Machine learning in trading: theory, models, practice and algo-trading - page 2724
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
200 pcs per observation
just train the TS for some period and show it on new data, on eurobucks h1 for example
just train for some period of time and show on new data, on eurobax h1 for example
I can't bring myself to finish writing the code, burned out....
I'm looking for an incentive, cheledge might help
can't bring myself to finish writing the code, burned out.....
I'm looking for an incentive, cheledge might help.
Challenging myself with different approaches is unlikely to help.
So take the code from the article and check it. Why should I understand what we're talking about if there is nothing yet.
I have a lot of variants of such implementations, including those with retraining and not retraining again, retraining by baselines and so on.
Especially when you start doing it, the result is different from the fantasy one that was originally planned
I already had a variant at that moment, but not enough experimentation, as with my method of training through the console it is not convenient, I wrote there about it.
At that point, I guess there was no "heap" as they referred to learning it.
That the final version may be different - I agree.
Don't say that I accused you of stealing, maybe you didn't understand anything at that time and learnt everything from books - the point is that a similar approach was already mentioned and it was not evaluated properly.
I already had an option at that moment, but not enough experiments, because with my method of learning through the console it is not convenient, I wrote there about it.
I guess there was no "heap" at that moment, as they referred to the study of this issue.
That the final version may be different - I agree.
Don't say that I accused you of stealing, maybe you didn't understand anything at that time and learnt everything from books - the point is that a similar approach was already mentioned and it was not evaluated properly.
I'm not saying anything
This method is >1.5 years old, I just wrote the article recently.
chelenge on different approaches is unlikely to help.
If there's only one dataset, what's the problem?
I can send a trained bot over right away.
I can send a trained bot over right away.
Okay, forget it.
chelenge on different approaches is unlikely to help.
answer the question logically: why no method for BP works on the market if the market is BP.
But for any other BP these methods work, in fact, they were created for that purpose.
There are methods that work. The difference for the market is that in this BP the data noise is too high. That's why you can't teach it head-on, you need all sorts of tricks. Like moonshine, you need to distill it several times, isolate the extract and work with it.