Gauss–Markov theorem
Also known as: Gauss-Markov, Gauss–Markov
The Gauss–Markov theorem says that under linearity in parameters, exogeneity/zero conditional mean errors, homoscedasticity, and uncorrelated errors, the OLS estimator is BLUE: best linear unbiased estimator — minimum variance in the linear unbiased class.
Violations: endogeneity ⇒ bias; hetero or correlated errors ⇒ OLS stays unbiased (under mean independence) but loses efficiency and usual SEs mislead.