`dataFrame.columns.stat — regression()`

Description

The regression() method of the stat object fits a regression model between two selected columns.

Signature

dataFrame.columns(...columnNames).stat.regression({ model: 'linear' })

Scope: columns
Family: stat
Returns: object

Arguments

...columnNames ( string[] ): The name of the columns from which to compute the regression.
options (object): Regression model options.

Option

model (string)

The regression model used to estimate the relationship.

linear (default)
theilSen
siegelRepeatedMedian

Returns

regression (number)

A regression model object containing the estimated parameters.

slope (number): The estimated slope coefficient.
intercept (number): The estimated intercept coefficient.

Notes

The method requires exactly two selected numeric columns.
The first selected column is interpreted as the predictor variable.
The second selected column is interpreted as the response variable.
The linear model uses ordinary least squares regression.
The theilSen model provides a robust estimate of the slope based on pairwise slopes.
The siegelRepeatedMedian model provides a highly robust estimate resistant to outliers.
The fitted model can be expressed as: y = mx + b
Robust regression models are generally less sensitive to outliers than ordinary least squares regression.

Example

// evaluate the regression between the values of 2 columns of the dataFrame
var regression = dataFrame.columns('groupA', 'groupB').stat.regression();

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