identifyOutlier: Identify outlier

In bozenne/butils: Various useful functions

Identify outlier

Description

Identify numeric or factor outliers

Usage

numOutlier(
  x,
  type = "auto",
  th.gaussian = 3,
  th.hampel = 3,
  th.boxplot = 1.5,
  na.rm = FALSE
)

factorOutlier(x, threshold = 0.01, use.NA = "ifany")

Arguments

x	a vector of numeric
type	the type of robust metric for assessing the "average" value
th.gaussian	the threshold for defining an outlier when using the mean
th.hampel	the threshold for defining an outlier when mad
th.boxplot	the threshold for defining the IQR
na.rm	should na be removed.
threshold	the threshold for defining an outlier regarding the prevalence (categorical variables)
use.NA	argument passed to table

References

idea from http://www.r-bloggers.com/finding-outliers-in-numerical-data/

Examples

## Not run: 
numOutlier(rnorm(1e3))

## End(Not run)

bozenne/butils documentation built on Oct. 14, 2023, 6:19 a.m.