effective degrees of freedom of the centroids per class if
subclass centroid shrinkage is performed.
tot.df
specification of the total degrees of freedom as an alternative
to sub.df.
dimension
dimension of the discriminant subspace to use for
prediction.
eps
numeric threshold for automatically truncating the dimension.
iter
limit on the total number of iterations.
method
regression function used in optimal scaling. The default of
linear regression is provided by polyreg from the
mda package. For penalized mixture discriminant models,
gen.ridge is appropriate. Other possibilities are
mars for multivariate adaptive regression splines and
bruto for adaptive backfitting of additive splines. Use
the . operator to quote specified functions.
trace
logical indicating whether iteration information is printed.
...
additional arguments to mda.start and method.
Details
Response types:
factor
Automatic tuning of grid parameter:
subclasses
The predict function for this model additionally accepts the
following argument.
prior
prior class membership probabilities for prediction data
if different from the training set.
Default argument values and further model details can be found in the source
See Also links below.