xvalidate {SDDA}R Documentation

Produces cross validated fitted values for sdda.

Description

Ten-fold cross-validation, or leave one out cross validation (default).

Usage

xvalidate(X, y, method=sdda, fold = NULL, trace = FALSE, ...)

Arguments

X real matrix with n rows (samples) and p columns (variables or genes)
y vector of class labels
method sdda is the only option in this version of the library
fold an integer specifying the number of folds to use in the cross validation. Default value NULL gives leave one out cross validation.
trace a logical variable with value TRUE if information about cross validation progress to be output and value FALSE if no output required.
... Other terms as appropriate

Details

The data is divided into fold groups. The fitted value for each group is obtained by using a model built from the data for the remaining groups. For sdda the usual cross validated class labels are produced.

Value

An n by 1 vector of cross validated fitted values.

Note

Missing values in X and y are not allowed

Author(s)

Glenn Stone


[Package SDDA version 1.0-4 Index]