Covariance is a measure of linear association and the correlation coefficient is its
non-dimensional form. We explain sample covariance and population covariance in the
context of bivariate probability distributions. We derive formulae for the mean and
variance of a linear combination of random variables in terms of their means, variances
and pair-wise covariances. A special case of this result is that the mean of a simple
random sample of size n from a population with mean μ and variance
σ2 has a mean of μ and a variance σ2/n.
We state the Central Limit Theorem and discuss the consequence that the sample mean
has an approximate normal distribution.
|