From theory to practice - page 1100

New comment
 

since there's an artist's competition:

of course by hand, but the point is that the maximum deviations are on the tails, which will make it difficult to separate sqrt from ln

 
Maxim Kuznetsov:

since there's an art competition here:

of course by hand, but the point is that the maximum deviations are on the tails, which will make it hard to separate sqrt from ln

In real life, such a pattern almost never happens.
But it's like pulling mathematics on a price;)
Any function, be it splines or polynomials, even complex conjugate spaces.)
Any signals, at least a simulation of an aeroplane flight along the MA trajectory;)
 
Martin Cheguevara:
In real life there's almost no such pattern.
But to stretch mathematics over the price is like two fingers;)
Any function, be it splines or polynomials, even complex conjugate spaces ahah))

ahahh, not ahahh

such a distribution will not happen, it is quite different. here are 6 pairs:

the only conclusion is that the correlation on large periods is 1, it's just that the pairs are shifted in time:

https://www.mql5.com/ru/forum/27421/page6#comment_11078270

But the distribution is such, because the cotier sometimes goes like this:

https://www.mql5.com/ru/forum/221552/page1085#comment_11076946

 

I once again checked the time intervals between OPEN prices of non-empty second bars (i.e. when there was at least 1 incoming tick in 1 second).

There are no doubts - unambiguously these time intervals form an Erlang distribution. I.e. events (quotes) in the market have an aftereffect, and thus the market is NOT a random process. All arguments on this subject may be ended.

Recall that the exponent divided by erlang gives the Pareto distribution.

I.e. instantaneous velocities in the market form a two-sided Pareto distribution. If only one could extract profit from this distribution... Ugh!

 
Alexander_K:

I once again checked the time intervals between OPEN prices of non-empty second bars (i.e. when there was at least 1 incoming tick in 1 second).

There are no doubts - unambiguously these time intervals form an Erlang distribution. I.e. events (quotes) in the market have an aftereffect, and thus the market is NOT a random process. All arguments on this subject may be ended.

The consequence is determined by memory(autocorrelation of increments), not the distribution of intervals.

 
secret:

Aftereffects are determined by memory (autocorrelation of increments), not by the distribution of intervals.

:))) Citizen Bass! I've been listening to this for two years now!

Random (stochastic) process is not a random walk! It also has such a thing as time. But your memory is a one-dimensional value. And where is time? It is absent!

However, judging by the results, you may be right...

 

I'm just going to fantasise about space-time transformations.

Well, we already talked about Minkowski space.

If we represent the ratio of the cathetuses (tg a) as instantaneous velocities and the hypotenuse as a radius-vector, we can represent the price motion in polar coordinates.

There too, there are likely to be some funny pictures...

 
Alexander_K:

By the way, where is your promised grail for the new year? The Zoroastrian New Year has already arrived and the grail is still missing) Stolen by Schrodinger's cat again and hidden in Russell's teapot?

 
Aleksey Nikolayev:

By the way, where is your promised grail for the new year? It's the Zoroastrian New Year and the grail is still missing) Stolen by Schrodinger's cat again and hidden in Russell's teapot?

I don't think anyone has said what New Year's Eve the magic goblet will be for. And as you know, you wait three years for what you've been promised. Well, in this case it could be thirty years and three years. Until torrents of Erlang form mighty channels and gain incalculable energy. It's a pity only most of local pensioners-forum members will not have to live in this wonderful time.

 
Alexander_K:

:))) Citizen Bass! I've been listening to this for two years now!

A random (stochastic) process is not random rambling! It also has such a thing as time. But your memory is a one-dimensional value. And where is time? It is absent!

However, judging by the results, you may be right...

You guys are crazy. Sometimes you must study the matter :)

Memory, by definition - the dependence of future price changes on the past ones.

In the simplest case - the dependence of increment with time t+1 from the increment with time t.

Time is involved here in the most direct way.

1...109310941095109610971098109911001101110211031104110511061107...1981
New comment
Reason: