MVN (version 4.0)

mvOutlier: Multivariate Outlier Detection

Description

This function detects multivariate outliers based on Mahalanobis distance and adjusted Mahalanobis distance.

Usage

mvOutlier(data, qqplot = TRUE, alpha = 0.5, tol = 1e-25, method = c("quan", "adj.quan"), label = TRUE, position = NULL, offset = 0.5)

Arguments

data
a numeric matrix or data frame
qqplot
if TRUE it creates a chi-square Q-Q plot
alpha
a numeric parameter controlling the size of the subsets over which the determinant is minimized. Allowed values for the alpha are between 0.5 and 1 and the default is 0.5.
tol
a numeric tolerance value which isused for inversion of the covariance matrix (default = 1e-25).
method
quan for Mahalanobis distance and adj.quan for adjusted Mahalanobis distance.
label
an optional term to display outlier labels (i.e. observation number) on the Q-Q plot.
position
a position specifier for the text. Values of 1, 2, 3 and 4, respectively indicate positions below, to the left of, above and to the right of the specified coordinates.
offset
when pos is specified, this value gives the offset of the label from the specified coordinate in fractions of a character width.

Value

outlier
an outlier set
newdData
new data set without possible outliers

See Also

mardiaTest roystonTest hzTest mvnPlot uniPlot uniNorm

Examples

Run this code
setosa = iris[1:50, 1:3] # Iris data only for setosa and three variables
result = mvOutlier(setosa, qqplot = TRUE, method = "quan", label = TRUE)
result

Run the code above in your browser using DataCamp Workspace