Machine learning in trading: theory, models, practice and algo-trading - page 3376
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
The first option, tables - Excel tables, to everything each row has a time marker. The most familiar form of financial data.
The second option, handwritten letters. Learning with a teacher, where as a teacher is a printed letter, and the column below it is variants of handwritten spelling of this letter.
Comparing bousting and NS. Which is more suitable and for which case? Or is it equivalent?
PS.
From Rattle, which has rpart (simple tree), rf, ada, SVM, glm, nnet (probably the simplest NS). The worst result is with rpart, second from the end is nnet, the other four are about the same, depends on the input data.
Theoretical question.
There is a TS, which is splendidly fitting. At the same time, it is precisely known that a certain set of input parameters profitably exploits the real pattern. That is, this set is not a fit.
Is it possible to find this set?
Theoretical question.
There is a TS, which is splendidly fitting. At the same time, it is precisely known that a certain set of input parameters profitably exploits the real pattern. I.e., this set is not a fit.
Is it possible to find this set?
No, since there are no patterns in NOT stationary data
Theoretical question.
There is a TS, which is splendidly fitting. At the same time, it is precisely known that a certain set of input parameters profitably exploits the real pattern. I.e., this set is not a fit.
Is it possible to find this set?
Well, you found it
You are talking about the input parameters of the TC, and I am talking about the non-stationarity of the input data. If you work with non-stationarity, it is possible to create TCs based on garch models or in the framework of MOEs that continue to struggle with non-stationarity. Input parameters play a secondary role.
Theoretical question.
There is a TS, which is splendidly fitting. At the same time, it is precisely known that a certain set of input parameters profitably exploits the real pattern. I.e., this set is not a fit.
Is it possible to find this set?
Theoretical question.
There is a TS, which is splendidly fitting. At the same time, it is precisely known that a certain set of input parameters profitably exploits the real pattern. I.e., this set is not a fit.
Is it possible to find this set?
Not a fit to what and for how long? I am embarrassed to ask :) it is probably more important to prove that it is not an adjustment to anything.
The nature of the profit curve does not change by OOS: Size(OOS_Left) = Size(OOS_Right) = Size(Sample). All in all, a result you can't pass by.
Theoretical question.
There is a TS, which is splendidly fitting. At the same time, it is precisely known that a certain set of input parameters profitably exploits the real pattern. I.e., this set is not a fit.
Is it possible to find this set?
If you remove the"TC that is a great fit" clause, youcan narrow the field considerably. And there's a lot to know about the pattern you're looking for right off the bat, too.
You can't take an arbitrary bunch of indicators, remove the obviously dependent ones, and from there get a "fitted TS" and in all of this isolate the exploitation of the real regularity
A good question is half of the answer - In the real world, where alg.optimisations and ML are applied, it is usually known what exactly is being searched for (darkness of characteristics) and it is necessary to highlight features, outline limits. And here nobody knows what he wants to find, but he knows how to run the optimiser :-)