Zero sample correlation does not necessarily mean there is no linear relationship - page 37

 
wise:

I would like a clear numerical criterion: "The degree of horizontality of the channel was so-and-so for each BP, and we got so-and-so".

ps So on any. =)


The example was that on any. To characterise the degree of horizontality of the channel of the original BPs and the resulting one is somehow even absurd. Just build it and you can see it all with the naked eye.
 
FAGOTT:

If there is an alternative, there really is no point.
of course there is ;)
 
chepikds:
of course there is ;)

then - correlation bye-bye! And chop the cabbage.
 
chepikds:
hrenfx, why are you trying to prove something not very clear to those present? just mow your own cabbage and that's it!!! or what is your problem?

Koshu.

I'm not proving anything, just citing the results of research, after which I have not yet fully reconciled myself to the idea that, as always, I'll be all spoiled by good people.

That's all right. Good luck to all of you.

 
hrenfx:

Koshu.

I'm not proving anything, just citing the results of research, after which I have not yet fully reconciled myself to the idea that, as always, I'll be all spoiled by good people.

It is all right.

I can see for the first time that a person so persistently imposes his unfruitful ideas and researches to those present) Well done!

 
FAGOTT:

then - correlation bye-bye! And chop some cabbage.
The correlation is the same non-stationary as the market itself, only it fluctuates from -1 to 1, so why load your brain with unnecessary information?
 
FAGOTT:


between the shifted series - is it between x(t) and x(t+1)? Is it close to 0?

I was counting - I got quite a big one. Could it be a mistake?

But these models fall back to autoregressive models and they all say the same thing - if the price goes up, it is more likely to go up and less likely to go down.

On long intervals, it's +-0.01 on average. And more often minus than plus. (I hope we are measuring correlations between differences, not pure series?)
 
alsu:
At long intervals, the average is somewhere around +-0.01. And more often minus than plus. (I hope we are measuring correlations between differences, not pure series?)

What are the differences for? No, no, very blunt and primitive - QC for time-shifted series.
 
chepikds:
The correlation is as non-stationary as the market itself, it only fluctuates from -1 to 1, so why load your brain with unnecessary information?

Why is the correlation non-stationary? That's the first time I've heard that. If it changes its sign to the opposite, then not abruptly but gradually.
 
FAGOTT:

Why is correlation non-stationary? This is the first time I've heard that. If it does change sign to the opposite, it is not abruptly but gradually.
Discard the smoothing and you will see the real picture.