Using neural networks in trading - page 26

New comment
 
Robot_al :
Leksus, I have a question for you. I want to find out more about neural networks... where do I start? What book, what program should I take to study it? Don't send me to Google, of course. Advise me on websites...hey, hey, Morda Raptors...it's not advertising, for fuck's sake...I need...I really do...please help...
 
Vinin, thank you. I'll start looking into it.
 
Robot_al :
Leksus, I have a question for you. I want to know more about neural networks, to try them out... where do I start? What book, what program should I take to study them? Don't send me to Google, of course. Advise me on websites...hey, hey, Morda Raptors...it's not advertising, for fuck's sake...I need...I really do...please help...

The nnet package from R. Not only fits the network (the nnet function), but also makes predictions (the predict.nnet function).

If you look more broadly into R (which is a specialized package for trading), you will be surprised to see that strictly speaking neural networks are a classification tool, which then, with good work, can be used in trading. But neural networks are not originally designed for trading. For example forecast package does not have any tools based on networks at all.

If you don't want to try (play), but need a specialty for the rest of your life with a stable source of income, then R is full of tools used by millions of people.

 
Well, EconModel, thank you for not joking. Yes, I need everything that concerns artificial intelligence... and I have solved the forex problem (= as it seems to me)... if forex and everything with this pricing model will feed me, I will be only glad... but I would like to create more complex algorithms... I lack education... I am a vocational student, only a vocational student.
 
Robot_al :
Well, EconModel, thank you for not joking. Yeah, I need all the AI stuff... If forex and everything with this pricing model will feed me, I'll be glad... but I'd like to create more complex algorithms... I lack education... I'm just a vocational student.

Hmm... It turns out I don't qualify as a vocational student. It turns out that vocational students are now very different.

And the forex problem is not solvable today. It is realistic to talk about the probability of prediction, which, by the way, the neural network does not give.

 
And the forex problem is not solvable today. It is realistic to talk about the probability of a prediction, which, by the way, the neural network does not provide.

Yeah, yeah. Maybe I was being too loud about "solving forex"...of course "predicting forex"...like, solving=100/100 probability...prediction=n/100, but agree, it all depends on the margin of error you're comfortable with...it's one thing to be 50/50 and another to be 90/100.

At the end of the day, nothing can be measured accurately...when I'm cutting sausage, why should I "catch microns"?

 
I think you will be the first to secure a bright future for yourself with the help of networks, unless, of course, you join the ranks of the disillusioned.
 

2 Robot_al

Vinin gave it just the right way. Right in the eye. I wanted to give you something similar.

P.S. How I like such friendly forums. To a direct question to a particular forum member, everyone else responds to the best of their ability. I'm going to wipe away a stingy man's tear.

 
Robot_al :
Vinin, thank you. I'll start looking into it.

You'll find a lot of interesting things for yourself https://www.mql5.com/ru/forum/127517
 
Vinin : Ф. WassermanNeurocomputing: Theory and Practice
The book is definitely good, but it does not relate to earning money in financial markets. You can certainly read it for the sake of self-education, but, again, it will not help you to earn money in financial markets ))))
1...192021222324252627282930313233...40
New comment