The market is a controlled dynamic system. - page 123

 
avtomat:

GAMMARASP(x;alpha;beta;integral)

x is the value for which the distribution is to be calculated.

Alpha is the parameter of the distribution.

Beta is the parameter of the distribution. If beta = 1, GAMMARASP returns the standard gamma distribution.

Integral is the logical value that defines the form of the function. If integral is TRUE, then the GAMMARASP function returns the integral distribution function; if this argument is FALSE, then the distribution density function is returned.

I made a typo, corrected it, look it up. Instead of H I typed AND in the last formula, I was tired at work, I didn't check it, sorry.

It is written::

yosuf:

How do you calculate the integrals? Please give the numerical values of I, P and H at the point of greatest divergence and the value of t at that.

Try counting them like this, for example, when t=2:

AND = GAMMARASP(t/t;n;1;1) = GAMMARASP(2/0.577292852;2.954197002;1;1)=0.682256914 ----------------- integral distribution function

P = GAMMARASP(t/t;n+1;1;1) = GAMMARASP(2/0.577292852;3.954197002;1;1)=0.465336551

AND = GAMMARASP(t/t;n+1;1;0) = GAMMARASP(2/0.577292852;3.954197002;1;0)=0.216920364 ----------------- distribution density function

N + I = 0.465336551 + 0.216920364 = 0.682256915

Where do you see the discrepancy?

There should be:

yosuf:

How do you calculate the integrals? Please give the numerical values of I, P and H at the point of greatest divergence and the value of t at that.

Try to calculate them like this, for example with t=2:

I = GAMMARASP(t/t;n;1;1) = GAMMARASP(2/0.577292852;2.954197002;1;1)=0.682256914 ----------------- integral distribution function

P = GAMMARASP(t/t;n+1;1;1) = GAMMARASP(2/0.577292852;3.954197002;1;1)=0.465336551

H = GAMMARASP(t/t;n+1;1;0) = GAMMARASP(2/0.577292852;3.954197002;1;0)=0.216920364 ----------------- density function

N + N = 0.465336551 + 0.216920364 = 0.682256915


 
Mathemat:

Two people arguing with each other - in different languages.

One is in the language of well-known mathematical formulas, and the other is answering him with the symbols used in Excel.

Fuck, how are you going to understand each other?

Since the language of mathematical formulas is more common and understandable, let Yusuf lay out how he understands the symbolic formulas of Excel in the language of integrals and exponents.

Otherwise you will argue ad infinitum and you will not understand each other.



You must know that the Exel variant does not differ from the usual, "common" notation, for example:


 
Forex is not just people but millions of computers with their own software bugs, I wonder if there is any way to account for this mathematically?
 

Well, let's go from the beginning, from the beginning, from the stove ;)

.

.

Newton-Leibniz formula.
Let the function f (x ) be continuous on the closed interval [a, b ]. If F (x ) is the first-order function f (x ) on[a, b ], then


.

.

.

.

.

i.e. the value H=0.216920364 coincide, but the matcadian integral P=0.268635468 and the Excel P=0.465336551 do not coincide. -- hence the further discrepancies.

My guess is that Excel is wrong and Matcad is correct.

 
konda:
Forex is not only about people but also about millions of computers with their own software bugs, I wonder if this can be accounted for mathematically?


Yes, we can. Due to the fact that "there are millions of them", the task is actually simplified, and mathematical statistics can easily handle it. But you have to understand the limits of what can be done and how it can be done.

These "mistakes in the millions" are what ultimately give rise to the noise component of the movement.

 
konda:
Forex is not just people but millions of computers with their own software bugs, I wonder if there is any way to account for this mathematically?


Unlike humans, computers don't make mistakes, they just do their software)))) Written by humans.
 
Sepulca:


Unlike humans, computers don't make mistakes, they stupidly do their own program)))) Written by humans.

well, they also malfunction, sometimes by human design and sometimes quite "sincerely" ;)))
 
And then there's the wacky brains of bank and fund managers and the equally wacky brains of traders, shall we count them as noise too?
 
Sepulca:


Unlike humans, computers don't make mistakes, they stupidly do their own program)))) Written by humans.

They don't make programs, they execute them like superfast idiots. :)
 
konda:
And then there's the wacky brains of bank and fund managers and the equally wacky brains of traders, shall we count them as noise too?


Sure. Up to certain limits.

Reason: