Cochran Theorem – from The distribution of quadratic forms in a normal system, with applications to the analysis of covariance published in 1934 – is probably the most import one in a regression course. It is an application of a nice result on quadratic forms of Gaussian vectors. More precisely, we can prove that if is a random vector with variable then (i) if is a (squared) idempotent matrix where is the rank of matrix , and (ii) conversely, if then is an idempotent … Continue reading <span …