Machine learning in trading: theory, models, practice and algo-trading - page 779
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
Thanks to this branch and without the retention of work and fun in this branch I get a dollar drop every week... Funny!!!!
The main thing is not to burn out. Good luck in trading.
In fact, this is one of the signs that the TS is adequate to the market...
I think that any movement in the market should be formed a prerequisite. There is no ungrounded market movement. Even the flat should be based on the precondition.
So our task is to correctly assess these prerequisites before the start of the movement.... IMHO naturally....
One of the options is to evaluate the market at the moments of the signal CONTROL of trend strategies. Whether it is a real reversal or a false one....
You're right, Misha.
You`re not the only one who thinks about it.
It's hard to find the answer, since no one who trades can look into all the other terminals.
And in 2008 even the DCs got burned, because apparently there is not enough information inside one DC too...You're right, Mish.
You're not the only one who's been racking his brains about this.
It's hard to find the answer, because none of the traders can look into all the other terminals.
And in 2008 they burned their terminals, because there is not enough information in one terminal.Only information from a real exchange will help you get a half percent advantage over others. Sometimes it's enough.... In complex situations...
1st version of the RL bot
Well, self-training is cool, at least the target does not need to pick up
The right half is OOS. This is the result of what I suggested for OOS in KB.
The right half is OOS. This is the result of what I suggested for OOS in KB.
I'll bang you on that later, haven't figured it out yet :)
very cool!
1st version of the RL bot
Well, self-training is cool, at least the target does not need to pick up
What do you mean you do not need to select? It knows that the master wants to make a profit? Or maybe the master wants the minimum drawdown? What is telepathy here?
:D
And gets the optimal strategy on the first 10 games, further significant improvements do not happen
1st game with a random policy, and often only it and negative, then immediately converges to the optimum... still do not understand why so quickly, probably just the algorithm so picked up
What is missing in these pictures - a forward test (and probably not by accident ;) ).
Any freshman programmer can build such a graph in the backtest.
There is something missing in these pictures - a forward test (and probably not by accident ;) ).
Any freshman programmer can build such a graph in the backtest.
In these pictures is exactly what was written about
about freshmen I have no idea, apparently the local inhabitants did not even get to the 1st
I showed exactly what the dull Vizard showed yesterday, and then got scared and deleted