Machine learning in trading: theory, models, practice and algo-trading - page 3245
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
Not for you.
so it's the usual.)
so the usual.)
Groundhog Day.
That's for sure.
By means of bards and gpts I tried to speed up my code in Python, which is not available in the Internet, i.e. without peeking. There are 20-30 lines. None of them worked, moreover, none of the suggested codes works at all. No matter how hard I try, there is no result. Which kind of hints too that these llm's don't know how to extrapolate.
Use bing, if it can't do it, no one can do it.
I'm dabbling in a quick pattern search.
I haven't been able to beat the MoD results yet
Associative rules?)
Associative rules?)
Correlation ) you can make sequences consisting of the values of different features and look for similarities in the stories
Along the lines of ensemble wood models....
If we can't take quality, let's take quantity....
Each tree separately predicts very badly, but if there is an ensemble of 100-1000 trees, we get more or less good forecasts due to noise suppression(here we have detailed information about it, and here).
If we make an analogy with trading - if we cannot take quality (make one good TS / one tree), we can try to take quantity by a set of TS / ensemble of trees.
And it is not necessary that each TC earns a lot and well, the main thing is that it earns exactly.
And so the task is to come up with a set of TCs.
I have developed a certain automatic generator of trading systems, it was found out that with one idea and one tool you can get about 10-20 weakly related / uncorrelated TS.
Anidea is, for example, one indicator that you have developed and on its basis you build some trading rules (TS).
The TSs themselves are quite simple, which gives hope that they did not turn out as a result of overtraining, but time will tell....
At the moment a robot for the euro on one idea, 15 TS, if the TS will work on the new data, the total trading set of TS should give a stable profit, but the TS should be at least 100 pieces.
Trading 15 TCs on the Jew looks pretty creepy.)
and so far in plus, but it's only the first day of testing, there were a lot of bugs...
I have not told anything super natural, and there is nothing new here, but still I find it interesting....
dabbling in a quick pattern search.
I haven't been able to beat the MoD results yet
You can also mine like this, I mined this on the clock today. I have been mining since 2020, everything before that is oos
You can also mine like this, today I mined this on a watch. I've been mining since 2020, everything before that is oos
Martin is good.