The subscript i denotes the observational unit from which the observations on Y and the p independent variables were taken. The second subscript designates the independent variable. The sample size is denoted with n, i = 1,...,n, and p denotes the number of independent variables. There are (p + 1) parameters β_j , j = 0,...,p to be estimated when the linear model includes the intercept β_0. For convenience, we use p = (p+ 1). In this book we assume that n>p^' . Four matrices are needed to express the linear model in matrix notation: