Fractals, fractal structures, their graphic images + Canvas - page 4

 
Uladzimir Izerski:

The price behaviour on any TF is identical. The lower the TF the more accurate the forecast.

No way! Full stop.

 
Physicists and mathematicians in substantiation of definition of a fractal allocate the basic property - self-similarity, in fact the fractal is "border" of two opposite tendencies and depending on prevalence of one of them (on an example of the market) the name short or long tendency is specified.

The medium of emergence

- I have no data for 250 million years on how the coastline of the Norwegian fjords has changed during that time, heavy option ;) , snow \ snowflakes - forming a "transition" to another state vapour-air environment goes to a solid state , and the constituent conditions are systemic in nature , density \ temperature \ presence or absence of wind create the characteristics of the snowflake shape . Markets - the USA dollar is the systemic link, the homogeneous not homogeneous elements of exchanges/trade operations/unions state not state are interconnected in the vast majority of transactions involving the US dollar.

 
Alexander_K:

Pops!

I've already explained - a fractal is the probability density of a stable, infinitely divisible distribution. That is, whichever slice of the process you take, any sample volume, you will always see a self-similar structure. An example is Brownian motion.

The market is NOT a self-similar structure. Mandelbrot, along with his associates, have deliberately brainwashed sufferers into claiming that trading will be the same on any TF. Crazy lovers have tried to scalp on low TFs... The result - Mandelbrot and his companions are stupidly lining their pockets at the expense of fools. That's it!

Oh, the unconcerned negativity is coming, so we are in the right direction.
 
т
Dmitry Fedoseev:

There's a running theme of 'radiation indicator' - it looks pretty funny. It seems to be in the articles.

Yes, thank you! Interesting topic, here it is https://www.mql5.com/ru/articles/26. There is a parallel, with one of my notions (which I am not going to particularly defend, just one of the models), that every point reached by a price is a source of probability amplitude waves, the whole set of which interfere in the whole space. Actually, Dukas also imagined something like that, when he identified the motion of quotations with the motion of certain emergent particles in the corresponding space, which interacted (collided) there. But if we represent these particles in a quantum way, then my idea will appear.
Построение излучений индикаторов в MQL5
Построение излучений индикаторов в MQL5
  • www.mql5.com
Наверняка, многих трейдеров и разработчиков торговых стратегий (ТС) интересовали подобные вопросы: Поиск ответов на эти вопросы в результате привёл меня к созданию нового направления исследования рынка: построение и анализ излучений от индикаторов. Чтобы было понятнее о чём идёт речь, предлагаю посмотреть на следующие рисунки. Здесь изображены...
 
Nikolai Semko:

Yes, I've been eyeing fractals for a long time too.

Of course not only a price, but our whole world can be described by a fractal.
(1) But it seems to me that it is unreal to calculate the formula of this fractal. Only God can do it. Anyway, I don't have enough guts for sure. ))

Clearly structured fractals won't do here.

The shoreline is also considered a fractal. But it is a random fractal structure. It cannot be predicted by modern mathematical methods. The same is true for the price.

(2) All that is now called fractals in forex is bullshit.

But personally I think that the more interesting application of kanvasses is in 3D-quotes or something like a heat map.

Something like this:


Such 3D quotes would be much more informative than traditional ones. There will be no need to switch between different TFs in order to understand the bigger picture. Just one general TF which will contain all the information from ticks to all the multi-year history.

I have no blank spots in my understanding of how this can be implemented. Will definitely implement it sometime when I have time. I don't want to talk about it in more detail ahead of time.

(1) First we have to understand conceptually what kind of fractals we may be talking about, then statistically identify them, and then work out an algorithm for their identification and an algorithm for their visualization.

(2) Totally agree that it's wrong to call them fractals, it's just a ponce.

 
Veniamin Skrepkov:
Physicists and mathematicians in substantiation of definition of a fractal allocate the basic property - self-similarity, in fact the fractal is "border" of two opposite tendencies and depending on prevalence of one of them (on an example of the market) the name specifies a short or long tendency.

The medium of emergence

- I have no data for 250 million years on how the coastline of the Norwegian fjords has changed during that time, heavy option ;) , snow \ snowflakes - forming a "transition" to another state vapour-air environment goes to a solid state , and the constituent conditions are systemic in nature , density \ temperature \ presence or absence of wind create snowflake shape characteristics . Markets - the systemic link is the USA dollar , the homogeneous not homogeneous elements of exchanges/trade operations/unions state not state are interrelated in the predominant majority of transactions involving the USA dollar.

In other words a fractal is an unstable position that switches from one state to another to compensate for the remainder of the division. For example, the Fibonacci ratio tends towards 1.618... but never reaches it, which is where the fractal self-similarity comes in.
The market has this too. In essence, the market is designed to distribute funds as evenly as possible between participants. And if there is no change in the number of participants and the volume of funds, then after a certain number of iterations the funds will become scarce. But there is an emission in the market (the currency is always simulated) and the number of participants changes, which prevents the equilibrium distribution of the means and each time out of equilibrium.
 
Alexander_K:

Pops!

I've already explained - a fractal is the probability density of a stable, infinitely divisible distribution. That is, whichever slice of the process you take, any sample volume, you will always see a self-similar structure. An example is Brownian motion.

The market is NOT a self-similar structure. (1) Mandelbrot, along with his associates, have deliberately brainwashed sufferers into claiming that trading will be the same on any TF. Crazy lovers have tried to scalp on low TFs... The result - Mandelbrot and his companions are stupidly lining their pockets at the expense of fools. That's it!

Hello dear colleague!

I'm not going to argue on point (1). Whether there are fractals in market quotations or not should be proved empirically. I see it in carrying out some specially developed clustering at different time scales. If the elements of the set of such clusters are similar, then this structure can be considered a fractal.

 
Aleksey Ivanov:

Hello dear colleague!

I'm not going to argue on point (1). Whether there are fractals in market quotations or not should be proved empirically. I see it in carrying out some specially developed clustering at different time scales. If the elements of the set of such clusters are similar, then this structure can be considered a fractal.

By the way, my self-adaptive robot is built according to the fractal principle, i.e. it uses the regularity but changes the scale of finding this regularity and perfectly adapts to any tool. On forex since 2008 with the same settings for 28 instruments it is stable and on the fund since 2013 on 28 stocks with the same settings it is stable.
 
Aleksey Ivanov:

I see it as carrying out some kind of clustering on different timescales, specifically designed for this purpose. If the elements of a set of such clusters are similar, then this structure can be considered a fractal.

Hi Alexey!

Yes, this is the right direction. Not to indiscriminately consider the market as a self-similar structure, but to reveal it by working with different time scales. I'm sticking to my opinion - time is primary on the market. Time to observe the process, time between the quotes, etc. It all forms some kind of space-time structure. Maybe 3D-modelling in dynamics will help to understand it...

P.S. I'm looking at your work - everything seems correct and canonical so to speak. I hope to see your personal trading signal. Good luck!

 
Alexander_K:

Pops!

I've already explained - a fractal is the probability density of a stable, infinitely divisible distribution. That is, whichever slice of the process you take, any sample volume, you will always see a self-similar structure. An example is Brownian motion.

The market is NOT a self-similar structure. Mandelbrot, along with his associates, have deliberately brainwashed sufferers into claiming that trading will be the same on any TF. Crazy lovers have tried to scalp on low TFs... The result - Mandelbrot and his companions are stupidly lining their pockets at the expense of fools. That's it!

Probably, the discussion should go in the direction of that - market data have a fractal structure. The probability levels out as the fractal structure appears, i.e. the probability changes into a fact if 0=point of fractal appearance, 1= extremum, 2= model correction and depending on TF to analyze fractal ranking in relation to other "chains" of fractals.