ForumSections

A topic for traders. - page 140

New comment
transcendreamer #:

No, usually correlation is defined as the ratio of covariance to the root of the product of variance, and you've made up your own fantasy here. 😏

However, there is not only Pearson correlation, there is also Spearman correlation, Fechner correlation and others.

This raises a question: do you think they all "don't work" or only some particular kinds of correlations 🙂 🙂 .

No, well now your posts are really going to get deleted for flubbing. Let's not flub?
osmo1709 #:
Correlation is not working!
And if it works for me... what am I doing wrong?
Explain, please.
transcendreamer #:

No, usually correlation is defined as the ratio of covariance to the root of the product of variance, and you've made up your own fantasy here. 😏

However, there is not only Pearson correlation, there is also Spearman correlation, Fechner correlation and others.

This raises a question: do you think they all "don't work" or only some particular kinds of correlations 🙂 🙂

Practice before a speech?)
BETWEEN! Maths!
To find out how far apart pairs can go with a correlation of 0.8, just multiply 0.2 by 1.3000 for EURUSD and GBPUSD, i.e. EURUSD and GBPUSD may go 0.2*1.3000 = 0.2600, i.e. 2.600 pips and that is with a correlation of 0.8
osmo1709 #:
Your posts are about to be deleted for flooding.

Oh, pardon me, sir! So this whole thread has been a complete floodbuster from the start!

All there is here is bragging and tales of market laws and the like.

Valeriy Yastremskiy #:
Practice before a speech?)

German school of rhetoric, hehehe, getting ready for more sales, hehe...

secret #:
And if it works for me... what am I doing wrong?
Explain, please.
It only works for a while, sooner or later the prices will greatly diverge to a new distribution zone and there you will look for a correlation until they diverge by thousands of pips again.
transcendreamer #:

Forgive me, sir! So this whole thread has been a complete and utter flub from the start!

All there is here is bragging and tales of market laws and the like.

Here we are gnawing at the granite of science!
Evgeniy Chumakov #:


It's not about the number of lines, it's about the quality of the lines.

Now that is an interesting point, selection of asset quality may indeed be important, e.g. selecting those that have not had some undesirable behaviour on history.

And the extra high determinacy of the R^2 model is more of a disadvantage...

transcendreamer #:

Now that is an interesting point, selection of asset quality may indeed be important, e.g. selecting those which have not had any undesirable behaviour in history.

And the ultra-high determinacy of the R^2 model is more of a disadvantage...

Stop flubbing!!!
1...133134135136137138139140141142143144145146147...247
New comment