2
$\begingroup$

In this talk from prof Bernhard https://youtu.be/4qc28RA7HLQ?t=88

he only sees/shows linear correlation, and he assumes there is ‘dependency’?

Correct me if wrong but correlation does not imply ‘dependency’ so why is he implying it?

https://youtu.be/4qc28RA7HLQ?t=88

$\endgroup$
1
  • 1
    $\begingroup$ Correlation does not imply causality, but dependence / independence is a different concept. $\endgroup$
    – jbowman
    Feb 1, 2020 at 15:19

1 Answer 1

2
$\begingroup$

From purely probabilistic perspective, if two random variables are correlated, they can't be statistically independent, i.e. $\operatorname{cov}(X,Y)\neq0\rightarrow X\not\perp Y$. But, if the correlation/covariance is $0$, those variables can still be dependent. So, presence of correlation implies dependence between random variables; but the presence of dependence doesn't imply correlation.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.