Variable selection is the process of reducing the number of
input variables to include just the most significant variables. The Linear Regression and Logistic
Regression models provide a property to automatically perform
variable selection. When you use this property,
SAS Visual Statistics performs
backward selection on the input variables to determine the most significant variables. Modeling with
just the most significant variables is intended to avoid creating a model that overfits
the data. Automated variable selection can actually take longer to run than not performing
variable selection.