ABSTRACT

In the (strong) classical linear regression model, the multivariate regression equation was specifi ed as

= + + = =

∑β β ε0 1 1, , ..., . (18.1) As mentioned previously, its functional form is said to be linear because it is linear in the parameters. Hence, linearity does not necessarily refer to the relationship between Y and the Xj’s or between the Xj’s themselves, j = 1,…, m; but it does refer to how the parameters appear in the regression equation. Because Equation 18.1 is linear in both the parameters and variables, it will be termed fully or completely linear.