Kernel Ridge Regression: Options

The Plots dialog provides options for specifying the number of crossvalidation folds, display options, plot settings, and items to save.

Number of crossvalidation folds
The number of splits or folds in crossvalidation with grid search for model selection. Enter an integer value larger than 1. The default value is 5. The setting is available only when Model selection is chosen as the Mode on the primary Kernel Ridge Regression dialog.
Display
Provides options for specifying which output to display when crossvalidation is in effect.
Best
The default setting displays only basic results for the chosen best model.
Compare
Displays basic results for all evaluated models.
Compare models and folds
Displays full results for each split or fold for each evaluated model.
Plot
Provides options for specifying plots of observed or residual values versus predicted values.
Observed vs. Predicted
Displays a scatterplot of observed versus predicted values for the specified or best model.
Residuals vs. Predicted
Displays a scatterplot of residuals versus predicted values for the specified or best model.
Save
The table provides options for specifying variables to save to the active dataset.
Predicted values
Saves predicted values from the specified or best model to the active dataset. An optional variable name can be included.
Residuals
Saves residuals from the specified or best model predictions to the active dataset. An optional variable name can be included.
Dual coefficients
Saves dual or kernel space weight coefficients from the specified model to the active dataset. An optional variable name can be included. The setting is not available when Model selection is chosen as the Mode on the primary Kernel Ridge Regression dialog.

Specifying options

This feature requires Custom Tables and Advanced Statistics.

  1. From the menus choose:

    Analyze > Regression > Kernel Ridge...

  2. In the Kernel Ridge Regression dialog, click Options.
  3. Select the desired options and click OK.