Machine learning in trading: theory, models, practice and algo-trading - page 1077

New comment
 
Maxim Dmitrievsky:

yes, PRDF

Ok, then you can create a switch case and add a bunch of polynomials there and call the function randomly to choose one polynomial...then, why did you said GMDH instead of PNN...then??:))))))))))))))))))))))))

I mean I unnecessarily wrote such big for loops:))))))))...

So now just use the same function which I gave to you and replace the for loops by a single polynomial of some degree and then, call that function for every line...

 
Maxim Dmitrievsky:

I said "like gmdh feature selection" I think:


So now just use the same function which I gave to you and replace the for loops by a single polynomial of some degree and then, call that function for every line...

I mean one condition in switch case= one different polynomial which will be chosen randomly...Now you can write any number of polynomials as you want since there are no for loops:))))

If you want to me write...I can write as well...but I need some time to search for some good combinations...

By the way, for me to better understand your goal can you please explain me exactly what are you looking for or trying to achieve in terms of:

1.OOB LOSS and LOG LOSS...values or conditions etc?

2.Sharp ratio, profit factor, recovery factor etc. or these things don't matter due to overfitting?

Anything else based on on which we can decide the reliability of the system for LIVE trading?

I mean which parameters are you trying to achieve or increase etc.?


 
Maxim Dmitrievsky:
Now I think I can drop out 1-st 2 lines, because 3-d line do it all

But how are you going to drop out the lines?

Are you going to write the code for every feature ... I mean all the 100 features if we use 100 features :))

 
Maxim Dmitrievsky:

yes, I leave 1-st line and combine together 2 and 3, 3-d (now 2nd) line will consist of all 2 features variants

so 1-st step: we are choosing best predictors without polynomial transformations

2-nd step: combining all with which others with different polynomials, by adding 3-d predictor, where its multiplication of 2 predictors with different degrees

and next - need to combine these predictors from step 2 to which others

Are you sure this line is correct? I mean the syntax is correct?

m[i].Set(2,MathPow(RDFpolicyMatrix[i][(int)models2[bf][1]],firstDegree)*MathPow(RDFpolicyMatrix[i][(int)models2[bf2][2]],secondDegree));
        

My brain already hurts by seeing these 2D matrix and all that stuff and so just checking if the syntax is correct:))))))

 
Maxim Dmitrievsky:
nope, I will use all 3 lines, all good now )

But what next...:)) are you going to write 4th line, 5th line,....so on for all features?

 
Maxim Dmitrievsky:

now we have: best triple features, transformed with polynomials

can combine all best features now, adding 6 inputs, next 12 inputs and so on

and then can choose the best model over all


No...I want to know if you are going to write the code again and again for every feature or will the existing loops and code will do it all?

 
Maxim Dmitrievsky:

this strings now contain 1 unit, which takes 3 inputs

we can operate with these units now, multiply it, so it can takes 3+3+1 = 7 inputs

This is holy sh.t.

Well, you decided to write it that way and till now I am not 100% sure as to why you are trying to write in that way... :)))

I am not an expert in writing programs of 2D arrays or matrix and hence, here my role is limited....

I can only understand important concepts quickly and help you somewhat to simplify things a little better... but you should know how to program it...:))))

 
Maxim Dmitrievsky:

I just try with 3 lines now, fixing an errors

if the error is reduced by line

OKKK ,, ..... Now I am getting closer in understanding why you are trying to implement in this way .... :))

If you want, you can provide me with the full code and I can try different variants of polynomials and it by replacing in this line:

m [i] .Set ( 2 , MathPow (RDFpolicyMatrix [i] [( int ) models2 [bf] [ 1 ]], firstDegree) * MathPow (RDFpolicyMatrix [i] [( int ) models2 [bf2] [ 2 ] secondDegree) );
 
Maxim Dmitrievsky:

now just some errors in heaps of loops . ^))) I try to fix

You mean syntax errors???...I thought LOG LOSS errors:)))))))))))

 
Gentlemen, why don't you go ... to the English part of the forum?
1...107010711072107310741075107610771078107910801081108210831084...3699
New comment