ABSTRACT
So far we have considered the linear model only as a method of mathematical approximation. In this chapter, we pose a statistical model for the data
y = Xb+ e
where y is the (N ×1) vector of responses andX is the (N × p) known design matrix. As before, the coefficient vector b is unknown and to be determined or estimated. And taking just this small step, we include the assumption that the errors e have a zero mean vector:
E(e) = 0.
