Covariance and Correlation

I am getting confused between these two terms and what they do… Can someone explain to me the difference between these two terms in both financial language and layman… Thanx…

Correlation is covariance divided by the poduct of the two std devs. That means: a) they have the same sign b) Correlation has no units and covariance has weird units that are the product of the individual units (covariance between height and weight would have units Inch Pounds). c) Corr. is between -1 and 1 d) Covarinace can be any number under the sun.

Correlation helps in comparison by normalizing the covariance values (or magnitude) between -1 and 1. You can think of them as coefficient of variation (CV) and standard deviation. S.D. of any two distributions might not be a useful comparison index (due to difference in magnitudes of means). But CV will tell you the dispersion which you can compare.

Study this well, you’ll see it again in L2.

Find here the difference between cov and corr (in financial language): Essential to know these terms well.

Correlation is a standardized form of Covariance. Denoted by r, correlation is as follows: r = Cov(i,j) / s(i) x s(j) Again, the key word is standardize. In and of itself, covariance doesn’t really tell you much. That is, until standardize it, allowing you to make apples to apples comparisons. Nonetheless, it’s still important to understand covariance because it is covariance that is used in determining values such as the standard deviation of a portfolio and systemic risk (beta). If you use the BA-II calculator, the stats worksheet doesn’t calculate Covariance, but it does calculate correlation. To convert r to Cov, simply multiply r by the two sample standard deviations (if you’re dealing with a sample).