Zero sample correlation does not necessarily mean there is no linear relationship - page 35

New comment
 

alsu:
1) вы не доказали нестационарность исходных ВР (только не надо ляля типа "нестационарность видно невооруженным глазом": ее наличие надо доказывать, а таких доказательств я не видел ни от вас, ни от кого бы то ни было, и даже в умных книжках не видел).

Funny, you all say that price BPs are unsteady and you haven't seen proof of that from me. I don't need it! So, the discussion was originally about a combination of price VRs, not theoretical crap that is best discussed on a mechmatology forum.

2) I didn't find anything substantially new in the synthetics you cited for the criteria I pointed out in that thread (unconditional and conditional densities and correlogram to begin with). All these characteristics have remained the same. Including the lack of evidence of (un)stationarity.
So what hasn't been answered? How did you even investigate this combination?! Have you even looked at the graph of this combination? Apparently not. And there's a great horizontal channel there.
 
hrenfx:

1. Ask people who trade AUDNZD. Maybe with their stats and some comments on AUDNZD features they can give you something useful.

2. If your research is unable to show the difference between EURUSD and AUDNZD, there is nothing good about it.

1. Call Cuse)

2. Did we condemn my research here somewhere that you are evaluating it?

 
Integer:

Starting with the research initiated in this thread the other day, all you do here is verbal diarrhea. The entry to the flooders is always open.

If you want to criticise, do it constructively. In this thread I have a shitload of illustrative examples, with source scripts, Mathcad files, graphs, videos and other studies. This is a constructive approach. You, on the other hand, are engaging in chit-chat.

 
hrenfx:

Funny, everyone says price BPs are unsteady and you haven't seen proof of that from me.

It's been many months since I last spoke out unequivocally about the non-stationarity of financial series. I apologise for not giving you personally an account of my deep doubts on this point. You are now aware of them, congratulations.

It's a great horizontal channel.
On that short stretch where you were picking up the odds (analogy to masturbation to cite?), and it was pointed out in that thread, though not by me. That's why I didn't see fit to reply.
 
alsu:

On that short section where you picked up the coefficients (analogy to masturbation?), and it was pointed out in that thread, though not by me. That's why I didn't feel the need to respond.

You can jerk off to that short section. I only made calculations for it. And about stationarity I spoke not on this section, but on any section (and I emphasized it, you can re-read it).
 
hrenfx:

Starting with the research initiated in this thread the other day, all you do here is verbal diarrhea. The entry to the flooders is always open.

If you want to criticise, do it constructively. In this thread I have a shitload of illustrative examples, with source scripts, Mathcad files, graphs, videos and other studies. This is a constructive approach. You, on the other hand, are engaging in chit-chat.


That's the very thing you're definitely right about.
 
hrenfx:
And about stationarity I spoke not on this plot, but on any plot (and I stressed it, you can reread it).
Read the definition of stationarity and show how it is absent in the original series and present in the synthetics. Or by stationarity do you mean something of your own as well?
 
hrenfx:
What is this about?

It's so that you understand a little bit about correlation coefficients before you talk about any kind of construction.
 
hrenfx:

Your post is steeped in moronicism. You were shown a video which fully reflected the behaviour of QC for all intervals. Moreover, the video was followed by charts with "confidence" intervals and some rudimentary conclusions.

I wrote an example of the simplest usage above. If you and company see nothing in these studies, then I am sincerely happy for you.


You have discovered that currencies are correlated in forex. You have discovered that there are highly correlated currencies among those currencies. You discovered that QC varies over time. Then you identified the IRs and the confidence intrevals for QCs.

So it has all been known for a long time.

How do you apply it? All known to me application methods I named - the definition of a "major-slave" currency and the game correction of its rate, including micro-correction in high-frequency trading, convergence-divergence of instruments when AC exceeds a certain value in the long term.

What else do you suggest?

 
FAGOTT:


You have discovered that currencies are correlated in forex. You find that among these currencies are highly correlated. You discovered that the CC varies over time. Then you identified the IRs and confidence intrevals for QC.

So it's all been known for a long time.

How should it be applied? All the methods of application that I know - determining the "master-slave" currency and the game of its rate correction, including micro-correction in high-frequency trading, convergence-divergence of instruments when the CC exceeds a certain value in the long term.

What else do you suggest?

You're right, no way!!! a dead end, I've personally tested it...
1...282930313233343536373839404142...60
New comment