Machine learning in trading: theory, models, practice and algo-trading - page 1666

 
mytarmailS:
Yeah, I guess I'm just dumb and wishful thinking.
That's the second step of reflection. Good for you. You can't bury yourself in your delusions. Good for you for understanding....
 
Mihail Marchukajtes:
This is the second step of reflection. Well done. You should not bury yourself in your own delusions. Good for you for understanding....

Thank you teacher)

 
mytarmailS:
Yeah, I guess I'm just dumb and wishful thinking.

That's normal, we all do it.

Although judging by your screenshots from wells, it all does not prove anything, and the fact that the test equity is not adjusted and that the plum is not just a drawdown, on the potential grail. It all depends on the system, I usually put a system stoploop when the drawdown is twice as much as on the test for the previous year. If the system is losing a week, it does not mean anything at all. The fact that some "gurus" are talking nonsense that "the grail should immediately mow dough and not fail" is a wet dream of some immature boys.

 
Kesha Rutov:

This is normal, we all do it.

Although judging by your screenshots from Wells, all this does not prove anything, and the fact that in the test equity is not adjusted and that the drain - not just a drawdown, on the potential grail. It all depends on the system, I usually put a system stoploop when the drawdown is twice as much as on the test for the previous year. If the system is losing a week, it does not mean anything at all. The fact that some "gurus" are saying nonsense that "the grail should immediately suck dough and not fail" - are wet dreams of some immature boys.

This is bullshit. When the system loses money, you don't know if it's just a drawdown or a systematic drain that will continue, so don't talk nonsense... The system should start gaining right away and period....
 
MytarmailS:

The funny thing is that it died after I trained it with new data...

I also noticed that after training the robot does not always start to trade normally immediately, sometimes I have to skip a day or two of trading and then...

I'll illustrate...

somewhere

1) I totally retrained here

2) The robot made a little profit and then it emptied

But after a day or so it started working again.




Does your robot trade normally at once?

By the way, my approach is very similar to yours

It should always trade the same way, how often do you train it? It should always be the same, I don't trade in real time, I've never seen anything like that.

For example, on 4 profitable transactions 1 unprofitable and another 2 open, probably as a result of the day will end with zero. But I'm not sad about it, because I was busy with my work all day and did not have time to monitor the situation, and it is not needed))), the system itself monitors everything.
 
Mihail Marchukajtes:
This is nonsense. When the system drawdown is just a drawdown or a systematic loss, you don't know, so don't talk garbage. The system should start gaining immediately and period....

Hmm, probably have to believe that you are not Jura, too clever, decisive, I am not.

As for me the drawdown may happen in the beginning of a profitable system, no one knows when it is supposed to happen, and you should start worrying when this drawdown exceeds that of the testing period.

 

Where is the fine line between randomness and chaos?

What does a neural network need to be fed to be able to predict PRNG?

Can a network be considered unfit for trading if it is unable to predict PRNG?

 
Rorschach:

Where is the fine line between randomness and chaos?

What does a neural network need to be fed for it to be able to predict PRNG?

Can a network be considered unfit for trading if it cannot predict RNG?

That's the thing: it should not predict RNG, and if it does, then there is an error or a fitting.

 
Kesha Root:

That's the thing: it shouldn't predict the RNG, and if it does, then there's an error or a fitting somewhere.

No, I'm talking about the PRNG. It's a specific algorithm.

Русские придумали гениальный способ обманывать игровые автоматы, от которого казино не могут защититься
Русские придумали гениальный способ обманывать игровые автоматы, от которого казино не могут защититься
  • habr.com
В начале июля 2014 года бухгалтера из казино Люмьер Плейс [Lumiere Place] в Сент-Луисе обнаружили, что несколько их игровых автоматов сошли с ума на пару дней. Программное обеспечение, одобренное правительством, даёт автоматам фиксированную фору математическими методами, так, чтобы казино были уверены в том, сколько они заработают в...
 
Rorschach:

No, I mean the PRNG. It's a specific algorithm.

I agree, we have a pseudo-random process. There's no point in predicting a random one and counting probabilities.