ISmLinearRegress.AutoSelection

Syntax

AutoSelection: ILinearRegressionAutoSelection;

Description

The AutoSelection property determines regressor autoselection parameters.

Example

Add a link to the Stat system assembly.

Sub AutoSel;
Var
    LR: ISmLinearRegress;
    res, i, Inc: Integer;
    y, x1, x2, x3: Array Of Double;
    ASel: ILinearRegressionAutoSelection;
Begin
    LR := New SmLinearRegress.Create;
    y := New Double[21];
    x1 := New Double[21];
    x2 := New Double[21];
    x3 := New Double[21];

    // Values of series y, x1, x2, x3
    y[00] := 0.401; x1[00] := -9999.99;
    y[01] := 0.225; x1[01] := 0.137142857;
    y[02] := 0.191; x1[02] := 0.306532663;
    y[03] := 0.534; x1[03] := 0.365384615;
    y[04] := 0.428; x1[04] := 0.38028169;
    y[05] := 0.249; x1[05] := -0.073469388;
    y[06] := 0.152; x1[06] := -0.00660793;
    y[07] := 0.13; x1[07] := 0.028824834;
    y[08] := 0.1; x1[08] := 0.045258621;
    y[09] := 0.051; x1[09] := 0.053608247;
    y[10] := 0.008; x1[10] := 0.056751468;
    y[11] := 0.046; x1[11] := 0.02962963;
    y[12] := 0.053; x1[12] := 0.03057554;
    y[13] := 0.03; x1[13] := 0.008726003;
    y[14] := 0.005; x1[14] := 0.01384083;
    y[15] := 0.018; x1[15] := 0.017064846;
    y[16] := 0.03; x1[16] := 0.025167785;
    y[17] := 0.01; x1[17] := 0.014729951;

    y[18] := 0.438; x1[18] := 1.591935484;
    y[19] := 0.233; x1[19] := 0.285003111;
    y[20] := 0.16; x1[20] := 0.17094431;
    x2[00] := 0.6; x3[00] := -9999.99;
    x2[01] := 0.43; x3[01] := 0.503916449;
    x2[02] := 0.32; x3[02] := 0.296875;
    x2[03] := 0.11; x3[03] := 0.269076305;
    x2[04] := 0.41; x3[04] := 0.097046414;
    x2[05] := 0.52; x3[05] := 0.458653846;
    x2[06] := 0.34; x3[06] := 0.147659855;
    x2[07] := 0.03; x3[07] := 0.22860425;
    x2[08] := 0.1; x3[08] := 0.095371669;
    x2[09] := 0.05; x3[09] := 0.103713188;
    x2[10] := 0.14; x3[10] := 0.034029389;
    x2[11] := 0.03; x3[11] := 0.07816006;
    x2[12] := 0.05; x3[12] := 0.060700659;
    x2[13] := 0.02; x3[13] := 0.151079137;
    x2[14] := -0.017; x3[14] := 0.03125;

    x2[15] := 0.01; x3[15] := 0.030578512;
    x2[16] := 0.03; x3[16] := -0.036621224;
    x2[17] := 0.008; x3[17] := 0.022752497;
    x2[18] := -0.011; x3[18] := -0.00759631;
    x2[19] := -0.005; x3[19] := 0.225533078;
    x2[20] := 0.053; x3[20] := 0.056881553;
    // Set explained and explanatory series
    LR.Explained.Value := y;
    LR.Explanatories.Add.Value := x1;
    LR.Explanatories.Add.Value := x2;
    LR.Explanatories.Add.Value := x3;
    // Set sample period parameters
    LR.ModelPeriod.FirstPoint := 1;
    LR.ModelPeriod.LastPoint := 21;
    // Exclude constant from calcualtion
    LR.ModelCoefficients.Intercept.Mode := InterceptMode.None;
    // Set regressor autoselection parameters
    ASel := LR.AutoSelection;
    ASel.IsActive := True;

    ASel.Min := 1;
    ASel.Max := 2;
    ASel.Criterion := CriterionType.R2adj;
    // Run calculation and show results
    res := LR.Execute;
    If res <> 0 Then
        Debug.WriteLine("Errors " + LR.Errors);
    Else
        For i := 0 To LR.Explanatories.Count - 1 Do
            Inc := ASel.Results[i];
            Debug.WriteLine(i.ToString + ": " + Inc.ToString);
        End For;
    End If;
End Sub AutoSel;

After executing the example the console window displays the vector that contains autoselection results:

Module execution started
0: 0
1: 1
2: 0
Module execution finished