DistanceOutlier

scalation.modeling.DistanceOutlier
object DistanceOutlier extends Outlier

Detect outliers in the vector by treating anything that falls outside of some distance from the mean as an outlier. Common number of standard deviation units are 2.5, 2.7, 3 and 3.5. The larger the dataset, the greater the number of units that should be used.

Attributes

Graph
Supertypes
trait Outlier
class Object
trait Matchable
class Any
Self type

Members list

Value members

Concrete methods

def calcBounds(y: VectorD, factor: Double): (Double, Double)

Calculate the lower and upper bound for acceptable values for vector y. Treat anything that falls outside of some factor standard deviation units from the mean as an outlier.

Calculate the lower and upper bound for acceptable values for vector y. Treat anything that falls outside of some factor standard deviation units from the mean as an outlier.

Value parameters

factor

the factor used in computing the bound (# std deviations)

y

the vector with the possible outlier values

Attributes

Inherited methods

def findOutliers(y: VectorD, bounds: (Double, Double)): Set[Int]

Find/detect all outliers in vector y outside the bounds and return their element indices.

Find/detect all outliers in vector y outside the bounds and return their element indices.

Value parameters

bounds

the acceptable lower and upper bounds for element values

y

the vector with the possible outlier values

Attributes

Inherited from:
Outlier
def removeOutliers(x: MatrixD, y: VectorD, toRemove: Set[Int]): (MatrixD, VectorD)

Remove all outliers from matrix x and vector y specified by indices in toRemove.

Remove all outliers from matrix x and vector y specified by indices in toRemove.

Value parameters

toRemove

the indices of elements to be removed

x

the predictor matrix: y = f(x)

y

the vector with the possible outlier values

Attributes

Inherited from:
Outlier
def removeOutliers(y: VectorD, toRemove: Set[Int]): VectorD

Remove all outliers from vector y specified by indices in toRemove.

Remove all outliers from vector y specified by indices in toRemove.

Value parameters

toRemove

the indices of elements to be removed

y

the vector with the possible outlier values

Attributes

Inherited from:
Outlier

Inherited fields

protected val flaw: (String, String) => Boolean

Attributes

Inherited from:
Outlier