`dataFrame.columns.stat — predict()`

Description

The predict() method of the stat object computes predicted values for each row using a fitted regression model.

Signature

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

Scope: columns
Family: stat
Returns: mask

Arguments

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

Option

model (string)

The regression model used to generate predictions.

linear (default)
theilSen
siegelRepeatedMedian

Returns

mask ( number[] ): The predicted values computed for each row.

Notes

The method requires exactly two selected numeric columns.
The first selected column is interpreted as the predictor variable.
The second selected column is used to fit the regression model.
One predicted value is returned for each observation of the predictor column.
The linear model uses ordinary least squares regression.
The theilSen model provides robust predictions based on pairwise slopes.
The siegelRepeatedMedian model provides highly robust predictions resistant to outliers.
Predicted values are computed from the fitted regression line:

Example

// computes predicted values for 2 columns of the dataFrame
var mask = dataFrame.columns('x', 'y').stat.predict();

// add the predicted values to the dataFrame
dataFrame.column('y_pred').set(mask);