Machine learning in trading: theory, models, practice and algo-trading - page 329
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
Actually, the random forest is a classification and does not deal with approximation at all
Classification is nothing but a special case of approximation in multidimensional space - the boundary of division into classes in space is some function, which is exactly what is approximated.
Forests, as I understand it, are used for predictor classification, roughly speaking, not for prediction :)
Classification is nothing but a special case of approximation in multidimensional space - the boundary of division into classes in space is some function, which is just approximated.
Into the annals!
P.S. Did you even understand what you said?
I knew that my phrase would throw the immature minds of young econometricians into a deep stupor, and when the brain has no time or cannot process the amount of information, it instinctively turns on aggression (a kind of protective reaction of the body).
But it's okay, relax and try to think again, thinking processes cause an increased need for oxygen, breathing becomes more frequent and deep - thinking is good for the whole body, not only for the cortex of the large hemispheres.
I knew that my phrase would throw the immature minds of young econometricians into a deep stupor, and when the brain has no time or cannot process the amount of information, it instinctively turns on aggression (a kind of protective reaction of the body).
But it's okay, relax and try to think again, thinking processes cause an increased need for oxygen, breathing becomes more frequent and deep - thinking is good for the whole body, not only for the cortex of the large hemispheres.
You gonna call the FSB again?
P.S. Random forests are used ONLY for classification. At most, to select predictors for a regression model.
P.S. Random forests are used ONLY for classification. At most - to select predictors for the regression model
Did I say otherwise? Hide your stupidity away, because it is very clearly protruding to the point of obscenity.
I told you - networks, unlike the same forests can do anything, including what forests do. go troll someone else, you can troll yourself in front of the mirror if no one else is around.
Did I say the opposite? Hide your stupidity away, because it is very clearly protruding to the point of obscenity.
I said that networks, unlike scaffolds, can do everything, including what scaffolds do.
Random forests don't solve approximation problems.
Oy-wey...
The R language has everything you need to trade both forex and stocks. There is an excellent MT/R combination. Just experiment and implement it. And you suggest to go where there is none of this.
If R was the first to catch your eye, and even provides a solution, it does not mean that the solution is the best or suitable for all tasks. I have already written that I use both R and SciLab, and I believe that in SciLab much is done much easier and with less work. I am not belittling the merits of R at all.
And the connection to MT is not a problem, there is nothing special to work out. There is a C/C++ API. There's no need in it at this stage, it's the last stage.
So many new words for me...
Would someone show me the result - is it worth the trouble or is it all just for fun???
So many new words for me...
Would somebody show me the results, is it worth the trouble, or is it just for fun?
I just need to read a couple of books and surf the Internet. I was in the same state a month ago). Started with SanSanych's article, and... I didn't understand a thing.)
Maxim Dmitrievsky has not quite a neural network yet, but the results are already good. And close to the middle of the topic someone published the results.
Nothing happens all at once.)