Machine learning in trading: theory, models, practice and algo-trading - page 2458
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
It was laid out on the fingers and, in general, the result showed, so that the goal was achieved.
In general, cool. Truth poked the link to the trained models, the car could not park, I thought there would be a hell of a box that parks like Schumacher
On the whole, cool. But I poked the link to the trained models, the car could not park, I thought there would be a hell of a box that parks like Schumacher
Nah, she's not trained yet. I left my tab to learn, already sometimes comes close )
But the algorithm is quite crude, schumacher there will not be a month.
What things?
A set of events aimed at making a profit. For example the quants now know what to do when the market closes and what market participants will do. They buy on purpose, knowing that the crowd on Friday, will sell at any price and on Monday the fall will continue, but without participants. The usual things that local statist-programmers know very well.
On the whole, cool. But I poked the link to the trained models, the car could not park, I thought there would be a hell of a box that parks like Schumacher
it's not bad, not bad at all. if you could see that the brakes were bad, it would be great.
every day I see people parking much worse)))))
A set of activities aimed at making a profit. For example now quants know what to do when the market closes and what market participants will do. Buying purposefully knowing that the crowd on Friday, will cover at any price and on Monday the fall will continue, but without participants. The usual things that local statist-programmers, very well know.
If quants and you "know," why aren't you billionaires yet?
If quants and you "know," why aren't you billionaires yet?
It's not bad, not bad at all. you can see that the brakes are bad - it would be great.
every day I see people parking much worse)))))
I had this thing for 2.5 days.
For some reason after the 9th gen. the error increased dramatically:
1st Best Car Genome for some reason shows an error of 1.52, even though the chart above has a minimum point of 0.67:
After switching to another tab and back started the 10th generation, the graph corrected itself, and there was a new leader immediately:
But overall raw, of course.
I satisfied my curiosity and that's enough.
But on the whole it's raw, of course.
I've satisfied my curiosity, and that's enough.
Of course it's just a toy, although fun.
I will leave a link to examples from Machine Learning for Algorithmic Trading in Financial Markets.Stefan Jansen
P.S. And, of course, I still believe in the limitations of machine learning for TS, although I admit its usefulness in RM
Machine learning in risk-management can use Matlab, Python, R
even though I haven't coded it yet... (to distinguish Trend and Flat to enable appropriate TS, you can of course rely on this kind of feedback [to analyze and vary MM algorithmically], but still I don't want to pay for market mood and my busyness, losses, even if reduced to a reduction by some algorithm and probability theory)... I'm still inclined to the clear distinction between good and bad trades in terms of market conditions and trader's awareness of them, and there is no way to inform the robot about it... The only way to do that is to teach it not to trade at a loss much (when the demand - supply conditions have changed, and the trader is not present at the terminal or he himself is not aware of them)
I will leave a link to examples from Machine Learning for Algorithmic Trading in Financial Markets.Stefan Jansen
P.S., and, of course, I still believe in the limitations of machine learning for TS, although I admit its usefulness in RM
although I haven't coded it yet... (to distinguish Trend and Flat, you can of course rely on this kind of feedback [to analyze and vary MM algorithmically], but still I don't want to pay for market mood and my busyness, losses, even if reduced to reduction by some algorithm and probability theory)... I am still inclined to the clear distinction between good and bad trades in terms of market conditions and trader's awareness of them, and there is no way to inform the robot about it... The only way to do that is to teach the robot not to trade at a loss (when the demand - supply conditions have changed, and the trader is not present at the terminal or he is unaware of them)
The usual review of the MO methods, with the prefix of trading.
I did all of the above, except that I did not try sostigal networks, but it does not matter, the important thing is that no miracle happened.
That's not what trading is all about.