Machine learning in trading: theory, models, practice and algo-trading - page 3078

 
Aleksey Vyazmikin #:

What exactly isn't possible?

I was just helping,

 
lynxntech #:

I was just helping out,

Have you had a negative experience? Here everyone does a common endeavour and almost immediately gets a result - a knowledge base is accumulated for everyone. So you can leave at any time with the result of your participation.

 
Aleksey Vyazmikin #:

Do you have a negative experience? Here everyone does a common cause and almost immediately gets a result - a knowledge base is accumulated for everyone. I.e. you can leave at any moment with the result of participation.

I'm just tired.

10 thousand pages that you have to not only read, but also delve into, and bear some responsibility.

not only here.

 
lynxntech #:

I'm just tired.

10 thousand pages in which it is necessary not only to read, but also to delve into, and a certain responsibility to bear.

not just here.

Your train of thought is slipping...

Are you afraid of making a mistake?

 
Aleksey Vyazmikin #:

I propose to join efforts to search for useful information in the code base, namely interesting indicators.

The task is time-consuming, but there is a probability that something underestimated will be found.

...

Why take code from CodeBase? It is possible that a completed idea has been published in indicators, but there is a chance that only a draft of the main idea has been published. In this case, the planned amount of work becomes disproportionately large.

 
Aleksey Vyazmikin #:

So nobody liked my idea?

Everyone thinks they are smarter than others and are sure that there can't be a useful idea in indicators?

Or they are simply not interested in getting useful information together with them? Like not for themselves or for others?

Or do you have 10 lives in reserve and hope to manage everything on your own?

No, it's just unpleasant to read when they write through the word with mistakes. And cheap NLP doesn't go over well.)

 
I like the idea of joining forces, I've been kicking around this idea for a long time....

I see it as creating a publicly accessible knowledge base (Google to help), something like sql, nosql databases, but smarter. In this KB can be loaded as rules and actions or just fuzzy entities, as well as add a graphical interface for input output information....

Well, certainly not the enumeration of identical indicators, it is difficult to think of a more dull occupation.
 

Some expert says Boosted Arima is better than deep learning models https://rdrr.io/cran/modeltime/man/arima_boost.html.R code includes the exponential smoothing.Does anyone how to implement Boosted Arima code in R code or Python?

Some experts say Boosted Arima is better than deep learning models https://rdrr.io/cran/modeltime/man/arima_boost.html. R code includes exponential smoothing. Does anyone know how to implement Boosted Arima code in R code or Python?
 
nevar #:

Some expert says Boosted Arima is better than deep learning models https://rdrr.io/cran/modeltime/man/arima_boost.html.R code includes the exponential smoothing.Does anyone how to implement Boosted Arima code in R code or Python?

Some experts say Boosted Arima is better than deep learning models https://rdrr.io/cran/modeltime/man/arima_boost.html. R code includes exponential smoothing. Does anyone know how to implement Boosted Arima code in R code or Python code?

You provide a link to an example on using the arima_boost function in R and ask - Does anyone know how to implement Boosted Arima code in R or Python code?

How is that even possible?

 
Lilita Bogachkova #:

Why take code from CodeBase? It is possible that a completed idea was published in the indicators, but there is a possibility that only a draft of the main idea was published. In this case, the planned scope of work becomes disproportionately large.

And how do you measure feasibility?

Even if there are 5% of useful indicators out there, out of those selected for the study - that's a great result for me!