ISummaryStatistics.LogL

Syntax

LogL: Double;

Description

The LogL property returns log-likelihood function.

Comments

It is used in regression model tests for redundant or missing variables.

Example

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

Sub UserProc;
Var
    LinearR: SmLinearRegress;
    can, fr: Array[9Of Double;
    res, i: Integer;
    Con: IIntercept;
    ss: ISummaryStatistics;
Begin
    LinearR := New SmLinearRegress.Create;
    For i := 0 To 8 Do
        can[i] := 1230 + i * 302;
        fr[i] := 579.5 + i * 9.4;
    End For;
    // Set model parameters
    LinearR.Explained.Value := can;
    LinearR.Explanatories.Add.Value := fr;
    Con := LinearR.ModelCoefficients.Intercept;
    con.Mode := InterceptMode.ManualEstimate;
    con.Estimate := 35.7;
    // Run calculation
    res := LinearR.Execute;
    ss := LinearR.SummaryStatistics;
    Debug.Write("Log-likelihood function: ");
    Debug.WriteLine(ss.LogL);
    Debug.Write("Restricted log-likelihood function: ");
    Debug.WriteLine(ss.RestrLogL);
    Debug.Write("p-value of probability for LR-statistics: ");
    Debug.WriteLine(ss.LRprobability);
    Debug.Write("LR-statistic: ");
    Debug.WriteLine(ss.LRstatistic);
End Sub UserProc;

After executing the example the console window displays summary statistics:

See also:

ISummaryStatistics