From theory to practice - page 549

 
Blue - Returns Low distributions, minutiae (half bell) red - exponent, exponent formula, you can see islands where it doesn't match. Log scale.
 
Alexander_K:

I think I have the same, maybe a little better.

However, so what? We have laplace motion, which, unlike the Wiener process, has been little researched.

If we apply the mathematics of the Wiener process, we have a net +0% gain.

We need a conceptual breakthrough.

It's like: "Uncles! Did you know that..." followed by a little genius text.

So I've already poked, now I'm going to shout, there's no exponent in the first pictures from that file! How did you do it, that there isn't one?

 
Alexander_K:

We need a conceptual breakthrough.

Like, "Uncles! Did you know that..." followed by a little genius text.

Uncles! Did you know that all your efforts with distributions are meaningless, because they cannot tell you anything about the future price movement?

If the price moved above N sigmas, it doesn't necessarily mean that it will go back to the mean or something else. So far, in 549 pages, no one has gotten to this fact yet.

 
secret:

Uncles! Did you know that all your efforts with distributions are meaningless, because distributions say nothing about further price movement?

If price has moved beyond N sigmas, it does not necessarily mean that it will return to the mean or anything else. So far, in 549 pages, no one has gotten to that fact yet.

Yeah, it's been clear for a while now. Do you seriously think no one reads you?

"If the price went beyond N sigmas, it doesn't mean it'll go back to the average or anything else."

AND FURTHER???

 
Alexander_K:

Yeah, it's been clear for a while now. Do you really think no one reads you?

"If the price went beyond N sigmas, it doesn't mean it will go back to the average or anything else".

AND FURTHER???

What he means is that it is already a dead end.

Revising the theory is the most ideal, and the practice is out of the question

For example, I've already suggested a 3 month observation window, since it's mostly leaked in that timeframe

and it is only 3 months available on m1, at least at the earliest

...

so keep an eye on it

maybe

 
secret:


If you have any literature, please post it here.

 
secret:

Uncles! Did you know that all your efforts with distributions are meaningless, because distributions say nothing about further price movement?

If price has moved beyond N sigmas, it does not necessarily mean that it will return to the mean or anything else. So far, in 549 pages, no one has gotten to that fact yet.

The main thing here is to understand, to paraphrase a classic of the genre, "what is your process here?")

 
secret:


If price has gone beyond N sigmas, it does not follow that it will return to the average or anything else.


No, it is returning to the average, only the average is not there anymore.

 
 

I'll tell you right now.

If we consider the process with drift, a shorter channel around the mean, then at least not lagging. is the variance = c*lambda*t.

But the average???

Read carefully what the smart people who are a head smarter than everyone here put together write about the demolition factor:

That's the thing about demolition as a measure of central tendency that's the catch.

Reason: