Machine learning in trading: theory, models, practice and algo-trading - page 2004
![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
dog, the console hung up again while I was leaving
They say that the new console in Windows 10 is like that.
no, stop. Yes, this row does not work in reverse.
)))
what akurasi? 63%?
What if you send two rows to the input?
In one column the current value, in the second forecast.
Max, it seems to me with the RNN you are fine, but the mentality of the bot is lame, comparison with the level of profitable trades (comparing with the bot on the NN):
On the transformed series it is clear why such good results at the backward stroke, but if at the normal price the indicators are better, then it is a paradox, it turns out that the past depends on the future.
And this is a great, no, even the coolest observation.
The paradox of the dependence of the present and past on the future is inherent only in non-Markovian processes. Thus the entire theory of random Markovian processes as applied to the market can be scrapped.
This is a great, no, even the coolest observation.
The paradox of the dependence of the present and the past on the future is inherent only in non-Markovian processes. Thus, the whole theory of random Markov processes, as applied to the market, can be sent to the furnace.
Well this observation has not yet been confirmed. None of the neural network wizards present here wants to conduct experiments on the normal price.
It turns out that if the price series is reversed and the tests are better at it than at a straight line, then the process in the market is not Markovian, i.e. not random? And it gives hope to those who suffer.
The process in the market is not Markovian, i.e. not random.
You were interested in the pound, you wanted to run something on it, some kind of chemical experience. You wrote that when your rating was2888.
Notice how the pound handles eights. And it's not just the pound anymore.
You were interested in the pound, you wanted to run something on it, some kind of chemical experience.
link to the post, I don't remember something like that.
link to the post, I don't remember something like that.
And you deleted or corrected a recent comment ;)
Well this observation has not yet been confirmed. None of the neural network experts here do not want to conduct experiments on the normal price.
It turns out that if the price series is reversed and the tests will be better on it than on a straight line, then the process in the market is not Markovian, i.e. not random? And this gives hope to those who suffer.
right
ever wonder why?
If you analyze further, you will also notice that not all pairs are like this