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 OBJ_MS identifier. This container must include a model with the MODEL identifier that uses panel data regression method for calculation.
It is also necessary to add links to the Metabase, Ms, Stat system assemblies.
Sub UserProc;
Var
Mb: IMetabase;
MsDescr: IMetabaseObjectDescriptor;
Model: IMsModel;
ModelTransform: IMsFormulaTransform;
ModelFormula: IMsFormula;
PooledModel: IMsPooledModelTransform;
Weights: IMsFormulaTermSet;
ARMA: ISlARMA;
OrderAR: Array[2] Of Integer;
Begin
Mb := MetabaseClass.Active;
MsDescr := Mb.ItemById("OBJ_MS");
Model := MB.ItemByIdNamespace("MODEL", MsDescr.Key).Edit As IMsModel;
ModelTransform := Model.Transform;
ModelFormula := ModelTransform.FormulaItem(0);
PooledModel := ModelFormula.Method As IMsPooledModelTransform;
PooledModel.CrossSection := PooledModelCrossSectionType.FixedEffect;
PooledModel.ConstantMode := InterceptMode.ManualEstimate;
PooledModel.ConstantValue := 0.78;
PooledModel.ConfidenceLevel := 0.75;
PooledModel.MissingData.Method := MissingDataMethod.SampleAverage;
Weights := PooledModel.Weights;
Weights.Clear;
ARMA := PooledModel.ARMA;
OrderAR[0] := 1;
OrderAR[1] := 3;
ARMA.OrderAR := OrderAR;
(Model As IMetabaseObject).Save;
End Sub UserProc;
After executing the example some parameters of model calculation are changed. The list of changed parameters and its new values:
Model type - model with fixed effects.
Mode of setting constant - value is determined by user.
Constant value - 0.78.
Significance of confidence limits - 0.75.
Missing data treatment method - average.
Weights are not taken place in model calculation.
Autoregression order is specified.
All changes in model calculation options are saved.
See also: