Machine learning in trading: theory, models, practice and algo-trading - page 690
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
More aggression, gentlemen, and more discontent with yourselves.
You have two weeks left to confirm your status as a quivering beast or a having a status.
It's not us but you, you've set such a deadline :)
Fucking up is not stealing boots. Where is the example after the conversion? At least a picture of what you got there. And it is desirable that the picture shows the balance of funds. Because this is the final goal of any market work, no matter how big or secret it may be....
You are a piz...dit here. You have been pointed a possible direction of search. It is up to you to check or ignore. That's because of such "smart guys" with claims prove it and it's all bullshit because I think so, and I do not want to check bullshit or not, the topic turns into a bazaar. Take part in a bazaar somehow do not want to. So I no longer write on the forum and do not put my development here. I'm already starting to regret that I posted the library for connecting the Python language. In general, it is a problem of moderation, and I don't like moderation on this site. Because squabbles are not stopped. I have no desire to read such comments and respond to them. They posted a graph of profitability for the week and you think it is an indicator? It's nothing. And the hubris is overblown. Have a nice life, everybody.
Dear Grigory! I quote you:"Moreover, after such transformation of intensity, the histograms of the increments take a strict form and it is no longer necessary to logarithm them, etc. " I would like to take a look, since they acquire a view, rather than take your word for it, and I know something, but I won't tell you. If they take any unique form, it would be a good idea to accompany pictures.....
It's not us, it's you, it's the deadline you set for yourself :)
(Ecclesiastes 3): "To all things there is a season, and a time for every thing under heaven: A time to be born, and a time to die; a time to plant, and a time to pluck out what is planted; a time to kill, and a time to heal; a time to destroy, and a time to build; a time to cry, and a time to laugh; a time to lament, and a time to dance; a time to scatter stones, and a time to gather stones; a time to embrace, and a time to evade embrace; a time to seek, and a time to lose; a time to save, and a time to cast away; a time to tear, and a time to sew; a time to be silent, and a time to speak; a time to love, and a time to hate; a time for war, and a time for peace."
(C) teacher Michael
(Ecclesiastes 3): "To all things there is a season, and a time for every thing under heaven: a time to be born, and a time to die; a time to plant, and a time to pluck out what is planted; a time to kill, and a time to heal; a time to destroy, and a time to build; a time to weep, and a time to laugh; a time to lament, and a time to dance; a time to scatter stones, and a time to gather stones; a time to embrace, and a time to shrink from embracing; a time to seek, and a time to lose; a time to save, and a time to cast away; a time to tear, and a time to sew; a time to be silent, and a time to speak; a time to love, and a time to hate; a time for war, and a time for peace."
(C) teacher Michael
Sincerely put!!!! Continuing the theme.... As you know I started to twist R and was able to twist the maximum is to count the VI between each input and output, but it was enough to reduce the input data from 110 to 20-30 with the input data remain those that have the maximum information about the output. As a result, the models began to pass my own tests more and more successfully. Let's see how it will be in the CB. The week will show.
But here I think that one VI metric will not be enough. I should try to calculate redundancy and try to reduce the number of columns.
Maybe there are already ready functions allowing to estimate input data to output besides mutual information????
Sincerely told!!!! In continuation of the topic.... As you know I started to twist R and was able to twist the maximum is to count the VI between each input and output, BUT even that was enough to reduce the amount of input data from 110 to 20-30 at the same time that the input data remain those that have maximum information about the output. As a result, the models began to pass my own tests more and more successfully. Let's see how it will be with the CB. The week will show.
But here I think that one VI metric will not be enough. I should try to calculate redundancy and try to reduce the number of columns.
Maybe there are already ready functions that allow to estimate input data to output besides mutual information????
They must be built into the model at once, that is, they must be evaluated with respect to the model and not just by themselves
The easiest way to do it in R is through a random forest. There are Gini MDI method (comparable to entropy method in quality of estimates), and method of average precision decline MDA
but I'm not an erodor and there are no ready-made codes... you can read Sanych's article about feature selection via forests, for example
or:
https://medium.com/@ceshine/feature-importance-measures-for-tree-models-part-i-47f187c1a2c3
http://blog.datadive.net/selecting-good-features-part-iii-random-forests/
prufhttps://habrahabr.ru/company/ods/blog/322534/
prufhttps://habrahabr.ru/company/ods/blog/322534/
Take a heart-to-heart... I'll see what's up...
Take a deep breath... I'll see what's up...
But all these statistical approaches are not relevant to forex :)
I'm just trying to rack my brains