varSelect {qtlmt}

Variable selection

Description

Select a specific number of variables via variable selection that are optimal for multivariate analysis.

Usage

varSelect(x, group, scope, nv, direction=c("backward","forward"))

Arguments

x

a data matrix/frame. Columns are variables to select from.

group

a grouping indicator of observations.

scope

which variables (i.e., columns of x) to select from.

nv

how many variables to be selected.

direction

forward selection or backward elimination.

Value

variables selected in the model.

See Also

varGroup and varStep

Examples

data(etrait)
g55<- varGroup(traits, mdat, zscope=55, method="best")
idx<- sample(1:nrow(traits), replace=FALSE)
varSelect(traits[idx,], group=mdat[,55], scope=1:ncol(traits[idx,]),
   nv=length(g55[[1]]))