prune_se: Prune an 'rpart' object

In bgreenwell/treemisc: Data Sets and Functions to Accompany "Tree-Based Methods for Statistical Learning in R"

Prune an rpart object

Description

Prune an rpart object using the standard error (SE) of the cross-validation results.

Usage

prune_se(object, prune = TRUE, se = 1)

Arguments

object	An object that inherits from class "rpart".
prune	Logical indicating whether or not to return the pruned decision tree. Default is TRUE. If FALSE, the optimal value of the cost-complexity parameter is returned instead.
se	Numeric specifying the number of standard errors to use when pruning the tree. Default is 1, which corresponds to the 1-SE rule described in Breiman et al. (1984).

Value

Either an object that inherits from class "rpart" (ideally, one that's been simplified using cost-complexity pruning with the 1-SE rule) or a numeric value representing the cost-complexity parameter to use for pruning.

References

Breiman, L., Friedman, J., and Charles J. Stone, R. A. O. (1984). Classification and Regression Trees. The Wadsworth and Brooks-Cole statistics-probability series. Taylor & Francis.

See Also

prune

bgreenwell/treemisc documentation built on Oct. 26, 2022, 12:56 a.m.