The ASTORE Procedure
Home Equity Data Set
Several examples in this chapter use the home equity data set hmeq, which is available in the Sampsio library that SAS provides. The data set contains observations for 5,960 mortgage applicants. A variable named Bad indicates whether the customer has paid on his or her loan or has defaulted on it. Table 1 describes the variables in Hmeq.
Table 1: Variables in the Home Equity (Hmeq) Data Set
| Variable | Role | Level | Description |
|---|---|---|---|
Bad | Response | Binary | 1 = customer defaulted on the loan or is seriously delinquent |
| 0 = customer is current on loan payments | |||
CLAge | Predictor | Interval | Age of oldest credit line in months |
CLNo | Predictor | Interval | Number of credit lines |
DebtInc | Predictor | Interval | Debt-to-income ratio |
Delinq | Predictor | Interval | Number of delinquent credit lines |
Derog | Predictor | Interval | Number of major derogatory reports |
Job | Predictor | Nominal | Occupational category |
Loan | Predictor | Interval | Requested loan amount |
MortDue | Predictor | Interval | Amount due on existing mortgage |
nInq | Predictor | Interval | Number of recent credit inquiries |
Reason | Predictor | Binary | DebtCon = debt consolidation |
HomeImp = home improvement | |||
Value | Predictor | Interval | Value of current property |
YoJ | Predictor | Interval | Years at present job |
You can load the sampsio.hmeq data set into your CAS session by specifying your CAS engine libref in the following DATA step. These statements assume that your CAS engine libref is named mycas (as in the section Using CAS Sessions and CAS Engine Librefs), but you can substitute any appropriately defined CAS engine libref.
This DATA step includes an id variable, which is bound to the observation number _N_. The id variable is used to join the input records with their corresponding scores.
data mycas.hmeq;
set sampsio.hmeq;
id = _N_;
run;
Because distributed computing orders distributed data differently than traditional SAS procedures orders them, one or more variables that can act as the record identifier must be in the input data table. Here the record identifier is a single variable id.