From theory to practice - page 410

 

Has anyone ever wondered why the intensity of night-time trading is lower?

The fact that it is lower has already been discussed for 400 pages, but why?

The logical chain of reasoning is as follows: spread is wider at night, there are less traders who want to trade with higher spreads.

And why is the spread higher? Well, market makers widen the spread due to uncertainty.

And what is the uncertainty for the market makers (remember, these are those who set limits on forex, we all conditionally trade with markets).

And here is an open question where the logical chain breaks down: Why the uncertainty for the MM is higher at night than during the day?

In other words: Where do they look that this cheat sheet is not available at night?

A similar question: why are there forex sessions, when trading is all day long and traders are located all over the world?

If you answer the question, you will understand what kind of scale is needed.

 
I've even prepared an answer for whoever answers:
Of course, it's elementary Watson.
 
Maxim Dmitrievsky:

it's just that there's an opinion that "memory" isn't a bad thing

e.g. such processes have centres of mass, which are sometimes well distinguishable. I have been trading this way for several years without automation

it is possible to automate with a validation section, define error tolerances and collapse automatically to try. All you need for that is an inverted acf, and you can also add a normal one with a history search for additional validation on similar pieces of history

Nah Max. There are 2 types of "memory" on the market, as it turns out:

1. Memory of the movement of the price itself. It's like a heavy particle that continues its movement despite the effects of surrounding light particles. A trend and "heavy" tails of distributions are formed.

2. Memory of appearance of events - occurrence of quotes. I.e. as if a heavy particle, is exposed to light particles at certain times too. As if the light particle remembers - when it is necessary to affect the heavy particle. Well, as if market participants are not making transactions chaotically, but purposefully.

There is no such thing in Markovian diffusion theory. The whole mathematical apparatus is designed for the impact to be made chaotically, without memory.

And whether the heavy particle will change the direction of motion or not is another question.

So, there should be no memory of events. Otherwise, everything goes to hell, and the branch can be closed.

 
Nikolay Demko:

Has anyone ever wondered why the intensity of trading at night is lower?

It is because those who trade more intensively do not work. Some people have morning, others have afternoon, even those who are in the same time zone rest at night, not work.

Isn't it so?

 
Evgeniy Chumakov:

It is because those who trade more intensively do not work at that time. Some traders have morning, others have afternoon, even those who are in the same time zone rest at night instead of working.

Not so?

So it's morning and night for some, it should be even, but no, when Chicago opens, the soul goes to heaven.

Lloyd's story.

 
Alexander_K2:

No, Max. There are two types of 'memory' on the market, as it turns out:

1. The memory of the movement of the price itself. It is like a heavy particle that continues its movement despite the influence of surrounding light particles. A trend and "heavy" tails of distributions are formed.

2. Memory of appearance of events - occurrence of quotes. I.e. as if a heavy particle, is exposed to light particles at certain times too. As if the light particle remembers - when it is necessary to affect the heavy particle. Well, as if market participants are not making transactions chaotically, but purposefully.

There is no such thing in Markovian diffusion theory. The whole mathematical apparatus is designed for the impact to be made chaotically, without memory.

And whether the heavy particle will change the direction of motion or not is another question.

So, there should be no memory of events. Otherwise, everything goes to hell, and the branch can be closed.

This is all the legacy of Bachelier and efficient market theory, and they "siphon off" every time some crisis starts, which according to their models is almost unbelievable. Garcham and Figarcham have not been able to conquer this topic for decades.

And memory in general has long been grasped by traders through self-similarity on charts and structure centres. If a process has memory, how can it be expressed in the case of quotes? in only one way - through the formation of self-similar figs. And they don't necessarily form a trend, they can also be flat. A simple example here:

The characteristic feature of all market patterns - at a certain moment they start to mirror themselves. An inverted acf would work well for that.

Just in case nothing works out of memorylessness, here is a peculiar alternative. The phenomenon is observed and reproduced regularly.

https://www.mql5.com/ru/forum/188282/page2#comment_7747448

Библиотеки: Кроссплатформенная библиотека оригинальных математических функций
Библиотеки: Кроссплатформенная библиотека оригинальных математических функций
  • 2017.11.21
  • www.mql5.com
Кроссплатформенная библиотека оригинальных математических функций: Автор: fxsaber...
 
Alexander_K2: So - there should be no memory of events. Otherwise, everything goes to hell, and the branch can be closed.


It's all right to think that vegetables are cheaper in summer than in winter. And if there was no memory, they would cost differently every time. Today you buy them for 30 roubles, tomorrow they cost 200 roubles, etc.

 
Evgeniy Chumakov:


And if there was no memory, they would cost differently every time. Today I bought them for 30 roubles, tomorrow they cost 200 roubles, and so on.

A very timely question.

A farmer needs money for fuel for planting, but he does not know whether he will have a harvest this year or whether a drought will dry everything out, so who takes the risks of the loan?

Or God forbid, a monstrous harvest will bring down the prices, so that he will not have enough to pay the loan for planting?

PS PS let's get to the point, the market trader buys and sells at the same time, he is the liquidity provider, and the market goes one way, the prices rise and MM sells, if the market goes back, no problem, he recovers it all, but if not - who compensates for the losses of MM, who has taken a position against the trend?

 
Alexander_K2:

No, Max. There are two types of 'memory' on the market, as it turns out:

1. The memory of the movement of the price itself. It is like a heavy particle that continues its movement despite the influence of surrounding light particles. A trend and "heavy" tails of distributions are formed.

2. Memory of appearance of events - occurrence of quotes. I.e. as if a heavy particle, is exposed to light particles at certain times too. As if the light particle remembers - when it is necessary to affect the heavy particle. Well, as if market participants are not making transactions chaotically, but purposefully.

There is no such thing in Markovian diffusion theory. The whole mathematical apparatus is designed for the impact to be made chaotically, without memory.

And whether the heavy particle will change the direction of motion or not is another question.

So, there should be no memory of events. Otherwise, everything goes to hell, and the branch can be closed.

So the memory of the market is permanent and there is no getting away from it. I remember that Matemat assumed a certain number of bars, but if the reference point was current, then this proportion will be saved from any other reference point. As a collective farmer I understand and can demonstrate.

 

eurusd


Live, will it be a take or a stop. By some calculations the reversal should have been on the vertical line. That there are two standard deviations (from something) . So far between 2 and 3 sko .

By the way the previous reversal was only a minute late .

By 8:22 pm MSC should turn around

Reason: