POCRE (version 0.6.0)

pocrescreen: Screen Variables Using Penalized Orthogonal-Components Regression (POCRE)

Description

Screen for a pre-specified number (i.e., maxvar) of covariates by constructing maxcmp components with POCRE. Each component will be constructed by selecting maxvar/macmp covariates which are most relevant to the response variable(s). Here POCRE selects covariates for their top relevance to the response variable(s) without penalization.

Usage

pocrescreen(y, x, maxvar=nrow(x), maxcmp=5, x.include=NULL,
            tol=1e-6, maxit=100)

Arguments

y

n*q matrix, values of q response variables (allow for multiple response variables).

x

n*p matrix, values of p predicting variables (excluding the intercept).

maxvar

maximum number of selected variables.

maxcmp

maximum number of components to be constructed.

x.include

a vector of indices indicating covariates which should always be included in the model (so not counted into selected maxvar covariates).

tol

tolerance of precision in iterations.

maxit

maximum number of iterations to be allowed.

Value

a vector of indices of selected covariates (excluding those in x.include).

References

Zhang D (2018). R package POCRE: Exploring high-dimensional data via supervised dimension reduction. Manuscript.

Zhang D, Lin Y, and Zhang M (2009). Penalized orthogonal-components regression for large p small n data. Electronic Journal of Statistics, 3: 781-796.

See Also

pocre, pocrepath, cvpocre.

Examples

# NOT RUN {
data(simdata)
xx <- simdata[,-1]
yy <- simdata[,1]

# Screen for 50 covariates
sidx <- pocrescreen(yy,xx,maxvar=50)

# Screen for 50 additional covariates besides the first 10
xinc <- 1:10
sidx <- pocrescreen(yy,xx,maxvar=50,x.include=xinc)
sidx <- c(xinc,sidx)
# }