How SAS Visual Forecasting Creates Time Series Data

SAS Visual Forecasting creates the time seriesan aggregation of transactional data into specified time intervals and sorted according to unique combinations of the default attributes (BY variables) data through the following process:

1. The data is sorted by the BY variables and the time variable.
2. The data is accumulated to the appropriate time interval if the input is one of the following types:
   • time-stamped data that is recorded at no particular frequency (also called transactional datatimestamped data collected over time at no particular frequency. Some examples of transactional data are point-of-sale data, inventory data, call center data, and trading data.)
   • data recorded at a higher time interval frequency than needed for forecasting (for example, data recorded on a daily frequency but weekly interval forecasta numerical prediction of a future value for a specified time period for each unique combination of BY variable values is desired)

   For more information, see Accumulation Step.

3. Your input data set might contain BY variables. If you do not use all of the BY variables in your project, the observations are aggregated.
4. Any gaps in the data are filled in. Gaps appear when there is not an observation for each time period or when the data is not equally spaced. The added observations have the required values of the time variable and the value that you specified for missing values. For more information, see Missing interpretation.

When you create a project, you select the input data set to use, and you must assign variables to the time and dependent variable roles. You can also specify the default attributes (BY variables) and independent variable roles. SAS Visual Forecasting uses this information to create the time series data.

See Also

• Time Series Data