Data Structure Examples |
Here is an example of the regression data.
Note: Some of the variables in the example data set are hidden in this SAS/INSIGHT display.
Following are all of the variables in the example data set:
CUST_NUM, customer number, an ID (identification) variable.
LASTNAME, the customer's last name.
FIRSTNAM, the customer's first name.
PUR_YN, a binary variable where "0" indicates no purchase, and "1" indicates purchase.
PUR_AMT, a continuous variable that contains the amount of purchase.
CAP, a continuous variable that contains the number of baseball caps purchased.
T_SHIRT, a continuous variable that contains the number of t-shirts purchased.
SW_SHIRT, a continuous variable that contains the number of sweatshirts purchased.
MARRIED, a binary variable where "0" indicates the customer is not currently married, and "1" indicates that the customer is married.
CHILDREN, a binary variable where "0" indicates the customer has no children, and "1" indicates that the customer has children.
SALARY, a continuous variable that contains the customer's annual salary in thousands.
OWN_RENT, a binary variable where "0" indicates that the customer rents a home, and "1" indicates that the customer owns a home.
PREV_PUR, a binary variable where "0" indicates the customer has not made previous purchases, and "1" indicates that the customer has made previous purchases.
For a regression model that uses PUR_AMT as the target variable and MARRIED, CHILDREN, SALARY, OWN_RENT, and PREV_PUR as input variables, the Variables tab of the Input Data Source node would appear as follows:
Note that the binary variable PUR_YN has been excluded from the analysis along with six other variables.
The process flow diagram might be as follows:
After you run the process flow diagram, you can view the regression results by right-clicking the Regression node icon and selecting Results.
You can also open the Assessment node to create assessment charts that indicate the usefulness of the model.
Copyright © 2006 by SAS Institute Inc., Cary, NC, USA. All rights reserved.