IModelling.Collapse

Collapse(Input: ITimeSeries;
         Method: MsCollapseType;
         TargetFrequency: MsFrequency;
         [MissingData: MissingDataMethod =16;]
         [NumberOfPoints: Integer = 1;]
         [SpecifiedValue: Double = 0;]
         [AdditionalSeries: ITimeSeries = Null]): Variant;

Fore.NET Syntax

Collapse(Input: Prognoz.Platform.Interop.Ms.ITimeSeries;
         Method: Prognoz.Platform.Interop.Ms.MsCollapseType;
         TargetFrequency: Prognoz.Platform.Interop.Ms.MsFrequency;
         Context: Prognoz.Platform.Interop.Fore.ForeRuntimeContext;
         MissingData: Prognoz.Platform.Interop.Stat.MissingDataMethod;
         NumberOfPoints: integer;
         SpecifiedValue: double;
         AdditionalSeries: Prognoz.Platform.Interop.Ms.ITimeSeries): object;

Parameters

NumberOfPoints. Additional parameter for missing data treatment method. It can take only positive values. It is used for the following missing data treatment methods:
• MissingDataMethod.NPointsAverage. N-Point Average. The parameter determines the number of neighbor points.
• PreviousGrowthRate, SucceedingGrowthRate. Previous Growth Rate, Succeeding Growth Rate. The parameter determines the number of periods.

SpecifiedValue. The value, with which missing data is substituted by the MissingDataMethod.SpecifiedValue Value method.

AdditionalSeries. The series that is used to substitute missing data by the MissingDataMethod.Pattern Pattern and MissingDataMethod.Overlay Overlay methods.

Description

Comments

This method aggregates data from the lower level to the higher level; it is used to calculate problems that require data aggregation. For example, data calculated by months must be summed for quarterly frequency.

On data aggregation, the following calendar frequency settings are taken into account: start period offset relative to its start/end.

The period under consideration: first two weeks of May 2012.

Aggregation method: Sum.

Calendar frequency settings: input variable contains weekly and daily frequencies; week start offset is not used; daily frequency contains days from Monday to Friday. Aggregation result:

Week	Period start date	Day of the week	Raw data	Calculated value
1st week	May 7 2012	Monday		0.50
	May 7 2012	Monday	0.10
	May 8 2012	Tuesday	0.10
	May 9 2012	Wednesday	0.10
	May 10 2012	Thursday	0.10
	May 11 2012	Friday	0.10
2nd week	May 14 2012	Monday		1.00
	May 14 2012	Monday	0.20
	May 15 2012	Tuesday	0.20
	May 16 2012	Wednesday	0.20
	May 17 2012	Thursday	0.20
	May 18 2012	Friday	0.20

Calendar frequency settings: input variable contains weekly and daily frequencies; week start is offset on 3 days forward; daily frequency contains days from Monday to Friday. Aggregation result:

Week	Period start date	Day of the week	Raw data	Calculated value
1st week	May 2 2012	Wednesday		0.50
	May 2 2012	Wednesday	0.10
	May 3 2012	Thursday	0.10
	May 4 2012	Friday	0.10
	May 7 2012	Monday	0.10
	May 8 2012	Tuesday	0.10
2nd week	May 9 2012	Wednesday		1.00
	May 9 2012	Wednesday	0.20
	May 10 2012	Thursday	0.20
	May 11 2012	Friday	0.20
	May 14 2012	Monday	0.20
	May 15 2012	Tuesday	0.20

Fore Example

Executing the example requires that the repository contains a modeling container with the MS identifier. This container contains a model with the MODEL_D_Q identifier and quarterly frequency, that is calculated using a determinate equation and includes at least one factor.

Var
    Mb: IMetabase;
    ModelSpace, ModelObj: IMetabaseObject;
    Transf: IMsFormulaTransform;
    Formula: IMsFormula;
    Model: IMsModel;
    Determ: IMsDeterministicTransform;
    TransVar: IMsFormulaTransformVariable;
    Slice: IMsFormulaTransformSlice;
    TermInfo: IMsFormulaTermInfo;
    Expr: IExpression;
Begin
    // Get current repository
    Mb := MetabaseClass.Active;
    // Get modeling container
    ModelSpace := Mb.ItemById("MS").Bind;
    // Get model
    ModelObj := Mb.ItemByIdNamespace("MODEL_D_Q", ModelSpace.Key).Edit;
    Model := ModelObj As IMsModel;
    // Get model parameters
    Transf := Model.Transform;
    // Get model calculation method
    Formula := Transf.FormulaItem(0);

    Determ := Formula.Method As IMsDeterministicTransform;
    // Get first factor of model
    TransVar := Transf.Inputs.Item(0);
    // Get slice, corresponding to factor
    Slice := TransVar.Slices.Item(0);
    // Create tern based on the first factor
    TermInfo := Transf.CreateTermInfo;
    TermInfo.Slice := Slice;
    // Get expression for model calculation

    Expr := Determ.Expression;
    // Set expression for model calculation
    Expr.AsString := "Collapse(" + TermInfo.TermInnerText + ", MsCollapseType.Total, " +
        "MsFrequency.Annual, MissingDataMethod.NPointsAverage, 4)";
    // Check expression correctness
    If Expr.Valid
        // If expression is correct, save changes
        Then ModelObj.Save;
        // If expression is incorrect, output message to console window
        Else Debug.WriteLine("Model is not saved: error in formula");
    End If;
End Sub UserProc;

After executing the example the model aggregates data of the first input variable with quarterly frequency for annual one with missing data treatment by means of the N-Point Average method where N = 4.

Fore.NET Example

The requirements and result of the Fore.NET example execution match with those in the Fore example.

Imports Prognoz.Platform.Interop.Ms;
…
Public Shared Sub Main(Params: StartParams);
Var
    Mb: IMetabase;
    ModelSpace, ModelObj: IMetabaseObject;
    Transf: IMsFormulaTransform;
    Formula: IMsFormula;
    Model: IMsModel;
    Determ: IMsDeterministicTransform;
    TransVar: IMsFormulaTransformVariable;
    Slice: IMsFormulaTransformSlice;
    TermInfo: IMsFormulaTermInfo;
    Expr: IExpression;
Begin
    // Get current repository
    Mb := Params.Metabase;
    // Get modeling container
    ModelSpace := Mb.ItemById["MS"].Bind();
    // Get model
    ModelObj := Mb.ItemByIdNamespace["MODEL_D_Q", ModelSpace.Key].Edit();
    Model := ModelObj As IMsModel;
    // Get model parameters
    Transf := Model.Transform;
    // Get model calculation method
    Formula := Transf.FormulaItem[0];
    Determ := Formula.Method As IMsDeterministicTransform;
    // Get first factor of model
    TransVar := Transf.Inputs.Item[0];

    // Get slice, corresponding to factor
    Slice := TransVar.Slices.Item[0];
    // Create term based on the first factor
    TermInfo := Transf.CreateTermInfo();
    TermInfo.Slice := Slice;
    // Get expression for model calculation
    Expr := Determ.Expression;
    // Set expression for model calculation
    Expr.AsString := "Collapse(" + TermInfo.TermInnerText + ", MsCollapseType.Total, " +
        "MsFrequency.Annual, MissingDataMethod.NPointsAverage, 4)";

    // Check expression correctness
    If Expr.Valid
        // If expression is correct, save changes
        Then ModelObj.Save();
        // If expression is incorrect, output message to console window
        Else System.Diagnostics.Debug.WriteLine("Model is not saved: error in formula");
    End If;
End Sub;

Example of Use in Expressions

collapse({Brazil|BCA}, MsCollapseType.Total, MsFrequency.Annual, MissingDataMethod.NPointsAverage, 4)

Result: data is aggregated for the Brazil|BCA time series for annual frequency by summing method with missing data treatment by the N-Point Average method where N = 4.

Use: it can be used in formulas of calculated series of time series database and in formulas of attribute-based models of modeling container.

Collapse(X1,MsCollapseType.Average,MsFrequency.Annual, MissingDataMethod.LinTrend)

Result: for the X1 factor data aggregation for annual frequency using average value calculation method with the Linear Trend missing data treatment method.