Machine learning in trading: theory, models, practice and algo-trading - page 2079

 
Maxim Dmitrievsky:
@Andrey Dik
Hi.
 
Andrey Dik:
hello.

hi )))

 
Andrey Dik:
hello.

Hi, do you have any experience in creating non-standard and most likely multi-criteria fitness functions?

 
mytarmailS:

Hi, do you have any experience in creating non-standard and most likely multi-criteria fitness functions?

I will.

funny russian, though)

 
mytarmailS:

multicriteria fitness functions.

Pareto sets? That would be interesting to see.

 
For some reason I couldn't find any discussion of using MO for news trading on this forum. Isn't there anything like that here?
 
Aleksey Nikolayev:
For some reason I couldn't find any discussion of using MO for news trading on this forum. Is there really nothing of the kind here?

It all rests in the archives... I'll have to look for normal ones

 
Maxim Dmitrievsky:

It all comes down to the archives... we'll have to find the right ones.

Even for some simple task like binary classification of days - was there significant news during the day or not?

Obviously, news has to be taken into account somehow, but so far I don't understand how to formalize it all correctly.

 
Maxim Dmitrievsky:

It all comes down to the archives... we'll have to look for normal ones

If you find it, send me the link.
Although it will be a hundred or more new features, for each type of news: forecast, reality, previous.
 
Andrey Dik:

help.

funny Russian, after all).

Let me show you a simple example...


We have a trading system based on two waveforms with periods of 10 and 20, trading by crossovers, classic...

The bag is a low pass filter, the period of the bag is the controlling parameter

In this case the controlling parameter is equal to the constants 10 and 20

the problem: on a given segment of the market to get DYNAMIC (not constants) control parameters of each sledgehammer, so they would be optimal in terms of profits

this is criterion 1


criterion ¹ 2 : received dynamic parameters of control should be similar to continuous function, instead of chaotic dispersion of points, that is it is necessary to introduce the concept of smoothness of function...


How do you see the solution of such a problem?