From theory to practice - page 47

 
Евгений:


And then, for some reason, you open an account with real money. And when you lose, "I told you money wasn't important to me".

It's all so strange.

I opened an account for my daughter. If everything goes well, it will be a New Year's present for her. I'm not trying to do it for myself. She asked me to take over - well, what can't you do for your daughter...
 
Alexander_K2:
I opened an account for my daughter. If it works out, it'll be a New Year's present for her. I don't do it for myself. She asked me to do something for her daughter.
So she says. Daddy, I want a grail for New Year.
Keep the medal. But the grail is mine))))). Next june will ask for the moon. Then a time machine.
Daddy can,
Daddy can,
Anything.
Because Daddy's not a wimp, he's a quantum physicist. It's a two-finger drive to the future for him. )))))
I'm afraid to ask what she wanted in years past.
The garter alley is probably the most innocuous of all. )))) I'd even get married. With a father-in-law like that, it's like having Christ behind my back)))).
 

ILNUR777:

It's a two-finger race to the future for him. )))))


I can race to the future too, only I haven't learned how to go back yet. )))

 
ILNUR777:
That's what he says. Daddy, I want a grail for the new year.
Keep the medal. But the grail is mine))))). Next june will ask for the moon. Then a time machine.
Daddy can,
Daddy can,
Anything.
Because Daddy's not a wimp, he's a quantum physicist. It's a two-finger drive to the future for him. )))))
I'm afraid to ask what she wanted in years past.
The garter alley is probably the most innocuous of all. )))) I'd even get married. With a father-in-law like Christ behind the bosom)))) .

.

 
Alexander_K2:

I can repeat the following - there is a link between two quotations received in series, a so-called "memory". We have 2 degrees of freedom. The increments form a non-Markov chain whose probability distribution must be described by a t2-distribution. It has to be there. Nothing else is there and can't be!

Do you mean the relationship between returns or between prices?

And what do you mean by "two degrees of freedom"?

 

The Big Bang Theory

Sheldon - "Penny, I'm a Physicist! I know how EVERYTHING in this universe functions!"

Penny - "Okay, what is Radiohit?"

Sheldon-"Well I know how EVERYTHING in the universe functions!"........ :))

 
Alexander_K2:

Here, Vladimir - especially for you.

This is the histogram of EURJPY increments for the past week at exactly exponential time intervals with quotes averaging


Here are the statistics

Column D shows real values of probabilities

Column E represents the calculated values for t2-distribution.

What more proof do you need????????

1. Proof of what? You didn't say.
2. The EURJPY.zip archive attached to your message 442 does not contain statistics at all, but 8 Mb of data to extract them.
3. The statistics itself is filled with Olympics-80 logo in the upper picture, as usual, the most part of it is not used, it is extremely uninformative. Isn't it possible to plot logarithms of values on the ordinate axis in Wissim?
4. The figure below shows the results of the analysis of the data you presented (your bottom figure) in terms of the nature of the sample frequencies. The exponential approximation of columns B and C does not fit well because of the heavy tails, as I told you before. Since you calculated theoretical probabilities for t2-distribution, I conclude that you wanted to prove the closeness of actual "probabilities" to them - closer than for any other distributions.
But the exponent turned out to be closer to the actual ones, see the right figure.


The actual data you presented show that the distribution type hypothesis you selected as the main hypothesis contradicts them.

 
Vladimir:

1. Proof of what? You didn't say.
2. The EURJPY.zip archive attached to your message 442 does not contain statistics at all, but 8 Mb of data to extract them.
3. The statistics itself is filled with Olympics-80 logo in the upper picture, as usual, the most part of it is not used, it is extremely uninformative. Isn't it possible to plot logarithms of values on the ordinate axis in Wissim?
4. The figure below shows the results of the analysis of the data you presented (your bottom figure) in terms of the nature of the sample frequencies. The exponential approximation of columns B and C does not fit well because of the heavy tails, as I told you before. Since you calculated theoretical probabilities for t2-distribution, I conclude that you wanted to prove the closeness of actual "probabilities" to them - closer than for any other distributions.
But the exponent turned out to be closer to the actual ones, see the right figure.


The actual data you presented show that the distribution type hypothesis you selected as the main hypothesis contradicts them.

Bloody hell... Vladimir - you are a real professional, I am just a child compared to you... So it's a two-sided geometric distribution? A-ya-ya-ya-ya... Shame on me. Don't scold the old man - he played the best he could...
 
No, it's probably a megatron complex somatic Eisenhower multidimensional distribution. Oh, yeah! That's it!
How have I not seen this before.
 
Alexander_K2:
Damn it... Vladimir - you're a real professional, I'm just a small child compared to you... So it's a bilateral geometric distribution? A-ya-ya-ya-ya... Shame on me.

Why do you need names? Brownian - not Brownian, chi-square or Statistic - what difference does it make... Why not use the identified regularity, even if it does not fit any distribution (of probability), even if the frequency does not have statistical stability (remember the reference to Gorban) and classical probability theory is not applicable. And what, to be afraid of that?

And why are you in such a hurry? Check my conclusions, double-check them, just in case.