`dataFrame.columns.stat — binaryAsymmetry()`

Description

The binaryAsymmetry() method of the stat object identifies binary columns whose majority category proportion exceeds the specified asymmetry threshold.

Signature

dataFrame.columns(...columnNames).stat.binaryAsymmetry({ threshold: 0.2 })

Scope: columns
Family: stat
Returns: array

Arguments

...columnNames ( string[] ): The name of the columns from which to compute the binary asymmetry detection.
options (object): Binary asymmetry detection options.

Option

threshold (string): The maximum tolerated proportion of the minority category. (Default: 0.2.)

Returns

stat ( object[] )

An array containing the asymmetry details for each selected column.

column (string)

The column name.

proportion (number)

The proportion of the majority category.

asymmetric (boolean)

Indicates whether the majority category proportion exceeds the specified threshold.

If asymmetric: true
Otherwise: false

Notes

Only binary columns are evaluated.
proportion represents the proportion of the majority category and ranges from 0.5 to 1.
A column is considered asymmetric when the proportion of its majority category exceeds 1 - threshold.
With the default threshold of 0.2, a binary column is considered asymmetric when one category represents more than 80% of the observations.

Example

// test the binary symmetry of 2 columns of the dataFrame
var stat = dataFrame.columns('groupA', 'groupB').stat.binaryAsymmetry();

// log the stat details
notebook.log(stat);