Discussing the article: "Developing a robot in Python and MQL5 (Part 1): Data preprocessing" - page 4

 
Maxim Dmitrievsky #:

few things fit, so all signs are at least detrending.

Taking the MA out of the price?

 
fxsaber #:

They take the Mashka out of the price?

Yeah, and sometimes divide the difference by the standard deviation. That adds to the pseudo-stationarity.

 
fxsaber #:

Like, MO is positioned as a being (mega-human) with super-computing abilities and memory.

And what attributes does an ordinary person use to create a TC?

That's already a question for the positioners :)

there are simple rules what is edible for ME and what is not.

 
Maxim Dmitrievsky #:

That's already a question for the positioners :)

So man creates in his own image. I.e. MO is a strong generalisation of standard approaches.

 
fxsaber #:

This is how man creates in his own image. I.e. MO is a strong generalisation of standard approaches.

If we are talking about classification, it is a generalisation tool for further prediction. You can make a hypothesis and look at the degree of generalisation. If it is high, it almost automatically means a good TS and one can predict further.

 
Thank you all for your feedback! Next article, if moderators allow, I will make a tester, and more extensive work with signs. The fact that they should be plus-minus in the form of oscillators, this has been known since Arima.
 

I have also noticed a strong influence of the risk/reward ratio on a positive forward. For example, a negative risk reward.

 

And here's a positive, 1:3 risk reward ratio in the tags.

 

Terminology: n/b - foul language.

n/b python, code for one version does not run on another n/b.

The author did not specify the n/b python version .

n/b code does not run. it asks for some n/b bible, what n/b.

I try to install the n/b bible, it says the version of n/b python is not the same n/b.

Is it difficult to write the version and how to install the necessary libraries? Or have you run out of letter limit?


 
Aleksandr Slavskii #:

Author, didn't specify the version of n/b python.


Judging by the screen, Python version 3.10.10 was used.