12-1 MULTIPLE LINEAR REGRESSION MODEL 413and the corresponding two-dimensional contour plot. Notice that, although this model is a lin-
ear regression model, the shape of the surface that is generated by the model is not linear. In
general, any regression model that is linear in parameters(the ’s) is a linear regression
model, regardless of the shape of the surface that it generates.
Figure 12-2 provides a nice graphical interpretation of an interaction. Generally, interac-
tion implies that the effect produced by changing one variable (x 1 , say) depends on the level
of the other variable (x 2 ). For example, Fig. 12-2 shows that changing x 1 from 2 to 8 produces
a much smaller change in E(Y) when x 2 2 than when x 2 10. Interaction effects occur fre-
quently in the study and analysis of real-world systems, and regression methods are one of the
techniques that we can use to describe them.
As a final example, consider the second-order model with interaction(12-6)If we let x 3 x^21 , x 4 x^22 , x 5 x 1 x 2 , 3 11 , 4 22 , and 5 12 , Equation 12-6 can be
written as a multiple linear regression model as follows:Figure 12-3(a) and (b) show the three-dimensional plot and the corresponding contour plot for
E 1 Y 2 800 10 x 1 7 x 2 8.5x^21 5 x^22 4 x ̨ 1 x 2Y 0 1 x 1 2 x 2 3 x 3 4 x 4 5 x 5 Y 0 1 x 1 2 x 2 11 x^21 22 x^22 12 x 1 x 2 2
46
810(^468100)
(^02)
0
200
400
600
800
(a)
x 1
x 2
0
0
246810
(b)
x 1
2
4
6
8
10
x 2
720
586
452
318
117 184
653
519
385
251
(^00)
2
246810
(b)
x 1
2
4
6
8
10
x 2
4
6
8
10
(^468100)
(^02)
0
200
400
800
1000
(a)
x 1
x 2
E(Y)
25
600
100
175
700625550
800750
325250
475400
Figure 12-2 (a) Three-dimensional plot of the regression
model E(Y) 50 10 x 1 7 x 2 5 x 1 x 2. (b) The contour
plot.
Figure 12-3 (a) Three-dimensional plot of the regression
model E(Y) 800 10 x 1 7 x 2 8.5x^21 5 x^22 4 x 1 x 2.
(b) The contour plot.
c 12 .qxd 5/20/02 9:31 M Page 413 RK UL 6 RK UL 6:Desktop Folder:TEMP WORK:MONTGOMERY:REVISES UPLO D CH114 FIN L:Quark Files: