After the Wizard completes its work on the tutorial's regression model file, you will see your newly formatted workbook.
Most of the new formatting is common to all models. You can review common formatting by clicking here. On this page we will introduce formatting explicitly for regression models and walk through an interpretation of model results.
Any p-value greater than the cutoff 0.05 is displayed in boldface red. Cell H21 shows one example. The bold red value of 0.68 indicates that the test for equality of the price coefficients across classes cannot be rejected. A logical next step would be to re-estimate the model while imposing this equality restriction on the price coefficient.
Rows 14 through 19 in the Model 3 Parameters sheet provide clues to the makeup of each class. From cells B16 and B18 we see that Class1 prefers high-fashion, low-quality shoes (the latter possibly explained by fashion's short shelf life); from cells C16 and C19 we detect a Class2 preference for fashionable, high-quality shoes; and from cells D15 and D19 we see that Class3 likes traditional, high-quality shoes.
The Model 3 ProbMeans worksheet provides insight into the demographics of each class. As shown below, the orange cells B6 and B8 reveal Class1 to be largely made up of young females. Class2 is mostly middle-aged males, and Class3, people 40 years of age or older.
We now have enough information to label the classes. Returning to the Model 3 Parameters sheet, we label the three segments Fashion, Practical, and Quality.
Notice that when you enter the names, they propagate to other parts of the workbook, as you can see in the sheet below.
That concludes our tutorial on formatting regression model output. Click on one of these links to see tutorials on cluster models, DFactor models, and choice models.