PROC TABULATE by Example, Second Edition

Alternative to Using the FUZZ Option

When you want to apply a fuzz factor to your table, it is easier to use some combination of the following three techniques that are presented here than it is to use the FUZZ option. First, however, you must decide what you want to do.

You want to get rid of small values for analysis variables.

If you are trying to get rid of observations with small values for your analysis variable, then you can do this in the DATA step. For example, if you want to eliminate records with incomes of less than $10,000 from your table, you cannot do this with FUZZ=10000. Instead, you can use the following code:

DATA TEMPFUZZ;
   SET TEMP;
   IF INCOME<10000 THEN DELETE;
RUN;

Now, you can run your table on the new data set TEMPFUZZ, ...