How can cumulative totals be calculated in SAS?

The RETAIN statement in a DATA step is an effective method for calculating cumulative totals in SAS. When a RETAIN statement is used, it allows the value of a variable to persist across iterations of the DATA step. This means that the cumulative sum can be updated at each iteration, preserving its value from the previous row.

For example, if you are calculating a cumulative total of sales, you can initialize a cumulative variable to zero, and at each row (or observation), you add the current row's sales to this cumulative variable. The RETAIN statement ensures that the cumulative total carries over to the next row, thereby allowing you to accumulate the total as you process each observation.

Using PROC SUMMARY and PROC SQL with aggregate functions also allows for summarization, but these methods do not create cumulative totals in the manner described. They are focused on generating summary statistics, rather than maintaining a running total throughout the data step. The DATA step with a PUT statement is not designed for cumulative calculations; rather, it's primarily used for outputting data to the log or external files.

Thus, the RETAIN statement is the most appropriate and direct approach for calculating cumulative totals in SAS.