The market is a controlled dynamic system. - page 132

New comment
 
ULAD:

It's all quite interesting, but flows seamlessly into a Moving average with a period of ? and ?

And fish do not want to be caught in such a net.

Just the opposite - we need to get away from the proverbial wipes with their lag and zero low predictive effect.

 
sergeyas:
A litre wouldn't hurt here;)


:-)


If with Yusuf on top of it, IMHO, it won't hurt the kajan either!!!!!!!!!!!!!!!!!!!!!!!!!!!

 
sergeyas:

Exactly the opposite - you need to get away from the proverbial wagons with their lag and zero low predictive effect.





You won't get anywhere in that way, and neither will the price. I'm 100 per cent sure the lag will be there.

Digging in this direction is not entirely hopeless, but it is no better than others.

 
ULAD:

It's all quite interesting, but flows seamlessly into a Moving average with a period of ? and ?

And fish do not want to be caught in such a net.

The Moving average is an integral function of the process and it's impossible to get the differential component out of it, due to the lack of functional dependence itself. Anyone wishing to describe any dynamic process must ultimately compare their result with a mashka. A mashka is a finished product, the recipe and dynamics of which we do not know.

In my case, as you correctly suggested, the I function is the prototypical mashka, but now we can break it (I) down into its components - the integral function P and the differential function H. Investigating the behavior of function H, checking the correspondence on AND, we directly come to an integral function B, which has the same nature as function AND, but it is broken down into its components by nature into components H and a function whose name I haven't thought of yet, like we broke history of AND into past P and present H. I suggest calling this function "perspective" - PP, defining it as PP = B - H. Outlook is a future, from which present H has split off.

So we have 5, interrelated functions of 2 genera of the same nature:

Functions of genus 1 from the class I have introduced, a new class of two-parameter superexponential functions:

I - Integral function of history;

B - Integral function of the future;

Functions of the 2nd kind from the class of density (H) and Erleng integral functions (P, PP):

H - Present time function;

P - Integral function of the past;

PP - Integral function of process perspective.

It remains unclear how the original B-function is created, we may never know, because it is created by nature or God, if you like, on the basis of the situation. At most we can recognise it early in the process.

 
ULAD:

You won't get anywhere much that way, and neither will the mashka from the price. I'm 100 per cent sure there will be a lag.

Digging in this direction is not entirely hopeless, but it's no better than others.

The right way of thinking is being driven away by pessimism. You understand the situation and the problem correctly, try to get rid of the feeling of hopelessness.
 
yosuf:

Please also draw the function B = 1-Y here

I agree that the tools, in the interpretation of TAU, are known. But, agree, exactly in such a perspective, space and time, and, the course of processes in them, are presented for the first time. None other family of known functions is capable to undergo the shown metamorphoses and still with clear physical meaning. Only I do not understand a role of the parameter n, while I am thinking intensively. So far it is known that the usual space-time dimension corresponds to the case n =0. And real processes show different values of it. As the model easily copes with both linear and nonlinear regularities, its properties have not yet been fully explored and understood by me. For example, the model can easily describe, like: "straight parabola", "parabolic hyperbola", "hyperbolic exponent", "straight parabolic hyperbola", ...., which we do not understand.



The inflection point of the transition function p is a parabolic point. This point p divides the transition function curve into a region of elliptical points e and a region of hyperbolic points h.

 
sergeyas:
Why don't we give the Yusuf model a zig-zag, tie the start and end points of the calculations to the vertices and collect statistics - maybe something useful will come out?!;)
I don't mind. That is, if I understood correctly, you suggest to investigate history of formation and future behaviour of fractals instead of bars.
 
avtomat:


The inflection point of the transition function p is a parabolic point. This point p divides the transition function curve into a region of elliptical points e and a region of hyperbolic points h.

Fine, let's get closer to understanding the nature and meaning of the parameter n.
 
MetaDriver:

You have no idea what idea you just gave me.

Thank you.
Keep it up!

Your idea that "we will get to the point where there is no Present" turned out to be prophetic. Indeed, there is no place for History (I) and the Future (B) in the system of the 1st kind, or rather, the Present (H) is clearly not present. H can be detected by breaking down AND and B into the components H, P and PP. Thank you for the clue.
 
sergeyas:

I see.

When a new bar arrives and the oldest one drops out, H is formally recalculated depending on what has passed and what has arrived, and that is not good.

That is, the presence of noise and presence or absence of signal at the current moment are not taken into account - everything is in one pile.

There is no reference to characteristic points or levels of the quotation chart from which the transition process started.



This is a cost of any study - noises and outliers cannot be predicted, but they do leave an imprint on the behaviour of the underlying function and, indirectly, can be accounted for by it. Noises, outliers and, other characteristic points, revolve around the underlying function.
1...125126127128129130131132133134135136137138139...551
New comment