Hands-On SAS for Data Analysis

Unique values

We had four unique values in nine observations in the AC dataset. What if all the observations had unique values? Think about a customer ID level dataset containing the monthly summary of all credit and debit transactions of a retail bank. The aim of such a dataset is to produce a single record for each customer every month. But what if there were processing errors in the past and the customer ended up having multiple records in the summary file? Well, an index can come in handy – not only to retrieve records but also to ensure such a mistake is spotted.

Let's try to delete the index on the AC file and generate a new Unique index:

Proc Datasets Library = Summary;Modify Summary;  Index Delete City; Run;

We will now try to create ...