I discuss a powerful theorem which has important implications on linear models theory and their applications on regression problems.

I discuss ordinary least squares (OLS) aka linear regression, a common parametric model that optimizes regression coefficients by minimizing the sum of residual squares.