Market model: constant throughput - page 13

 
gip:

...........

How do we evaluate information in the market? Information changes the market, so the arrival of information gives deviations from some basic values. And it is desirable to have an adequate criterion, something that is measurable and repeatable on the market. You can find several such things. Say range or mileage, maybe some other characteristic, the same distribution to compare torque with the average. The deviation can be non-linear. And ambiguous. And an increase in mileage or range, is an input, and a decrease too.

There is no talking about uniformity of information coming in.

Get a graph and compare it to a calendar and news graph. That's pretty much it.

I think this trick has been done a hundred times already, you can probably search the web.

Information is not "distributed" evenly over the market (i.e. in the same volumes) and for all participants at the same time. Also, it cannot be assumed that once the market has received some information, it will act rationally based on this information. The result (function) of the information and the market's reaction to it is the price. After the price appears in the terminal as a tick, it has become information, the basis for this or that decision and the above-mentioned cycle is repeated to form a new price higher/lower than the previous one. And the question where the price moved is answered by a particular market participant who perceives the received information according to his own understanding ("right" or "wrong").

Thus, if we test the market for a strong form of its efficiency, we need to learn how to (1) collect absolutely all information (and assume that the whole market does so) and (2) determine that each particular information always has an unambiguous influence. on the actions of all market participants (i.e., to put it simply, to assume that a situation where the price goes up and this is a signal for some to enter the market to go up and for others to close long positions - is impossible, because everyone is smart and simply will not allow this).

All this makes the approach to studying the market discussed in this thread not much more effective than, say, studying classic Price Action patterns....

 
sanyooooook:

I certainly understand that the man writes well, let him write, but so that others could understand him, he has a dry text, I understand that he is clever and probably well versed in the correlations of currency pairs, has done much research in this area and will probably soon receive a Nobel Prize for his research, but how does he apply his knowledge and findings in practice, why all his efforts, is there any practical sense from all this?

Reshetov's laurels do not give him any peace. =)
 

Hm, I wonder if anyone has made an indicator of their own information?

It would seem that if there is an assumption about the permanence of information, there should be an indicator to investigate it :)


P.S. By the way, an entropy indicator would probably do.
 

What would it be for? There is no way to test the hypothesis of the permanence of information in the market with this tool. (And what for?).

And then it is not clear what to apply it to, something tells me that the application of these tools to bars is wrong.

 

"- Mr Faraday, everything you have shown is very interesting. What is it good for, this electromagnetic induction of yours?

- What's a newborn baby good for?"

What I think. The original message is wrong. Or rather, it may be correct, but in terms of abstract, non-applied to the purpose of the trade (making a profit) information. But even that is not the case.

Profit making is possible when you work in the Tao of a given context. Ergo, the information value in the market is the signals that bring the system into a different state. Or, if you like, changing the rules of the game. Working within a given context, on the other hand, is a trivial task. This is not what we are talking about.

Do I need to prove that context-setting (rule-changing, rule-throwing to a new ent.level - whatever))) impulses by any consideration should not be smeared in an even layer over time. Ha, even if one raves about Fibonacci time zones, this is not the case. The Fibo series is not 1, 2, 3, 4 etc.

But putting that reasoning "for context" aside, even then, think for yourself, is the information value equal between the period when markmakers hold the corridor jerking each other off and the period when important news comes out that dramatically changes hmm... context? )))

===

In short, so far I don't see the point of this (with such a drawn out premise) approach. So far it reminds me of a sophisticated form of syllogism. Pardon.

 
Man, that's some fucked-up idea.)
 
Moped didn't dickfix, he provided the link and it's from somewhere else quite far away.
 
gip:

1. What would it be for? To check the hypothesis about constancy of information in the market with this tool is impossible.

And then it is not clear what to apply it to, something prompts that the application of these tools to bars is incorrect. 2.

1. I don't actually believe in trivial constancy. But I don't rule out the existence of some kind of pattern.

2. I would not say "unclear" but "unknown" :) . That is, there is a potential field for research

Svinozavr:

Do I need to prove that context-setting (rule-changing, entrancing - whatever))) impulses for whatever reason should not be smeared in an even layer over time.

But if we set aside this reasoning "for the context", then again, think about it yourself, is the information value equal between the period when the markmakers hold the corridor, jerking each other off, and the period when important news comes out, dramatically changing hmm... context? )))

If you look more closely at the definition of intrinsic information, you can see that the rarer the event, the more of it there is.

 
Candid:

1. I do not actually believe in trivial constancy. But I do not exclude the presence of some regularities.

2. I would not say "unclear" but "unknown" :) . That is, there is a potential field for research

If you look more closely at the definition of intrinsic information, you can see that the rarer the event, the more of it there is.


more often, less often, but what if time is not taken into account?
 
He might not like it.
Reason: