Machine learning in trading: theory, models, practice and algo-trading - page 1452
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 think this is a reprint of a material I read about 6-7 years ago, but here about trading volatility, a couple of times I wondered how to simulate options trading through regular orders - I couldn't find anything
This is relatively new, as far as I know. Although the approaches are not new in principle. The example on python is approximately dated when the method was proposed. There are conferences held by people there.
This is relatively new, as far as I know. The Python example is from around the time the method was proposed. People are having conferences there.
No, the rough volatility is googled and the 2014 articles, but never mind - we are talking about trading via derivatives, right?
No, the rough volatility is googled and 2014 articles, but okay, not the point - we are talking about trading through derivative financial instruments, right?
I don't think so, options are volatility trading, but you can use strats for spot.
There are essentially 2 strats there - mean reversion and volatility breakout. How exactly to fix one of them we need to think.
PS yes, the article in the archive was published in 2014
Have you read too much fiction?
haha
Aleksey Vyazmikin:
The second strategy with the result 0.65 - is not a fantasy, there the decision is made only after the unit appears, but it is possible to identify the unit in 23% of cases and correctly identify it in 65% of these 23%. And it should be taken into account that the risks are approximately 1 to 1.3 when opening a position but they are partially covered by trawl - the overall balance will be wavy enough (it makes no difference whether it is equity or balance because of reasonable stops).
I'm not arguing, it's easy to get 95% as for ZZ, but it is useless, I mean 65% quality of pure future price change prediction without admixture of the past, on what ASR directly depends.
Senior brothers in the trade, somewhere in the wilds of the branch suggested to test on the SB, take the price of the SB and see what will be akurasi and everything else, if it is clearly over 55% then obviously somewhere a shitty, because the SB can not predict much more than 50%, but with ZZ that price that SB equally "cool" predicted, this means what? That it is possible to trade on SB?Not necessarily, options are volatility trading, but you can apply the strats to the spot
There are essentially 2 strats there - mean reversion and volatility breakout. How exactly to fix one of them we need to think.
PS yes, the article in the archive was published in 2014
i still can't read your links, i'm still dragged into this q-learning, i have a lot of reading to do
i don't know how to apply valotility trading to spot, the best i can offer is simple grids))
I can't read your links, I keep getting drawn into q-learning, I need to read a lot
I don't know how to apply valotility trading to spot, the best I can offer is trivial grids )))
Sutton and Barto on cunnilingiling. I have the old version of the book in Russian, the new one is only in english on google. The new one has examples in python.
on cunnilingualism by Sutton, Barto. The old version of the book is in Russian, the new one is only in English on google. The new one has examples in python.
Yes downloaded, there is a lot to read
SZY: there are examples of CNTK in the network, it seems that it's not difficult to make LSTM in C#, one problem, Microsoft lazy, even on the CNTK official page they send to study the API from Python, like this tutorial, use there also
https://bhrnjica.net/2017/12/07/cntk-106-tutorial-time-series-prediction-with-lstm-using-c/
Yes, I downloaded it, there is a lot to read, and you also need to check this kuni)))
ZS: there are examples of CNTK in the network, it seems it's not difficult to make LSTM in C#, one problem, Microsoft lazy, even on the CNTK public page they send to study the API from Python, they say here is the manual, use there also
https://bhrnjica.net/2017/12/07/cntk-106-tutorial-time-series-prediction-with-lstm-using-c/
they have some kind of illiquidity, I don't know who uses it
try 2 layers and reduce the number of neurons in layers, down to 1 in each layer.
before the white vertical line - sample, after - oos
The more neurons - the more probability of fitting (more degrees of freedom), try to reduce the number of neurons as long as the neuron can produce at least somewhat sane results.
That is, the clearer the information in the inputs and the rougher the mesh, the better.
Vladimir, hello!
How are you doing with the script I sent you, have you tried to experiment with it? Maybe you've developed the idea and the regression approach?