Machine learning in trading: theory, models, practice and algo-trading - page 945
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
Do you think these cycles exist on small TFs? Let's see, there are more bars there!
I'm not interested in small ones, I'm interested in total pattern changes by quarters and higher!
I've got an article, check what I need!
I'm not interested in small ones, I'm interested in total pattern changes quarterly and above
There's an article, check it out.
Yes, yes, I read the article. As expected, there are actually functions that describe trends (directional movement), if the function has stopped describing, we generate a new one, and so on.
I understand correctly that the amplitude on the graphs shows the width reached by the function to describe the selected window?
Where is Alyoshenka the son with his 100% accuracy? Help, save us suffering - right here, lay out the Grail. For it will be rewarded.
What have I got to do with you, with your 10% errors? When you go to zero or in deficit, then we'll talk, I wish everyone to go in deficit.
Here is another interesting experiment.
You have a profit for each trade entry (finRez_Buy). You can configure the tree not for classification, but for regression - give it as a target for these very profits. The tree is unlikely to be able to predict specific values, but it will at least split the target into several groups, depending on the allowed branching. Then you can draw the resulting tree, and following the branches try to understand - where there are more profits. Maybe it will help you in your information retrieval.
A positive forecast of such a tree can be regarded as an entry signal.
Am I correct in assuming that the amplitude on the charts shows the width reached by the function for the description of the selected window?
i broke my brain trying to understand the question. i'm going to rest and weep bitterly over the plight of poor traders
and Aleshina individually
What have I got to do with you, with your 10% errors? When you go to zero or in minus, then we'll talk, I wish everyone to go in minus.
Alesha, seriously, I'm already in zero, because I went on a slippery slope streams Erlang, separating them from the tick BP. I ran into trouble right there... Didn't calculate everything... But, I'm not making excuses, I just want to know - aren't you going to convert BP in the same way or using what is there, i.e. OPEN, CLOSE on M1, M5, etc.?
I do not need your secrets of preparing the predictors. Just asking for an answer - are you converting BP or not? Help an old man, please...
Sincerely,
A_K2
Here is another interesting experiment.
You have a profit for each entry into the trade (finRez_Buy). You can set up the tree not for classification, but for regression - give it as a target for these very profits. The tree is unlikely to be able to predict specific values, but it will at least split the target into several groups, depending on the allowed branching. Then you can draw the resulting tree, and following the branches try to understand - where there are more profits. Maybe it will help you in data mining.
A positive forecast of such a tree can be regarded as an entry signal.
We're in the wrong business!
A picture created by a neural network graced the cover of the magazine (photo)
Bloomberg Businessweek magazine published an issue entitled Sooner Than You Think, dedicated to artificial intelligence.
On its cover, the magazine placed several pictures drawn by a neural network. The editors enlisted the help of American programmer Robby Barratt, who not so long ago "taught" a neural network to draw nudes.
Withthe help of Barratt's previous developments, this time his neural network was able to draw several landscapes that are hard to distinguish from the brush of a real artistThe programmer himself in an interview with the publication expressed the opinion that neural networks will certainly become one of the main instruments in the art of the XXI century, and even very soon developed by him artificial intelligence will be able to "experience" mood swings, which will also be reflected in the work of
I got in trouble right away... I don't have it all figured out...
You play in the casino, on casino property, according to the rules of the casino. Renat and I warned you. You'd better read on the forum about real exchanges, mt5 allows you to do that.
I broke my brain trying to understand the question. I'm going to rest and cry bitter tears over the plight of poor traders
and Aleshina individually.
The article talks about theEMD method, which consists of the construction of envelope extremum lines and the subsequent processing that results in functions. So, I understand that in this indicator chart, we see the amplitude width that depends on the difference between two extrema within a certain period of time.