ISlPDLTerm.InitValues

Syntax

InitValues: Array;

Description

The InitValues property determines initial approximations of lag variable.

Comments

To get lag variable coefficients, use the ISlPDLTerm.BetaCoefficients property.

Example

To execute the example, add a link to the Stat system assembly.

Sub UserProc;
Var
    Method: SmLinearRegress;
    Factors: ISlSeries;
    Serie, Factor, ger, jpn: Array[20] Of Double;
    arr: array Of Double;
    PDLTermCollect: ISlPDLTermCollection;
    PDLTerm: ISlPDLTerm;
    BetaCoef: ICoefficients;
    res, i: Integer;
Begin
    Method := New SmLinearRegress.Create;
    // explained variable
    Serie[00] := 6209; Serie[10] := 7132;
    Serie[01] := 6385; Serie[11] := 7137;
    Serie[02] := 6752; Serie[12] := 7473;
    Serie[03] := 6837; Serie[13] := 7722;
    Serie[04] := 6495; Serie[14] := 8088;
    Serie[05] := 6907; Serie[15] := 7022;
    Serie[06] := 7349; Serie[16] := 8282;
    Serie[07] := 7213; Serie[17] := 9064;
    Serie[08] := 7061; Serie[18] := 9380;
    Serie[09] := 7180; Serie[19] := 9746;
    // explanatory variable

    Factor[00] := 4110; Factor[10] := 5948;
    Factor[01] := 4280; Factor[11] := 6218;
    Factor[02] := 4459; Factor[12] := 6521;
    Factor[03] := 4545; Factor[13] := 6788;
    Factor[04] := 4664; Factor[14] := 7222;
    Factor[05] := 4861; Factor[15] := 7486;
    Factor[06] := 5195; Factor[16] := 7832;
    Factor[07] := 5389; Factor[17] := 8153;
    Factor[08] := 5463; Factor[18] := 8468;
    Factor[09] := 5610; Factor[19] := 9054;
    // first lag variable

    ger[00] := 3415; ger[10] := 6633;
    ger[01] := 3673; ger[11] := 6910;
    ger[02] := 4013; ger[12] := 7146;
    ger[03] := 4278; ger[13] := 7248;
    ger[04] := 4577; ger[14] := 7689;
    ger[05] := 5135; ger[15] := 8046;
    ger[06] := 5388; ger[16] := 8143;
    ger[07] := 5610; ger[17] := 8064;
    ger[08] := 5787; ger[18] := 8556;
    ger[09] := 6181; ger[19] := 9177;
    // second lag variable

    jpn[00] := 1475; jpn[10] := 3052;
    jpn[01] := 1649; jpn[11] := 3453;
    jpn[02] := 1787; jpn[12] := 3666;
    jpn[03] := 1884; jpn[13] := 4008;
    jpn[04] := 1972; jpn[14] := 4486;
    jpn[05] := 2108; jpn[15] := 4663;
    jpn[06] := 2249; jpn[16] := 5115;
    jpn[07] := 2394; jpn[17] := 5655;
    jpn[08] := 2505; jpn[18] := 6358;
    jpn[09] := 2714; jpn[19] := 6995;

    // set explained and explanatory variables
    Method.Explained.Value := Serie;
    Factors := Method.Explanatories;
    Factors.Add.Value := Factor;
    // define regression parameters
    Method.ModelPeriod.FirstPoint := 1;
    Method.ModelPeriod.LastPoint := 20;
    // Set last forecast point
    Method.Forecast.LastPoint := 30;
    // set lag variables
    PDLTermCollect := Method.PDLTermCollection;
    If PDLTermCollect.Count > 0 Then
        PDLTermCollect.Clear;
    End If;

    // set parameters of first lag variable
    PDLTerm := PDLTermCollect.Add;
    PDLTerm.Explanatory.Value := ger;
    PDLTerm.Explanatory.Include := True;
    PDLTerm.PDLConstraint := PDLConstraintType.Both;
    PDLTerm.PolinomialDegreeP := 4;
    PDLTerm.LagLengthK := 1;
    arr:=New Double[PDLTerm.LagLengthK+1];
    For i:=0 To arr.Length-1 Do
        arr[i]:=0.5-i*0.1;
    End For;
    PDLTerm.InitValues := arr;
    // set parameters of second lag variable
    PDLTerm := PDLTermCollect.Add;
    PDLTerm.Explanatory.Value := jpn;
    PDLTerm.PDLConstraint := PDLConstraintType.FarEnd;
    PDLTerm.PolinomialDegreeP := 2;
    PDLTerm.LagLengthK := 2;

    // Autofit parameters
    Method.AutoSelection.IsActive := True;
    Method.AutoSelection.Min := 1;
    Method.AutoSelection.Max := 2;
    // calculate model
    res := Method.Execute;
    // calculate method with lag variables, output results
    BetaCoef := PDLTerm.BetaCoefficients;
    Debug.WriteLine("=== Value of coefficients of first lag variable  ===");
    PDLTerm := PDLTermCollect.Item(0);
    For i := 0 To BetaCoef.Estimate.Length-1 Do
        Debug.WriteLine((i+1).ToString + " " + BetaCoef.Estimate[i].ToString);
    End For;
    Debug.WriteLine("=== Value of coefficients of second lag variable  ===");
    PDLTerm := PDLTermCollect.Item(1);
    For i := 0 To BetaCoef.Estimate.Length-1 Do
        Debug.WriteLine((i+1).ToString + " " + BetaCoef.Estimate[i].ToString);
    End For;
End Sub UserProc;

After executing the example the console window displays values of coefficients of the first and second lag variables.