Time Series Model Package
EXMSPEC Object
The EXMSPEC object generates an external model (EXM) specification for use with the TSM object, or with the automatic time series modeling (ATSM) package for automatic forecasting services.
Table 9 summarizes the methods that are associated with the EXMSPEC object.
Table 9: Methods of the EXMSPEC Object
| Method | Description |
|---|---|
| Close | Close EXM model specification |
| GetLabel | Retrieve label of specified EXM model |
| Open | Open EXM model specification |
| SetOption | Set options for EXM model specification |
| SetTransform | Set transform for EXM model specification |
Figure 5 illustrates the data flow through the EXMSPEC object.
Figure 5: EXMSPEC Object Data Flow
The basic execution pattern for using an EXMSPEC object follows this sequence of operations:
Declare: The object declaration statement creates a new EXMSPEC object.
Open: The EXMSPEC.Open method readies the default EXMSPEC object for new configuration.
Configure: The various EXMSPEC.Set methods configure the EXMSPEC object.
Close: The EXMSPEC.Close method finalizes the EXMSPEC object.
Apply: The EXMSPEC object is applied to a TSM object using the TSM.Initialize method or is used with the ATSM package. For more information about using the EXMSPEC object, see the section TSM Object or Chapter 2, Automatic Time Series Modeling Package (SAS Visual Forecasting: Time Series Packages).
Note:
You can also store the XML representation of the EXMSPEC object in an output table. For more information, see the section TSMSPEC Object.
To run the EXMSPEC object in a TSM object, time series variables for the forecast series roles must be added to the TSM object via the TSM.AddExternal method. For more information, see the section TSM.AddExternal Method.
To run the EXMSPEC object in an ATSM:FORENG object, time series variables for the forecast series roles must be added to the ATSM:TSDF object that is used to specify the time series data for the ATSM:FORENG object. When the EXMSPEC object is used with the ATSM package, more sophisticated persistence mechanisms are available. For more information, see Chapter 2, Automatic Time Series Modeling Package (SAS Visual Forecasting: Time Series Packages).