it's already useful

Examples of brute force what is in kodobase - I do not understand.

Well, and I alone can not pore over this reading, which would be an application code to write and check the feasibility of these actions. Isn't it possible to create new predictors on the basis of existing ones in Python by the method described in the book?

 
Aleksey Vyazmikin:

Example of brute-force what is in the codebase - I do not understand.

Well, I alone can't read this reading to write the application code and check the feasibility of these actions. Is it possible to create new predictors on the basis of existing ones by the method described in the book not implemented in Python?

Maxim wrote -"I don't know how I can help, because my approach is different and I keep developing it", i.e. material for general development.

 
Yuriy Asaulenko:

Maxim wrote:"I don't know how I can help, because my approach is different and I continue to develop it," i.e. the material is for general development.

I was surprised, because I assumed that he tested the method described, since he speaks about a significant improvement in results when using it.

I have already met similar ideas - the implementation is interesting...
 
Aleksey Vyazmikin:

I was surprised, because I assumed that he had tried the method described, since he says there is a significant improvement in results when using it.

I have already met similar ideas - the implementation is interesting...

We all have.) The devil is in the details.

 
Yuriy Asaulenko:

We've all met). The devil is in the details.

That's what I mean, and the details are in the tricky formulas...

 
Maxim Dmitrievsky:

of features through the analog of the Kolmogorov-Gabor polynomial

This is one of the simplest books, the dawn of the MO, so to speak, so it is easy to understand. But there is a lot of useful information on auto-construction and feature selection, i.e. bruteforce. This is the topic we touched on.

Modern books on MoD are much harder to read

If you carefully examine their site you will find examples of codes, google python, I myself have written since it is simple. I've been chatting with an indian about it for a few months now, I don't think you've been here very long.

I can't seem to use search engine, I can't find MQL script/library which you mentioned. Can you give me a link to it?

(Although the post has already been corrected and all references have been removed)

I'll study the book carefully if possible and may have some ideas.

 
Maxim Dmitrievsky:

did I tweak it?

RL algorithms library in my codes

why do you need codes if you don't understand how it works, don't want to learn

it's the same with catbust in the form of obvious experimental results

I talked to one moron - a good programmer but doesn't understand what he's doing, I gave him the codes and he said, oh, it's very simple, let's try this and that, let's try that, and what's this and that, let's try removing it... I told him to read a book, or you'll end up poking around and poking around. In the end he didn't do anything.

The main thing is to study something that somehow works, so I'm curious to see if it makes sense. I don't understand, what functionality from your library is offered to use?

A successful person is not someone who is smart, but someone who surrounds himself with smart people. I am a bad programmer, a bad mathematician, but a good dreamer and experimenter - my skills are not enough for success, so I am looking for people with whom we will have a mutual compensation of shortcomings at the expense of sharing virtues.

The results of the experiment are not obvious, look carefully at the post where I painted the sampling periods - the results are very unusual!