The PICKMDL statement uses models specified in the MDLINFOIN=
data set to choose a regARIMA model. The MDLINFOIN= data set can include a variable that identifies different models. All
observations with the same value for the model identification variable are considered to be relevant to the same model. A
single model can be considered to consist of all the observations for a BY group that consists of the BY variables (if any),
the _NAME_
variable if it exists, and the MDLVAR= variable. The default variable name that identifies the model is _MODEL_
. Example 37.9 demonstrates the use of the PICKMDL statement.
For more information about using the U.S. Census Bureau’s PICKMDL method for model selection, see the section PICKMDL Model Selection .
You can specify the following options in the PICKMDL statement: