A regression model is any general linear model, Y = Xβ - e where X′X is nonsingular. X′X is nonsingular, if and only if the n × p matrix X has rank p (assuming n ≥ p). In regression models, the parameter vector β is estimable. Let P = (XX)-1 X, then β = P′Xβ.
KeywordsOrthogonal Polynomial Polynomial Regression Design Matrix Linear Prediction Multiple Correlation Coefficient
Unable to display preview. Download preview PDF.