From theory to practice - page 153

 
Alexander_K2:

A point of principle.

Everyone is saying that we should check the history and, preferably, test it in real time for 50 years, then yes, OK, we will take it into operation.

Let us think about it. Here we have a plot of the behavior of the probability amplitude (the lower one from those I have posted for AUDCAD), in fact it is the solution of the Fokker-Planck equation for this currency pair.

Read Feynman:

I.e., what we see on the chart is the behaviour of the wave function. It CANNOT change, it is the only one for a particular currency pair. This means that all we have to do is test it on a meaningful sample size (think 1-2 months) and that's it - it's done.

Prepare your pockets, dust them off!


OK, she's the only one.

Let's take the pair EURCAD, it has a unique trajectory.

Would this trajectory change if the news that the instrument "OIL Brend" became cheaper?

Recall that Canada is an oil supplier and Europe is a consumer.

 
Nikolay Demko:

Well, bicycles can be useful. I used to study the market that way. When you invent a bicycle and then find out that it's already been done, it's useful in that you have a deep understanding of the problem.

On the other hand, bicycles show that the inventor is amateurish. A person without having studied the experience of those who came before him, ventures into an unknown realm. Naturally, one will be in for a world of hurt.

Personally, I am impressed by his energy, I even write off some boastfulness to inexperience in developments, so I will not judge strictly.

That's my point. It would be a simple beginner, bullshit question that such a great. It's just strange when a scientific man invents detrended mashka and boundaries of deviation, and does not understand it. Isn't it weird.
 
ILNUR777:
That's what I'm saying. It would be a simple beginner - bullshit question that such a great one. It is just strange when a scientific person invents a detrended dash and the limits of deviation and does not understand it. Isn't it weird.

It's easy for you to invent a machine, but physicists need five years of study for that)))

 
Alexander_K2:

Reading Feynman:

By the style of presentation, probably the Feynman Lectures in Physics. Written, by the way, for students in the mid-60s.

And what did the Landau-Lifshits have to say on the subject?

 
Alexander_K2:

A point of principle.

Everyone is saying that we should check the history and, preferably, test it in real time for 50 years, then yes, OK, we will take it into operation.

Let us think about it. Here we have a plot of the behavior of the probability amplitude (the lower one from those I have posted for AUDCAD), in fact it is the solution of the Fokker-Planck equation for this currency pair.

Read Feynman:

I.e., what we see on the chart is the behaviour of the wave function. It CANNOT change, it is the only one for a particular currency pair. This means that all we have to do is test it on a meaningful sample size (think 1-2 months) and that's it - it's done.

Get your pockets ready, dust them off!

Alexander, in the financial market a hike from point A to point B is a price movement from the highest risk just drained to the next highest risk, which of course is less than the previous one.

Personally, I don't see any parralliance to probability theory, much less Schrödinger's equation.

So I am waiting for the result of real trades.

 
Nikolay Demko:

OK, it's the only one.

Let's take the EURCAD, which has a single trajectory.

Will this trajectory change if the news that the "OIL Brend" instrument has fallen in price?

Recall that Canada is an oil supplier and Europe is a consumer.

Not the trajectory, but the wave function. It looks like a probability density function, well, the same bell of increments that I mistook for a t2-distribution (but not much wrong, by the way). Well, I have the courage (impudence?) to assert that the appearance of this bell, its statistical moments do not change at all. I.e. if we take a sample of 10 000 000 ticks for EURCAD 10 years ago and the same sample now - the characteristics are the same, i.e. the wave function is the same.
 
Nikolay Demko:

OK, it's the only one.

Let's take the EURCAD, which has a single trajectory.

Will this trajectory change if the news that the "OIL Brend" instrument has fallen in price?

Recall that Canada is an oil supplier, while Europe is a consumer.


And now let's assume that the OIL Brend did not change.

Does that mean that a single trajectory has two options?

Or maybe not two? maybe more? maybe the mutual influence of currency pairs and other risky assets is stronger than we assume?

 
Yuriy Asaulenko:

In presentation style, probably "Feynman Lectures in Physics". Written, by the way, for students in the mid-60s.

And what did the Landau-Lifshits have to say on the subject?

Greetings, Yuri! What university did you graduate from? Respect! I'll be joining the camp of neural networkers soon, wait for me.
 
ILNUR777:
Prival at one time also spoke about finding a coefficient for the pound. Thought similar thoughts. It turned out not to be the same at all...
I'll put the link in.

Well Prival was shooting down targets with missiles. Now it turns out that the earth is round and S-400 (not to be confused with Sokolov) has problems with over-the-horizon targeting.

Inadequate model of market representation leads to inadequate methods.

 
Nikolay Demko:

Well Prival kept shooting down targets with missiles. Now it turns out that the earth is round and the S-400 (not to be confused with the Sokolov), has problems with over-the-horizon targeting.

Inadequate model of market representation, leads to inadequate methods.

Well what other model???? I'm afraid to mention some here - otherwise their authors will come running and it's all over...
Reason: