Machine learning in trading: theory, models, practice and algo-trading - page 1080
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
As for Max's discussion with the bourgeois, that's great and I personally support those who are in favor of it.
You should always look for the problem not in someone else, but in yourself.
Why aren't we doing this yet? It's the grail)
Why aren't we doing this yet? it's the grail)
Well... I don't know who doesn't, but for me it's already been done :).
What did you do?
What did you do?
I did not apply the genetic algorithm, as I have not finalized the mechanism of trade analysis by the training, and such flows (trade in training) for one instrument should be a few, to select the best and continue its development, because I found that the subsequent learning worsens the quality of the result, most likely made a mistake somewhere, my natural inattention.
In my case, errors in the code, can be regarded as a mutation ))))))
About Max's discussion with the bourgeois, that's great and I personally support those who are for it.
You should always look for the problem not in someone else but in yourself.
:)
People who associate themselves with a particular language or nation are already infected.
at
Good afternoon, Nicholas!
Judging by your comments, mts is based on knowledge base or machine learning. Can you explain the essence of the theory for a layman on the fingers. Suppose I have two criteria - the color of a candle and the breakout of the maximum. I want the machine to "learn" by these criteria. How does training take place and what is the forecast - price movement on the next bar, price movement on the specified horizon or certain event - whether it will/will not break through the previous extremum?
Regards, Yury.
Answer:
Good afternoon!
To simplify it.
the essence is that you create a base of attributes and a base of decision-making rules,
and the algorithm of synthesis of the decision on the basis of the initial bases.
After that, you start training the system.
You tell the system where you think the decision should be made.
The system synthesizes the decision-making rule in the places you specified according to a certain set of criteria.
If the rule has been successfully synthesized, it is entered into the database of decisions and then used for work.
If no solution can be found, your decision-making opinion is rejected.
Then the natural selection of decisive rules comes into effect.
Some decision rules are further strengthened by evolution,
and some are weakened or killed altogether.
This is how the system learns to make the right decisions and works later on.
http://www.kamynin.ru/2012/03/22/xroniki-robota-22-03-12/#comments
Hindus don't write.
Misha doesn't write.
It's autumn evening...
So why are you complaining.... No fish :-( Life sucks, but I have a shovel :-)