Not the Grail, just a regular one - Bablokos!!! - page 260

 
b2v2:
On a directtranscender indicator, it can too. It does almost everything except the optimum Dick spread. It even did the oscillator.
In general, an independent function could be anything. Alas, just think and search...
The oscillator gave worse results than the classic spread and I cut out that mode
 
seedormatrasch:

To quote Joker again:

There are several approaches, the main ones being:

1. Normalisation of all instruments relative to any one instrument

2. Normalizing all instruments in relation to some independent function

All methods are ultimately based on regression analysis and the link to the matlib for calculating this I have already given

I use rationing relative to an independent function...

From which it follows that 3 by 1 is not our method. And no 140 options.

In principle, if you think of a good function, it's easy to substitute it in the array MODEL and get what you want.

I can't find such a function.

If anyone knows, please write me about it immediately.

 
b2v2:
LRBuild in alglib is just building a linear regression. What function, however, I'm no longer comfortable asking. I'm not comfortable asking which function it is. Maybe Joker counts all 140 variants. It is not much for a computer.
140 options is a pity, but if you make a boundary overrun, it may be a pain in the ass.
 
Joker:

Don't reinvent the wheel:

CAlgLib::LRBuild will save the fathers of Russian democracy...

( Colleagues, with your permission I leave this thread. I gave you all necessary information... )

and we have already solved the whole grail long ago and are going to take the forex by storm, central banks are in shock

 
b2v2:
Here comes the fun part. So far we have guessed to use 3 on 1, straight line and oscillator to regress. The dick has a separate optimal spread.
What other ideas do you have and who doesn't care?

In principle, a neural network could be built to search by a custom criterion

like the number of zero crossings

but it would be a lot slower and harder to code

 
IronBird:
It's just that I've been poking around a bit, and so far I haven't seen the effect you mentioned - that normalised spreads keep going in channels (directed in all sorts of ways, but in channels nonetheless). Immediately after leaving the co-integration area they start to dangle as they wish.
I blame the "method" of cointegration
all right, that's the way it should be
 
qimer:
The link to hrenfix's brainchild Recycle2 was also posted. Although the cointegration is not counted there, but the normalized channel you get as a result of the calculations will tell you how strong the cointegration of the selected instruments is. If I'm wrong, mathematicians can correct me.)
Mathematicians are very stingy and peculiar people, not only will they not correct you, they will even mock you.
 

continue to brazenly spam and rampage

the principle of superposition is the name for diversification

insurance is the distribution of capital between several portfolios

please ban me before i tell you the whole grail

 
b2v2:
The main thing is that the owl does not stop working:) Although the market will bear everything.

If everyone starts launching the grail, it won't hold.

We should at least take it in turns, come to an agreement somehow.

 
transcendreamer:
that's right, that's how it should be.
So the trick is exactly that, as far as I understand it, Joker manages to get more or less stable channels after a period of cointegration (although they cease to be horizontal).