Machine learning in trading: theory, models, practice and algo-trading - page 2366
![MQL5 - Language of trade strategies built-in the MetaTrader 5 client terminal](https://c.mql5.com/i/registerlandings/logo-2.png)
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
I see, it's a kind of time compression - throwing out lines where almost nothing happened. I guess that makes sense. But there probably aren't very many of these lines, about 5 percent? And mostly nighttime?
The point is to remove autocorrelation in the rest of the model, otherwise it retrains on the series. In sliding windows, if there are a lot of trades, it's 95 percent
You have to look at specific datasets.
The point is to remove the autocorrelation in the model's residuals, otherwise it retrains on the series. In sliding windows, if there are a lot of trades, percent 95
You have to look at specific datasets.
You throw out 95% of the rows? Tough)
You're throwing out 95% of the lines? That's harsh.)
they're about nothing, maybe less, but a lot
They're nothing, maybe less, but a lot.
It's kind of like going from M1 to M20. The spread should be almost imperceptible by now. It's strange that it's even hard to overcome.
On what TF were you thinning the lines like that?
It's like going from M1 to M20. The spread should be almost imperceptible by now. It's strange that it's even hard to overcome.
On what TF did you thin the lines like that?
I do not remember, I think it was H1. There is no transition there, it's probably my fault with the spread.
Ideally, the signs should be cleaned in such a way that they do not correlate either with themselves or with others.
then you can resample, if there's not much left. resampling via GMM does all this by default, as far as I understand.
So we should check the predictors for stability and comparability at different time intervals.
What's the point?
Well, you checked, you saw that there is no stability (there is no stability), and then what? And then you have to do what I say.
In computing resources.
And how much power powerful?
I remember that you once wrote that when you create python scripts you make a graphical representation on paper, like a roadmap
It would be interesting to see, if you don't mind sending a couple of photos)
I remember that you once wrote that when you create python scripts you make a graphical representation on paper, like a roadmap.
It would be interesting to see, if you don't mind sending me a photo or two)
I do not remember this, no, I'm writing directly from my head
What's the point?
So you checked, you saw that there is no resistance (And there is no resistance) and then what? And then you have to do what I say.
And how much power are powerful?
If it does not, then the fit makes no sense.
There is a server iron E5-2660/E5-2670.
If it's not there, then the fitting makes no sense.
Doooo long sigh... forgotten, tired)
I have E5-2660/E5-2670 server hardware.
384 GB RAM ?