Machine learning in trading: theory, models, practice and algo-trading - page 802

 
Uladzimir Izerski:

On item 1, I can assume that for predictors incorrectly selected common factors acceptable for different instruments.

By point 2 it follows from the error of point 1.

On item 3, most likely, the wave structure is not taken into account.

On item 1, I can suggest that thecommon factors accepted for different instruments were selected incorrectly for the predictors.

Atitem 2 follows from the error in item 1.

I do not think so, since the choice of predictors is much luck. The ones I've been using for years I got from selecting out of 200. Just a fact.

On point 3, the wave structure is most likely not taken into account.

Here is a graph of the predictor with sd = 1%


And here is the chart with sd=10%.


 
Alexander Ivanov:

p.s. And give me your home address...

You'll have a hard time getting from your Kolyma yurt.

 
SanSanych Fomenko:

On item 1, I can assume that for predictors incorrectly selectedcommon factors acceptable for different instruments.

On item 2 follows from the error of item 1.

I do not think, as the choice of predictors in many respects luck. The ones I've been using for years I got from selecting out of 200. Just a fact.

On point 3, the wave structure is most likely not taken into account.

Here is a graph of the predictor with sd = 1%


And here is the chart with sd=10%.


Without time reference to the instrument(s) it's hard to draw any conclusions.

 
Maxim Dmitrievsky:

You'll have a hard time getting from your Kolyma yurt

hahahaahah))) LoL

 
Alexander Ivanov:

hahahaahahah))) LoL

on the unfed reindeer

 

Young neuralists, aren't you invited by Sberbank?

Herman Gref is just dreaming of developing AI ....

 
Alexander Ivanov:

Young neuralists, aren't you invited by Sberbank?

Herman Gref is just dreaming of developing AI ....

he writes all this under different nicknames

 
Maxim Dmitrievsky:

This is all he writes, under different nicknames

hahaahahahaha))))

DOWNSHIFTERS))))
 
Maxim Dmitrievsky:

on unfed reindeer.

What's the difference between reindeer and elk? Conditionalities. Why react.

Profitability of TS will always depend on the forecast.

The NS is designed to quickly react to the current situation and the ability to predict a possible scenario in the future.

I emphasize POSSIBLE.

A neural network can never guarantee, but only assume.

Do you have to agree with this?

 
Uladzimir Izerski:

I never understood your statements, your pictures or your philosophy at all

Ok, I agree, what do I care

Reason: