[ top | up ]

functions for subset selection

Usage

subsets(x=, ...)

subsets.formula(formula=, data=, weights=rep(1, length(y)), nbest=1, nvmax=8, force.in=NULL, force.out=NULL, intercept=T, method=c("exhaustive", "backward", "forward", "seqrep"), really.big=F)

subsets.default(x=, y=, weights=rep(1, length(y)), nbest=1, nvmax=8, force.in=NULL, force.out=NULL, intercept=T, method=c("exhaustive", "backward", "forward", "seqrep"), really.big=F)

Arguments

formula model formula for full model
data Optional data frame
x design matrix with all predictors
y response vector
weights weight vector
nbest number of subsets of each size to record
nvmax maximum size of subsets to examine
force.in index to columns of design matrix that should be in all models
force.out index to columns of design matrix that should be in no models
intercept Add an intercept?
method Use exhaustive search, forward selection, backward selection or sequential replacement to search.
really.big Must be T to performe exhaustive search on more than 50 variables.

Description

Generic function for regression subset selection with methods for formula and matrix arguments.

Value

An object of class "leaps" containing no user-serviceable parts. It is designed to be processed by summary.leaps. If you want to understand the components of this object then read the source.

Note

This function improves on leaps() in several ways. The design matrix need not be of full rank. The ability to restrict nvmax speeds up exhaustive searches considerably. There is no hard-coded limit to the number of variables.

See Also

leaps(), summary.leaps

Examples

data(swiss)
a<-subsets(as.matrix(swiss[,-1]),swiss[,1])
b<-summary(a)