Outliers.jl

An outlier discovering package for Julia.

At this stage, a naive outlier discovering for numerical vectors is implemented, using the BoxPlot Rules.

This package will be extended for more functionalities in the future.

Outliers and BoxPlot Rules

This method works with 1 dimensional numerical vectors. Suppose we have a random vector of float numbers, then manually append a relatively big number at the end, and we want to find out the outlier(s) among such vector. This can be done using outliers function as following:

using Outliers

a = [randn(10), 100.] # Generate 10 random float numbers, and append a relatively large number as "outlier"
o, idc = outliers(a) # returns the outliers, and according indices in vector