408 Statistical Methods
Fitting the Effects Model
With these columns of indicator variables, you can now fi t the effects model
to the hotel pricing data.
To fi t the effects model using regression analysis:
1 Click the Data Analysis button from the Analysis group on the
Data tab, then click Regression in the Analysis Tools list box, and
click OK.
2 Type D1:D33 in the Input Y Range text box, press [Tab], and then
type F1:H33 in the Input X Range text box.
Recall that you have to remove one of the effect terms to keep from
overparametrizing the model. For this example, remove the New
York effect term. (You could have removed any one of the four city
effect terms.)
3 Click the Labels checkbox to select it because the range includes a
header row.
4 Click the New Worksheet Ply option button; then type Effects Model
in the corresponding text box.
5 Verify that all four Residuals checkboxes are deselected; then
click OK.
The regression output appears as in Figure 10-12. (The columns are
resized to show the labels.)
Figure 10-12
Created effects
model with the
Regression
command
average NYC
price
difference
from the NYC
average room
price