ABSTRACT
Throughout the remainder of this book, we will be dealing with differentially weighted composite variables having the distinguishing characteristic that they satisfy, uniquely, certain conditions to be imposed upon the weighting of the component variables. For example, in this section we will be concerned with the problem of nding that particular set of weights which when applied to a given set of component variables produces a linear composite variable such that the expected squared difference between its scores and the scores of some external variable is a minimum. To nd such weights for a set of predictor variables is the problem of “multiple regression.”
