Machine learning in trading: theory, models, practice and algo-trading - page 2968

New comment
 
СанСаныч Фоменко #:

but an extremely unpleasant peculiarity came out: half of the prediction error, i.e. 10% of the error, gives about the same loss as all other positive entries.

because the target is incorrect and the problem statement itself

 
СанСаныч Фоменко #:
half of the prediction error, i.e. 10% error, gives about the same loss as all other positive inputs.

50/50 in the money

mytarmailS #:

because the target is incorrect and the problem statement itself is incorrect.

The problem is not the targeting, there are probably 10 people here who are running tests with MO and everyone has a different version of the targeting. But everyone has the same 50/50.
The problem is that the market is efficient, for those who don't have big money to move the market or at least insider knowledge about big money to follow them.
 
Maxim Dmitrievsky #:

At this point, perhaps we can safely enter an evolutionary dead end and make way for other normal species :D

Julia is coming up, for example. But it will be funny if some ChatGPT 6 will remove the need for such languages and you can use a normal language)

 
Forester #:

50/50 on the money.

What is that? If not the answer that the target is wrong.


As an example:
"I want to maximise profit in money, but I'm predicting a zigzag."
 
Aleksey Nikolayev #:

Julia is coming up, for example. But it will be funny if some ChatGPT 6 will remove the need for such languages and you can use a normal language).

We probably won't see that stage. And Julia is really very promising and well integrated into R. But while it is being intensively developed we can only experiment. But it's not for sure :)

 
Aleksey Nikolayev #:

Julia is coming, for example. But it will be funny if some ChatGPT 6 removes the need for such languages and you can use a normal language)

They will definitely not bypass python, there is no point. If you need speed, there are all sorts of accelerators for code pieces. But I can't imagine that a bunch of fans will run away from the benchmark of high-level languages. It was the same with GO, not many people ran to make MO in it.
And python is as simple and intuitive as possible.
R, with its cramped syntax, is not a competitor either. It seems to me that a stubby syntax somehow distorts the mind and spoils the mood.

That's why I initially chose it and didn't lose out. Although I'm not a programmer and it's all the same in essence.

 
Maxim Dmitrievsky #:
R, with its jerky syntax, doesn't appeal to many people either, it's not a competitor. It seems to me, that the crooked syntax somehow distorts the mind and spoils the mood.
That's why I initially chose it and didn't lose out. Although I'm not a programmer and it's all the same in essence.

but for some reason the whole branch is sitting on R, and doesn't want to touch python without need...

All the programmers I know who know both python and R, all of them choose R.


and to call something you don't know snarky is a sign of ignorance... and to judge what you don't know is also a sign of ignorance....

 
mytarmailS #:

but for some reason the whole branch is sitting on R and won't touch python unless it's needed.

All the programmers I know who know both python and R very well, all of them choose R as the only one


and to call something you don't know snarky is a sign of ignorance... and to judge what you don't know is also a sign of ignorance....

It's the opposite. No one uses R from the major implementers. And they're not the last people in IT. Although there's not much demand for Python there either, usually scala.

What ignorance? I originally downloaded R and python. I poked around in both and understood everything at once, without anyone even prompting me. On the contrary, an unbiased opinion. From R-studio in general left an unpleasant residue.
 
Maxim Dmitrievsky #:
The opposite situation. No one uses R from large implementers. And they're not the last people in IT.

because that's not what it's for.

Why doesn't anyone write websites in C++?

Each language has its own task, that's why there are so many of them.

Maxim Dmitrievsky #:

What ignorance? I initially downloaded R and python. I poked around there and there and understood everything at once, even without anyone prompting me. From R-studio in general unpleasant residue remained.

I just chose what was more familiar, more similar to mql, and that which was less familiar (not familiar) called cramped. And you should have studied...

 
mytarmailS #:

because that's not what it's for.

Why doesn't anyone write websites in C++?

I just chose what was more familiar, more similar to mql, and what was less familiar (not familiar) was called jerky. You should have learnt...

It's not hard to learn anything. But the rational grain suggested that there is no sense or advantages for their crafts.

I started writing code in Python at once, without almost studying it. Then a couple of basic libraries were needed and that was it.

There are no tasks where R wins. Statistics is a myth. Some people justify their choice in this way, like Sanych.
1...296129622963296429652966296729682969297029712972297329742975...3682
New comment