[Archive!] FOREX - Trends, Forecasts & Consequences (Episode 12: February 2012) - page 614

New comment
 
solar:

I strongly recommend that you learn the basics of neural networks, for the market the fewer neurons the better )))) (so as not to say that)
For the market it is definitely better. Kolya will confirm it. But I was gathering my Expert Advisor not for market, but for me. I am so selfish and self-centered. The Wild West is in crisis, and I only care about my own pocket.
 
dssolov:

I have my own trading methods and I don't care about all this neuro bullshit. I see a target at the top or at the bottom and immediately analyse what could be the reason for it to get there. And I don't bother myself with all this nonsense.


extra knowledge never hurts (I haven't seen such cases) )))) so don't boast about what you don't know ))))
 
solar:

extra knowledge never hurts (I haven't seen any such cases) )))) so don't brag about what you don't know ))))
I wasn't even trying to boast!!! It's just that neural networks are more of a probabilistic nature. The market is an interesting thing. And there are very different rules at play. Both mathematical and economic and political.
 
Reshetov:
For the market, definitely better. Kolyan will confirm it. But I didn't build the EA for the market, I built it for myself. That's how selfish and selfish I am.

Just like me ))) ... What's important is the structure of the neural network and what's in the input... all the rest is usually secondary
 
solar:

Just like me ))) ... it's the neural network structure and what's in the input that counts... everything else is usually secondary


neural networks are good (sorry I don't know how to do it), but it could be simpler (without them).

 
dssolov:


And the market is an interesting thing. And there are very different rules at play. Both mathematical and economic and political.

The rules are quite primitive: take the dough from other traders, or else they will take it from you, because every exchange contract has two sides: the loser and the winner. Nothing interesting.
 
solar:

Just like me ))) ... it's the neural network structure and the input that counts... everything else is usually secondary
What the hell? You hear all sorts of nerdy nerd nerd nerd nerd nerd nerd nerd nerd nerd nerd nerd nerd nerd. With neural networks the most important thing is the output, because the output is trading signals.
 
I don't care if you kill me! It's so much simpler than that! Let's say a trader sits in his brokerage house and thinks about how to make more money. He does not care how to earn money, as long as he earns it. So, he gets information via his own channels that the news on Greece will be positive. Does he think about the probability model of the chart or something else? No, he just buys and waits for the maximum price. And that's it! Profit is in his pocket. And he does not care about anything.
 
Reshetov:
What the hell? You hear all sorts of nerdy nerd nerd nerd nerd nerd nerd nerd nerd nerd nerd nerd nerd nerd. The most important thing in neural networks is the output, because the output is the trading signals.

I laughed silently )))) I won't even continue.
 
dssolov:


And then he gets word through his channels that there is going to be positive news on Greece. Does he think about the probability model of the chart or something. No! He just buys and waits for the maximum price.

Kolya likes dumb ones like that if the news is not from insiders, but from analysts.
1...607608609610611612613614615616617618619620621...793
New comment