From theory to practice - page 80

 
Alexander_K2:
Aren't increments over time a process? A process and what a process it is! With expectation = 0 and strict rank statistics.
Where have you seen that increments are tested for stationarity? dispersion is a variance, sigma is also a variance.

You're making a series out of increments?



That would be an artificial series. How can it be traded?
 
Alexander_K2:
Wrong again. Still a known direction! But I won't talk about it. Otherwise everyone will say - I gave you advice, but your pockets are still empty. If you go down this road, you will still get robbed of the spread and the commission. Although... If you try very hard.... No! I'm personally not interested!
Well, build a robot with a takeprofit of 1 and a stoploss of 100. it will not make money.
 
Alexander_K2:
Again, it does not. It also knows the direction! But I'm not going to talk about it. Otherwise everyone will say - I gave you my advice, but my pockets are empty anyway. If you go in this direction, you will be robbed of the spread and commission anyway. Although... If you try very hard.... No! I'm personally not interested!

the bell of distributions (constructed from increments) does not show directions, it only shows the size of the most frequent increments.

 

Thanks, Oleg, but I was asking specifically about standards. Tell me what I should do now with the phrase in this link of yours:

Random mechanisms can be different; more often considered are SS generated by summation of independent random variables or Markov chains. There is no precise , generally accepted definition of S. B.

End of quote.

And I had so naively hoped that the people who reported the SB standard had something specific in mind. Pity.

 
Vladimir:

Thank you, Oleg, but I was asking specifically about standards. Tell me what I should do now with the phrase in this link of yours:

Random mechanisms can be different; more often one considers SS generated by summation of independent random variables or Markov chains. There is no precise , generally accepted definition of S. B.

End of quote.

And I had so naively hoped that the people who reported the SB standard had something specific in mind. Pity.

how to live now?)

 
Vladimir:

Thank you, Oleg, but I was asking specifically about standards. Tell me what I should do now with the phrase in this link of yours:

Random mechanisms can be different; more often one considers SS generated by summation of independent random variables or Markov chains. There is no precise , generally accepted definition of S. B.

End of quote.

And I had so naively hoped that the people who reported the SB standard had something specific in mind. Pity.


There is no standard. But do you necessarily need a standard? Can't you do without a standard?

But "a special kind of random process which can be interpreted as a model" -- as a model is fine. And do all fully working research models have an approved standard?

 


.


Mathematical Encyclopaedic Dictionary. Moscow: Sov.Encyclopaedia, 1988. -- 847 с.

 
Vladimir:
And this is going to be a random rambling model? Exactly SB with a double logarithm in the denominator under the square root?

I do not know what it will be, you are the mathematicians and I'm just out for a walk ))

The only thing I know is that PRNG will not trade, and eventually any strategy on PRNG will have to go bust.

 
Vladimir:
Why generate? Are there not enough archive sources. I don't understand the purpose of identifying the type of distribution at all. What good is the Kolmogorov-Pearson criterion to give so much (a) and so much (b) we need for 97% confidence that we have not rejected a false hypothesis. When estimating by method of moments (if you estimate by method of maximum likelihood or Bayes, a and b will be different). Who needs the whole distribution? Even Alexander_K said he cares about tails first and foremost. Coin flips, by the way, have also been done up to 50,000 times by at least a dozen different scientists.

Read my post above.

 
Nikolay Demko:

I do not know what it will be, you are the mathematicians and I'm just out for a walk ))

The only thing I know is that PRNG trading will not work, and eventually any strategy on PRNG must go bankrupt.


Why not?

Yes, it will. It's not about trading every sneeze.