`dataFrame.columns.stat — residuals()`

Description

The residuals() method of the stat object computes the residuals between two selected columns for each row.

Signature

dataFrame.columns(...columnNames).stat.residuals()

Scope: columns
Family: stat
Returns: mask

Argument

...columnNames ( string[] ): The name of the columns from which to compute the residuals.

Returns

mask (array): An array containing the residual value for each row.

Notes

The method requires exactly two selected columns.
The first selected column is interpreted as the observed values.
The second selected column is interpreted as the predicted values.
Residuals are computed as observed - predicted.
Positive values indicate underestimation by the prediction.
Negative values indicate overestimation by the prediction.
A value of 0 indicates a perfect prediction.

Example

// calculate the residuals between the values of 2 columns of the dataFrame
var mask = dataFrame.columns('actual', 'predicted').stat.residuals();

// add the residuals to the dataFrame
dataFrame.column('residuals').set(mask);