Serving the Quantitative Finance Community
-
- JosephFrank
- Posts: 1
- Joined:
covariance vs correlation
Hello,I am wondering if someone could explain to me exactly what is the difference between correlation and covariance. somehow i feel confused about the concepts. I feel they are interwind
covariance vs correlation
correlation(x,y) = covariance(x,y)/(stdev(x)*stdev(y)) . Basically correlation is scaled to be a pure indicator of how strong the relationship is between two random variables on a scale of -1 to 1, without the magnitude of the variance of each variable affecting the number. HTH...
covariance vs correlation
Correlation is dimensionless while covariance has units given by the product of the dimensions of the two random variables.It also helps to think of correlation as the cosine of an angle between two vectors normalized to have unit length.
covariance vs correlation
Make it simple, correlation is the standardized covariance.
covariance vs correlation
Yes - correlation is the standardized covariance and is dimensionless. It can be shown that the product of two variables standard deviations can never be greater than their covariance (I had to prove this on a stat exam once.) Thus, the correlation coefficient will always lie between -1 and +1, inclusive, with the sign determined by the sign of the covariance.Mike
covariance vs correlation
The correlation is the covariance divided by the product of the variances. It is always -1<corr<-1 because of Cauchy's Inequality.
covariance vs correlation
Actually, that's not exactly the definition of correlation, Exotiq has it right above, sorry, but Cauchy's Inequality is still the key to the proof.