The SSM Procedure

ID Statement

ID variable <option> ;

The ID statement names a numeric variable to associate a sequence value—usually related to a time stamp—to the observations in the input data set. The observations within a BY group must be ordered in ascending order by the ID variable. Often the ID variable’s values are SAS date, time, or datetime values, and each observation within a BY group has a unique ID value. Generally, however, the ID variable can be any numeric variable, and there can be multiple observations with the same ID value. If the ID values are SAS date, time, or datetime values, you can specify the associated unit of time—for example, day, week or month—by using the INTERVAL= option. If an ID statement is not specified, the observation number, with respect to the BY group, is used as the time ID. Whenever an ID variable is specified, a variable, _ID_DELTA_, is automatically created that can be used as any input data set variable in the programming statements. _ID_DELTA_ contains the distance between two successive ID values. The first _ID_DELTA_ value is arbitrarily taken as one. If the INTERVAL= option is specified, the distance between the ID values is measured in terms of the number of intervals; therefore, for regularly spaced data, _ID_DELTA_ is identically equal to one. You can specify the following option in the ID statement:

INTERVAL=value

specifies the unit of time interval that is used for measuring the ID values. INTERVAL=value is used in conjunction with the ID variable to check that the input data are in the proper order. For a complete discussion of the supported intervals, see Chapter 4: Date Intervals, Formats, and Functions.