loess.ifit(interpolator, x.evaluate, parametric)

interpolator:

k-d tree obtained from loess.raw.

x.evaluate:

a data frame expanded from a rectangular marginal grid points (points of evaluations) in the space of the predictors (see expand.grid).

parametric:

for two or more numeric predictors, this argument specifies those variables that should be conditionally-parametric. It should be specified as a logical vector of length equal to the number of columns in x.evaluate.