vslSSEditOutliersDetection

Modifies array pointers related to multivariate outliers detection.

Syntax

status = vslssseditoutliersdetection(task, nparams, params, w)

status = vsldsseditoutliersdetection(task, nparams, params, w)

Include Files

  • mkl_vsl.f90

Input Parameters

Name

Type

Description

task

Fortran: TYPE(VSL_SS_TASK)

Descriptor of the task

nparams

Fortran: INTEGER

Pointer to the number of method parameters

params

Fortran: REAL(KIND=4) DIMENSION(*) for vslssseditoutliersdetection

REAL(KIND=8) DIMENSION(*) for vsldsseditoutliersdetection

Pointer to the array of method parameters

w

Fortran: REAL(KIND=4) DIMENSION(*) for vslssseditoutliersdetection

REAL(KIND=8) DIMENSION(*) for vsldsseditoutliersdetection

Pointer to an array of size n. The array holds the weights of observations to be marked as outliers.

Output Parameters

Name

Type

Description

status

Fortran: INTEGER

Current status of the task

Description

The vslSSEditOutliersDetection routine uses the parameters passed to replace

  • the pointers to the number of method parameters and to the array of the method parameters of size nparams

  • the pointer to the array that holds the calculated weights of the observations

If you pass a value of NULL for a specific input parameter, the value of that parameter in the task descriptor is unchanged.

Intel MKL provides the BACON algorithm ([Billor00]) for the detection of multivariate outliers. Pack the parameters of the BACON algorithm into the params array and pass them into the editor. Table "Structure of the Array of BACON Parameters" describes the params structure.

Structure of the Array of BACON Parameters

Array Position

Algorithm Parameter

Description

0

Method to start the algorithm

The parameter takes one of the following possible values:

VSL_SS_METHOD_BACON_MEDIAN_INIT, if the algorithm is started using the median estimate. This is the default value of the parameter.

VSL_SS_METHOD_BACON_MAHALANOBIS_INIT, if the algorithm is started using the Mahalanobis distances.

1

α

One-tailed probability that defines the (1 - α) quantile of χ2 distribution with p degrees of freedom. The recommended value is α/ n, where n is the number of observations. By default, the value is 0.05.

2

δ

Stopping criterion; the algorithm is terminated if the size of the basic subset is changed less than δ. By default, the value is 0.005.

Output of the algorithm is the vector of weights, BaconWeights, such that BaconWeights(i) = 0 if i-th observation is detected as an outlier. Otherwise BaconWeights(i) = w(i), where w is the vector of input weights. If you do not provide the vector of input weights, BaconWeights(i) is set to 1 if the i-th observation is not detected as an outlier.

See additional details about usage model of the algorithm in the Intel(R) MKL Summary Statistics Application Notes document [SS Notes].

For more complete information about compiler optimizations, see our Optimization Notice.