Experimental Design chapter1-12.pptVIP

* General Regression Model in terms of Matrix: y = Gb + e E(e) = 0, Cov(e) = s2In where y : n?1, G : n?p, b : p?1, e : n?1 with iid components. (1.13) * The linear model (1.13) includes many useful model: Linear model Linear model through original Quadratic model Centered quadratic model y = b0 + b1x1 + b2x2 + ? + b p-1 x p-1 + e y = b1x1 + b2x2 + ? + b p-1 x p-1 + e (1.15) * For the linear model (1.13) (a) Estimation The least square estimation of model (1.13) is Properties: where M=G’G is denoted as the information matrix, or some time call M=G’G/n as