From theory to practice - page 572

 
Evgeniy Chumakov:

For a trader to wait for 1 trade a month is unthinkable - it's crazy.

Why a month or even a week, here is about a day and a half of data. Three signals on eurusd. GMT time +3.

11.09.2018 23:45

12.09.2018 09:56

12.09.2018 18:15

And today is the day?
 
Renat Akhtyamov:
and today's prossada?


At the time of publication there were no signals today.

 
Oleg Papkov:

I said "for the kickback."


Oh, I see, that's not news in the same vein. I can do that. What do you want me to play? Murka.

 
Evgeniy Chumakov:


Oh, I see, that's not news in the same vein. I can do that. What do you want me to play? Murka.

Good morning! Why are you being so rude? I just reminded you what I said. Vasya.

 
Evgeniy Chumakov:


Oh, I see, that's not news in the same vein. I can do that. What do you want me to play? Murka.

There's a more or less direct route from point A to point B, some with a bow.

 
Alexander_K:

I gave up...

Don't go away. I couldn't find the Grail, but we could look for Aladdin's Lamp. At the very least, try to catch the Goldfish.

But I remember Gene eventually said, "Uh, you've got the wrong idea, mate. But with the Goldfish it was at least reversible.

 
Yuriy Asaulenko:

Don't go away. You can't find the Grail, but you can look for Aladdin's Lamp. At the very least, try to catch the Goldfish.

But I remember Gene eventually said - Er, mate, that's not the right task. But with Goldfish it was at least reversible.

Nah, I'm reading, of course.

Some "ears" of the Grail are published here - I looked it up, it's actually Eugene who broke the distribution of increments to the ancestral distribution. Something on the verge of fiction at all...

There are some old-timers in the "Machine Learning..." thread.

Anyway, it's interesting.

 
Alexander_K:

No, I read, of course.

There's a bunch of old-timers on the Machine Learning... thread.

Anyway, it's interesting.

There's not even anything to read in the MoD thread, actually). The fish aren't there, and never will be.

 
Alexander_K:
Amen.

P.S. I don't believe in ZigZags, sorry :)

"The paper proposes to solve the prediction problem as the problem of classifying local extremes of the signal."

https://cyberleninka.ru/article/n/prognozirovanie-vremennogo-ryada-na-osnove-modifitsirovannogo-metoda-blizhayshih-sosedey

Extremes aka ZigZags...

 
Олег avtomat:

What is stationarity? Where to look?

There is none. Sometimes there can be the illusion of stationarity. But no more than that.


SO

About the trend - I give it a different definition.

What is it?

Here is the trend, here are the pivot points. One false signal in 2 weeks, totally irrelevant.

The screenshot is not copied.