[R] Simulations of GAM and MARS models : sample size ; Y-outliers and missing X-data

[Abby Spurdle]

> For me better all the x variables (collectively), to have m% missing
values.

I checked the mgcv documentation.
Observations with (any) missing values are ignored.
(i.e. The entire row, from your input data).

"If there are missing values in the reponse or covariates of a GAM then the
default is simply to use only the ‘complete cases’."

I haven't checked the ISLR package.

	[[alternative HTML version deleted]]