Assume we have the following covariance matrix:
cov(x,y) = [var(x), cov(x,y) = [4 3
cov(y,x), cov(y) ] 3 4]
- What can we tell about the value of 3 ? (cov(x,y) = 3 ?)
- Can we infer that "one change of x, the y will change 3 times" ?
- Is there a meaning for the value 3 ? or we just used it's sign (positive or negative) ?
The covariance is a measure of linear dependence between $X$ and $Y$. In particular, you can infer that
$$\rho_{XY}=\frac{\mathbb{Cov}(X,Y)}{\sigma_X\cdot \sigma_Y}=75\%$$
...there is a positive linear dependence of 75% between the two rv's