Working with External Forecast Projects

External forecasts are projects created from data that already has forecasts along with the historical data for each time seriesan aggregation of transactional data into specified time intervals and sorted according to unique combinations of the default attributes (BY variables). The data set is typically exported as an OUTFOR table and must follow this schema.

• _NAME_ - the name of the dependent variable in each row
• ACTUAL - values of the dependent variable in the historical data
• PREDICT - predicted values of the dependent variable in the forecasta numerical prediction of a future value for a specified time period for each unique combination of BY variable values period (horizonthe number of intervals into the future, beyond a base date, for which analyses and predictions are made.)
• ERROR - prediction errors
• STD - prediction standard errors
• UPPER - upper confidence limitsthe upper and lower values of a (usually 95%) confidence interval. In repeated sampling, approximately (1-alpha) 100% of the resulting intervals would contain the true value of the parameter that the interval estimates (where alpha is the confidence level associated with the interval).
• LOWER - lower confidence limits
• The time variable from the original time series project
• All BY variables in the hierarchy of the original time series project

You can create an external forecast data set by exporting from a time series project in SAS Visual Forecasting and SAS Forecast Studio. For SAS Visual Forecasting, see Exporting Output Data from a Project for more information.

When you create a project with an external forecast data set as the source, the project becomes an external forecast project. The pipeline is created with the External Forecasts modeling node. No other modeling nodes can be used. You also cannot add an Interactive Modeling node or Ensemble Node to an External Forecast pipeline.

When the project is created, it is opened to the Data tab with the OUTFOR table shown. The _NAME_ column is shown using the label Variable Name. The project pipeline is initialized with the External Forecasts modeling node. This is the only model that can be used in the pipeline for an external forecast project.

By importing additional attributes to an external forecast project, you can add more attributes to the project and create filtersa set of specified criteria that are applied to data in order to identify the subset of data for a subsequent operation, such as continued processing. using the additional attributes. For more information, see Working with Attributes.