From theory to practice - page 1355

 
Yuriy Asaulenko:
Yes, yes. Even Einstein said - God does not play dice.
By the way, the Present does not exist. A recorded event is already the past).
A_K, do you know what he said that about? Don't look on the internet. Fucking physicist.)

I see, but you can't do without the H function. It is a link between functions B and P. It appears, as I have already noted, when integrating function B.

It is necessary to define the notion of the "Present". It is not as obvious as it seems and depends on the choice of time unit in the H function.

 
Yousufkhodja Sultonov:

All our misfortunes stem from the fact that we are all hostages of the B-function, of which we know the type, but we are not allowed to know its parameters! What principle is used to form parameters of the B-function is a super-task which mankind will probably never be able to solve.

Well, why not? The future is an unknown function of many variables, most of which we do not know. So it is a random process.
It is already known that even in classical physics it is impossible to predict anything reliably. I cannot quote, because I am writing from a mobile phone.
 
Yousufkhodja Sultonov:

For a view of the functions seehttps://www.mql5.com/ru/articles/250

Dear Yusuf, From your link I read"Nevertheless, to be able to make correct decisions about buying or selling goods, including currency or shares, at the current moment in time, one must be able to anticipate and predict with a reasonable degree of accuracy the behaviour of the market price in the future."

Please write what do you think is the differencebetweenpredicting with a reasonable degree of accuracy and assuming with a reasonable degree of probability?

 
Yuriy Asaulenko:
Well, why is that? The future is an unknown function of many variables, most of which we do not know. That is, it is a random process.
It is already known that even in classical physics it is impossible to predict anything. I can't quote you as I'm writing from my mobile phone.

In fact, there are no "random" processes. With this term we are forced to admit our impotence in knowing the B function.

 
Yousufkhodja Sultonov:

In fact, there are no "random" processes. With this term, we are forced to admit our impotence in knowing the B-function.

Wrong. Random processes have been and will be.
We cannot, by definition, know everything and absolutely precisely. At least that one thing is randomness.
A billiard ball, by the laws of billiards, has to hit the pocket. But these are not the only laws in our world. And by those laws, it won't hit the pocket.)
 
apr73:

Dear Yusuf, from your link I read"Nevertheless, to be able to make correct decisions about buying or selling commodities, including currency or shares, at the current moment in time, one must be able to foresee and predict with a fair degree of accuracy the behaviour of the market price in the future."

Please write what is the difference between:forecasting with a reasonable degree of accuracy and assuming with a reasonable degree of probability in your opinion?

Because, in reality, the course of all processes is probabilistic in nature. How to take into account this probability in formulas is a problem, over which I think incessantly, but, I do not find an acceptable solution, to be finally convinced in "God does not play dice". It is not clear when and why the dice are wedged into the process?

 
Yuriy Asaulenko:
Wrong. Random processes have been and will be.
We, by definition, cannot know everything and absolutely precisely. At least that one thing is chance.
A cue ball, by the laws of billiards, has to hit the pocket. But these are not the only laws in our world. And by these laws, it will not fall into the pocket.)

Randomness is a manifestation of necessity. (K. Marx). It is an unacknowledged regularity. On the other hand, the random motion of gas atoms leads to the Boyle-Mariotte law. Paradox.

 
Yousufkhodja Sultonov:

In reality, all processes have a probabilistic nature. How to take into account this probability in formulas is a problem, over which I think incessantly, but I do not find an acceptable solution to be finally convinced in "God does not play dice". It is not clear, when and why the dice are wedged into the process?

On this issue I found the opinion of a very respected person, I hope many will understand whose words these are:

"In fact you can predict anything you want, there are no parametric limits.But in parallel to the procedure of classical prediction there is (but not in opposition to) also a slightly different procedure - the procedure of evaluation of correspondence to something.What is the difference between them? In classical forecasting, we are trying to estimate parameters of what the bazaar will give us in the future with one or another probability.The procedure of conformity assessment, although also based on statistical methods, is that we do not estimate forthcoming parameters of bazaar but parameters of the system itself.At first glance, the difference between the two statistical procedures is very arbitrary and that is why it often generates various,sometimes heated debates about harm or benefit of forecasting. However, if you look more closely, there is a difference between them and in practice it is really essential.What is the essence of this essentiality? In a classical prediction sooner or later there comes an effect when the desired expectation in the mind of the expectant becomes almost axiomaticand it has a very negative impact on the result, because every time it will mean an attempt to have a showdown with the bazaar or to prove him/her right.Compliance assessment, on the other hand, does not make any predicates with respect to the bazaar, but continuously compares a priori received statistical targets for the system under playwith what the bazaar presents to us at this very moment.In the latter case, statistical advantage is assessed by the degree of closeness of a prioritargeting inherent to a given system and current values from the bazaar, whereas in the former case statistical advantage is assessed by the degree of closeness of projected future and current values of the bazaar.If you think about it, the second implementation is more preferable because it atrophies harmful urge to prove anything to Bazaar anduses only principle - if it matches enough - we trade, if it does not - we wait for next trading opportunity.Besides,system targetets defined in the procedure of matching give rise by themselves to estimate other parameters of a possible trade, whilein the classical forecasting, even with a very successful forecast, we can rely on the only parameter important for us - the entry".

 

Here are yesterday's currency values.

Change is the absolute value of the currency change for the period t = 1440 minutes

buy & sell is sum of positive and negative change for the period separately


And this is Change in dynamics for a period (sum of changes for a period at each step)