R: Recipes for Analysis, Visualization and Machine Learning

Diagnosing a generalized additive model

GAM also provides diagnostic information about the fitting procedure and results of the generalized additive model. In this recipe, we demonstrate how to plot diagnostic plots through the gam.check function.

Getting ready

Ensure that the previous recipe is completed with the gam fitted model assigned to the fit variable.

How to do it...

Perform the following step to diagnose the generalized additive model:

1. Generate the diagnostic plot using gam.check on the fitted model:

   > gam.check(fit)
   
   Method: GCV   Optimizer: magic
   Smoothing parameter selection converged after 7 iterations.
   The RMS GCV score gradient at convergence was 8.79622e-06 .
   The Hessian was positive definite.
   The estimated model rank was 10 (maximum ...