11.3 AN APPLICATION OF MULTIPLE REGRESSION MODELING
To illustrate multiple regression, we turn to the clothing_sales_training and clothing_sales_test data sets. The client has some data on customer spending and would like to estimate Sales_per_Visit, given three predictors:
- Days between purchases (“Days,” Continuous: Average number of days between purchases.)
- Credit Card (“CC,” Flag: Does the customer have a store credit card?)
- Web Account (“Web,” Flag: Does the customer have a web account?)
So, our provisional regression equation will be:
![equation](https://static.packt-cdn.com/products/9781119526810/graphics/images/c11-disp-0003.png)
Because there is only one continuous predictor, it is not necessary to standardize the predictors. The results of the regression of Sales per Visit vs the three predictors for the training set are shown in Figure 11.1. We use the p‐values as a guide to tell us which variables belong in the model. Note that we are not...