IExponentialSmoothingAutoSearch.AlphaSearch

Syntax

AlphaSearch: Boolean;

Description

The AlphaSearch property determines whether the Alpha parameter is to be searched.

Comments

AlphaSearch = True. Alpha parameter is found automatically.

AlphaSearch = False. Default value. The Alpha parameter is defined by the user (use the IExponentialSmoothingParameters.Alpha property).

Example

The following example describes setting parameters and calculation of exponential smoothing for data series set by the real array "serie". To execute the example you need to add a link to the system assembly Stat.

Sub UserProc;
Var
    Method: SmExponentialSmoothing;
    serie: Array Of Double;
    status: Integer;
    Seasonal: ISeasonal;
    Auto: IExponentialSmoothingAutoSearch;
    Grid: IExponentialSmoothingQuickestDescentMethod;
Begin
    Method := New SmExponentialSmoothing.Create;
    serie := New Double[15]; serie[7] := 1064.74221;
    serie[0] := 670.2000183; serie[8] := 1033.324656;
    serie[1] := 576.0680563; serie[9] := 780.8584552;
    serie[2] := 717.6484268; serie[10] := 657.5033113;
    serie[3] := 856.9105808; serie[11] := 654.5472579;
    serie[4] := 885.4609516; serie[12] := 678.2380139;
    serie[5] := 1011.846431; serie[13] := 642.4128544;
    serie[6] := 995.4496292; serie[14] := 751.9611194;
    Method.Serie.Value := serie;
    Method.Forecast.LastPoint := 30;
    Seasonal := Method.SeasonalComponent;
    Seasonal.Mode := SeasonalityType.Additive;
    Seasonal.Cycle := 4;
    Method.TrendComponent := TrendType.Damped;
    Auto := Method.AutoSearch;
    Auto.Criterion := CriterionType.SumOfSquareError;
    Auto.Mode := SearchType.Grid;
    Grid := Auto.QuickestDescentMethod;
    Grid.GridStep := 0.1;
    Auto.AlphaSearch := True;
    Auto.GammaSearch := True;
    Auto.DeltaSearch := False;
    Grid.InitialApproximation.Delta := 0.001;
    Auto.PhiSearch := False;
    Grid.InitialApproximation.Phi := 0.4;
    status := Method.Execute;
    If status <> 0 Then
        Debug.WriteLine(Method.Errors);
    Else
        Debug.WriteLine("Alpha " + Method.BestModelCoefficients.Alpha.ToString);
        Debug.WriteLine("Delta " + Method.BestModelCoefficients.Delta.ToString);
        Debug.WriteLine("Gamma " + Method.BestModelCoefficients.Gamma.ToString);
        Debug.WriteLine("Phi " + Method.BestModelCoefficients.Phi.ToString);
    End If;
End Sub UserProc;

After executing the example the console window displays values of exponential smoothing parameters. Alpha and Gamma parameters are estimated automatically, while Delta and Phi parameters are given initially.