Machine learning in trading: theory, models, practice and algo-trading - page 3153

 
Maxim Dmitrievsky #:
Level up, at least writing code and a basic understanding of algorithms, as written about in books. Then there will be something to talk about. Otherwise, cleverness (an attempt to imitate formal scientific style), with grammatical and other errors, causes only a smile :).

It's funny to hear that. Oh well, anyway I'm glad that at least you are smiling. There is no sense to talk to arrogant people - instead of arguments they have theses about their greatness, which is not confirmed by the result.

I will say that psychologists have very interesting things about modelling - there is even something similar to this UpLift.

 
Aleksey Vyazmikin #:

That's funny to hear. Anyway, I'm glad you're at least smiling. There is no point in talking to arrogant people - instead of arguments they have theses about their greatness, which is not confirmed by the result.

I will say that psychologists have very interesting things about modelling - there is even something similar to this UpLift.

I've already chewed it up and put it in my mouth, you don't even want to swallow it. Maybe the psychological background isn't right.

 
Maxim Dmitrievsky #:

I've chewed it up and put it in your mouth, you don't even want to swallow it. Probably the psychological background is not right. I am not a psychologist, so I cannot comment on your other theses.

If there are any other diagnoses or name-calling, you still have time to voice them today, while at the computer :)

#62

You were asked a specific question about the validity of splitting the sample in two, specifically at what point you suggest splitting the sample and why. You reported that it doesn't matter, which makes me doubtful. The fact that you got some effect is not consistent with the theory, which is why I wanted to hear your logic. It doesn't take much intelligence to just copy and paste code - you have to think what you are doing. Instead of answering, you switched to discussion of personal characteristics and started attributing to me attributes that have nothing to do with reality, and you need them only to demonstrate your uniqueness.

You don't have to write here - nobody will cry.

I myself do not see an opportunity to spend a lot of time on this branch of the forum, as it is impossible to communicate constructively here, and you prove it in essence.

 
Aleksey Vyazmikin #:

You were asked a specific question about the validity of splitting the sample into two parts, namely at what point you propose to split the sample and why. You said that it was not important, which makes me doubtful. The fact that you got some effect is not consistent with the theory, which is why I wanted to hear your logic. It doesn't take much intelligence to just copy and paste code - you need to think what you are doing. Instead of answering, you switched to discussing personal characteristics and started attributing to me attributes that have nothing to do with reality, and you need them only to demonstrate your uniqueness.

You don't have to write here - no one will cry.

I myself do not see an opportunity to spend a lot of time on this branch of the forum, as it is not possible to communicate constructively here, and you prove it in essence.

I specifically answered your specific question - in any thread you want. It is your personal business. It's even strange why I was asked. It had nothing to do with the topic I was interested in.

I wrote from the very beginning - if you don't bother to read at least diagonally, I won't put a dummy in your mouth. Apparently you're used to that.

And the whole thread will be made up of posts like this, thanks to these psychopaths. One thing after another, in a circle. I think they call them cycloids.
 
I didn't even realise that a suggestion, already quite long ago, to discuss the topic of causal inference, would cause such a far cry from it

😁😁😁
 
I'm gonna bust a gut soon. Farewell, my flock, you are now on free bread.
 
Maxim Dmitrievsky #:
I'm gonna bust a gut soon. Farewell, my flock, you are now on free bread.

I hope you'll get down to business and demonstrate your outstanding achievements in the marketplace instead of rubbish?

 
СанСаныч Фоменко #:

I hope you'll get busy and showcase your outstanding achievements in the marketplace in lieu of rubbish?

Why call masterpieces rubbish? I offered you free of charge, including on your supermarks, as well as Alexey. But I was sent to the kodobazu. And now he, the complainer, is running around calling me more names.

If you and he really think that you have cool signs, the algorithm will learn on them very well.

If you just suffer from bullshit, then keep on suffering.

You don't let me out of the subject, do you?

 
How can you pick signs for so long and not have any in the end? At least the bad ones should be there, forgotten and not needed by anyone.

There is everything, including ready-made bots on the way out. Training a couple of minutes on your super signs. On the output bot source for mt5 and 4.

Top signs according to my version: volatility, autocorrelation in a sliding window with different periods, the same with autoregression, deviations not in the form of increments from the price, but from the global / local trend line, with tracking of its change.
 
Aleksey Nikolayev #:

Anarticle with an approach similar to the one promoted by Aleksey Vyazmikin. Instead of a classification tree, a "difference tree" is constructed, in which each leaf corresponds to a different probability of an event (for example, fire frequency). In essence, this is some variant of clustering.

I will say at once that I am not ready to recount the article in detail, as I have only glanced through it in passing.

Thank you for the article!

Translated it, read it.

Can you help me translate the algorithm in the form of formulas into a more understandable language? And I will try to write the code of this method of building a tree.