Using neural networks in trading - page 20

New comment
 
faa1947:

Thank you.

I cannot accept the answer. It is not substantive. Trading on patterns is called "Technical Analysis". You are trying to find another pattern, nothing more. But you cannot go beyond the limitations of TA.

And what makes you think that pattern trading is only pattern trading? If someone says he uses TA, it doesn't mean anything at all, because most likely he is just not telling us a lot. )) In addition to the entry/exit block (which in principle is most often discussed), there may be ten other blocks that make up the trading system and these ten blocks are no longer related to TA.
 
faa1947:

Trading on patterns is called "Technical Analysis". You are trying to find another pattern, nothing more. But you cannot go beyond the limitations of TA.

And econometric approaches that allow themselves to be coded into an automatic TS won't be solid TA? Isn't the regression you mentioned, and indeed regression analysis of time series TA?

It's up to you to use NS or not, believe it or not, otherwise it's pure demagoguery.

 

Hello all.

I have not been lazy and have read the whole thread. I've been doing neural network forecasting for a few years now.

But in many ways I'm still a novice.

And here's what I've come to:

1.There are people smarter than me.

2. one grid won't solve the problem.

3. Lack of quality data.

4.What to teach.

5.There are more powerful computers than my 4.

But more specifically on the subject, I really can't find "what to teach?"

 
TimeMaster:

The best and most powerful neural network is your brain, which has billions of neurons.

So if your neural network, in the form of a brain, cannot predict the market, what do you want from an artificial neural network with a few tens or even hundreds of neurons? Of course it can't predict the market either, whatever data you give it..... ))))

 
LeoV:

The best and most powerful neural network is your brain, which has billions of neurons

A neural network is just a model, the simplest model.
 
tara: A neural network is just a model, the simplest model.

That it is a model is clear. It all depends on what it is a model of ))))
 
A primitive model of vegetative processes.
 
What is the essence of the autonomic process model?
 

The essence of the model is the image of the original.

 
tara: The essence of the model is the image of the original.

Neurons are not enough to represent the original )))). If a human neural network cannot represent the original (or rather predict it), let an artificial neural network do the same )))

Although, at the same time, the simplest neural network with 5 or more neurons would be happy to learn 10 years of any instrument's minute chart with minimal error.....

Only with prediction will be a big problem.....))))

1...131415161718192021222324252627...40
New comment