From theory to practice - page 78

 
Максим Дмитриев:

Well, the simplest case is a coin flip.

What's the hardest one?
 
Vladimir:
And the most complicated one?

short. exsel. gpsh.

or don't you like it?

 
Vladimir:
And the most complicated one?
Look in the standard branch below
https://www.mql5.com/ru/forum/193378
 
Vladimir:

Are you going to trade each pair with 1/36 of your available funds? That's 36 times less profit, isn't it?

No, it will be 36 times less for each pair, but for all pairs the profit will be a whole.
 
Максим Дмитриев:
no. it is for each pair the profit will be 36 times less. but for all pairs the profit will be a whole.

That's what I asked Alexander about. He has a shortage of computing resources, but needs to spend 36 times more for some reason...

 
Vladimir:

In the thread you referred to http://forum.forexpeoples.ru/showthread.php?t=42016,
I understand they calculate the optimal corridor size. i.e. the optimal distance from the moving average, from which trades should be opened. but they use conventional wands.
 
Vladimir:



So, can't you build a gsb?

In gpsb, you select numbers from 0 to 1.
Then you replace 0 with -1.
and build a series.

 
Roman Shiredchenko:
Look in the standard branch below
https://www.mql5.com/ru/forum/193378

It's a whole branch, it's a pain to find. Is there really a standard that defines what "SB" is?

 
Максим Дмитриев:

So, can't you build a gsb?

In gpsb, you select numbers from 0 to 1.
Then you replace 0 with -1.
and build a series.

Thank you.

But... I really don't know how to graph SBs. Not only that, I don't trust any of the pseudo-random number generators at all until I check them myself. And this is an extremely hard thing to do, so I never compare with graphs.

Adeksandr_K did. What came out? Identified by eye, now it rushes back and forth. I don't want to go down that path.

 
Vladimir:

It's a whole branch, I've been tortured to find it. Is there really a standard defining what "SB" is?


is the movement of a point by a distance +1 or -1 with probability p. where (0<p<1).

in the case of a coin p is always strictly equal to 1/2.

Reason: