IValidationStatisticalOutliers.LowerPercentile

Syntax

LowerPercentile: Integer;

Description

The LowerPercentile property determines lower outlier border probability.

Comments

Range of property available values: [0;100].

To set upper outlier border probability, use the IValidationStatisticalOutliers.UpperPercentile property.

Example

Executing the example requires that the repository contains a time series database with the TSDB_VALIDATION_PARAMETERS identifier containing a validation rule with the VALIDATION identifier.

Add links to the Dimensions, Metabase, Ms system assemblies.

Sub UserProc;
Var
    Mb: IMetabase;
    RubrKey: Integer;
    ValidObj: IMetabaseObject;
    ValidFilter: IValidationFilter;
    StatisticalOutliers: IValidationStatisticalOutliers;
Begin
    Mb := MetabaseClass.Active;
    // Get time series database key
    RubrKey := Mb.GetObjectKeyById("TSDB_VALIDATION_PARAMETERS");
    // Get validation rule
    ValidObj := Mb.ItemByIdNamespace("VALIDATION", RubrKey).Edit;
    ValidFilter := ValidObj As IValidationFilter;
    // Set calendar frequency of checked data
    ValidFilter.Level := DimCalendarLevel.Year;
    // Set validation rule type
    ValidFilter.Kind := ValidationDetailsKind.StatisticalOutliers;
    // Set up rule parameters
    StatisticalOutliers := ValidFilter.Details As IValidationStatisticalOutliers;
    // Set lower and upper outlier border probability
    StatisticalOutliers.LowerPercentile := 10;
    StatisticalOutliers.UpperPercentile := 20;
    // Use error inverting
    StatisticalOutliers.InvertException := True;
    // Find statistical outliers by frequency points
    StatisticalOutliers.OutliersType := ValidationStatisticalOutliersType.CrossDate;
    // Clear execution area and validation filter
    StatisticalOutliers.SelectionTargetData.Clear;
    StatisticalOutliers.SelectionFilter.Clear;
    // Save the rule
    ValidObj.Save;
End Sub UserProc;

After executing the example the VALIDATION validation rule is reconfigured. The following parameters are set for the rule:

The Statistical Outliers Analysis rule type.
Lower outlier border probability - 10.
Upper outlier border probability - 20.
Filtering of checked data is not executed.
Outliers are found by annual frequency points.