Detect outliers using MAD from the median for asymmetric distributions

Outlier detection based on Median Absolute Deviation (MAD) for asymmetric distributions. The function calculates the distance to the median for every value in the distribution relative to the left or right side MAD. It then compares the value to your threshold and labels the outliers.

Usage

is_outlier_double_mad(x, zero_mad_action = NULL, threshold = 3.5)

Arguments

x

A vector of numeric values.

zero_mad_action

Determines the action in the event of an MAD of zero. Defaults to NULL. The options are:

• NULL: process runs with no warning

• "warn": a warning will be displayed

• "stop": process is stopped

threshold

Z-score threshold (defaults to 3.5).

Value

A logical vector.

Examples

x <- c(1, 2, 3, 3, 4, 4, 4, 5, 5.5, 6, 6, 6.5, 7, 7, 7.5, 8, 9, 12, 52, 90)

is_outlier_double_mad(x)
#>  [1] FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE FALSE
#> [13] FALSE FALSE FALSE FALSE FALSE  TRUE  TRUE  TRUE