Machine learning in trading: theory, models, practice and algo-trading - page 1300
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
Okay, try it, tell me if you want to. And if possible, compare my selection and ML selection, if at all possible without too much work on my part.
Yes, I don't thoroughly understand how a catbust works, but I already have the knowledge and experience to operate it, and all that takes time, which leads to a comprehensive understanding. To take something from scratch and try to apply in the work, when there is no sufficient information, well, for me it is not comfortable. Even with Catbust I have to look and understand everything, translate it, and it's good that I have people, who understand me better, who I can ask technical questions about the code.
It's just not clear what to discuss in this vein - well, we found some combinations of features, ok... then maybe there's some fundamental reason why they and not others, and why some specifically targeted specifically for them
as a rule, these questions are impossible to answer.
It's just not clear what to discuss in this vein - well, they found some combination of features, ok... then maybe there is some fundamental reason why they and not others, and why some specifically targeted specifically for them
as a rule, it is impossible to answer these questions
Again I lost my train of thought. If we still speak about the selection of models, then all models have almost the same predictors (metrics, probability distribution, error balance, etc.), it's more difficult with the target ones. I mean that the topic is global and complex, and not only relevant to the MO in trading, so there is something to discuss, in my opinion.
Again I lost my train of thought. If we are still talking about the selection of models, then all models have almost the same predictors(metrics, probability distribution, error balance, etc. ), with the target it is more complicated, I mean that the topic is global and comprehensive, relevant not only for MO in trading, so there is something to discuss, in my opinion.
What are these predictors?
i don't really understand what you are doing.
what are these predictors?
This is different information about the model behavior on the sample and even about the model structure (for example, the number of splits and/or trees), it is quite good, at least for me.
I have attached information on the model I am collecting but there is also specialized information for trading.
A sample is formed from such data and the model can be trained to select the necessary models.
I do not understand what you are doing in general.
I do what I think I should do :)
This is different information about the behavior of the model on the sample and even about the structure of the model (for example, the number of splits and/or trees), it turns out pretty good, at least for me.
I have attached information about the model that I collect, but there is also specialized information purely for trading.
I mean they are not predictors but indicators for each model, because predictors are what is fed to the model input
So, you have a bunch of different indicators.
you need to make one weighted metric out of all useful indicators and use it to choose the best model on the automaton, by argmax
I mean they are not predictors but indicators of each model then maybe? because predictors are what is fed to the input of the model
So you have a bunch of different indicators, what should you do with this bunch?
you need to make a weighted metric out of all useful indicators and use it to choose the best model automatically, by argmax
Are not predictors the indicators influencing the target? What a play on words :)
The thing is that there is no formula for this "weighted metric" and it's up to the Defense Ministry to find it.
Are predictors not indicators that affect the target? What a play on words :)
That's the thing, there is no formula for this "weighted metric", and finding it is the task of the IO.
Your table shows WHAT? I see there model estimates, acuras and other things. What does this have to do with predictors?
Exactly what a play on words, I read and do not understand what I read, what it's about))
Why don't you sleep at night? Or do you live somewhere in America?
Insomnia from stress...
What do you have in your table? I see there model scores, acuras, and so on. What does this have to do with predictors?
Exactly what is this play on words, I read and do not understand what I read, what are we talking about))
Doesn't the model estimate affect its efficiency when applied to an unknown sample?