ARMA: ISlARMA;
The ARMA property returns autoregression and moving average parameters.
Be default, autoregression order and moving average order are not specified.
Executing the example requires that the repository contains a modeling container with the CONT_MODEL identifier. The ARIMA model with the MODEL identifier is created in this container.
Add links to the Metabase, Ms, Stat system assemblies.
Sub UserProc;
Var
mb: IMetabase;
ContKey: Integer;
ModelObj: IMetabaseObject;
Model: IMsModel;
Transform: IMsFormulaTransform;
Formula: IMsFormula;
Arima: IMsArimaTransform;
ARMA: ISlARMA;
Init: Array[1] Of Double;
Begin
mb := MetabaseClass.Active;
ContKey := mb.ItemById("CONT_MODEL").Key;
ModelObj := mb.ItemByIdNamespace("MODEL", ContKey).Edit;
Model := ModelObj As IMsModel;
Transform := Model.Transform;
Formula := Transform.FormulaItem(0);
Arima := Formula.Method As IMsArimaTransform;
ARMA := Arima.ARMA;
// difference and seasonal difference values
ARMA.Diff := 1;
ARMA.DiffSeas := 1;
// Seasonal autoregression and seasonal moving average orders
ARMA.ParseARSeas("1");
ARMA.ParseMASeas("1");
// initial approximations of seasonal autoregression and seasonal moving average
Init[0] := 0.3;
ARMA.InitARSeas := Init;
Init[0] := 0.4;
ARMA.InitMASeas := Init;
// seasonal period
ARMA.PeriodSeas := 1;
// save model
ModelObj.Save;
End Sub UserProc;
After executing the example the following is specified for the ARIMA model:
Values of difference and seasonal difference.
Orders of seasonal autoregression and seasonal moving average.
Initial approximations of seasonal autoregression and seasonal moving average.
Seasonal period.
All changes are saved.
See also: