getfinalSCADcoef {SIS}R Documentation

SCAD regularized loglikelihood for generalized linear models

Description

This function gets the final regression coefficients for the SCAD regularized loglikelihood for generalized linear models after applying (I)SIS.

Usage

getfinalSCADcoef(x, y, pickind, folds=NULL, eps0=1e-3, family=binomial(), 
tune.method="AIC", inittype='NoPen', detailed=FALSE)

INDEPgetfinalSCADcoef(x, y, pickind,  folds=NULL, xtune, ytune, 
family=binomial(), inittype='NoPen', eps0=1e-3, detailed=FALSE) 

Arguments

x an (n * p) matrix of features.
y an (n) vector of response.
pickind predictor indice selected by (I)SIS.
folds fold information for cross validation.
family a description of the error distribution and link function to be used in the model.
tune.method method for tuning regularization parameter.
inittype inittype specifies the type of initial solution for the one-step SCAD. It can be either NoPen or L1.
xtune, ytune independent tuning dataset.
eps0 an effective zero.
detailed indicates whether detailed information will be provided. Default is FALSE.

Value

An initial solution vector wt.initsoln and final solution (p+1) vector SCADcoef.

Author(s)

Jianqing Fan, Yang Feng, Richard Samworth, and Yichao Wu

References

Jianqing Fan and Runze Li (2001) Variable selection via nonconcave penalized likelihood and its oracle properties. Journal of American Statistical Association, 96, 1348-1360.

Hui Zou and Runze Li (2008) One-step Sparse Estimates in Nonconcave Penalized Likelihood Models (with discussion). The Annals of Statistics, 36, 1509-1533

See Also

scadglm, fullscadglm

Examples

set.seed(0)
b <- c(4,4,4,-6*sqrt(2))
n=150
p=200
truerho=0.5
corrmat=diag(rep(1-truerho, p))+matrix(truerho, p, p)
corrmat[,4]=sqrt(truerho)
corrmat[4, ]=sqrt(truerho)
corrmat[4,4]=1
cholmat=chol(corrmat)
x=matrix(rnorm(n*p, mean=0, sd=1), n, p)
x=x%*%cholmat
feta=x[, 1:4]%*%b
fprob=exp(feta)/(1+exp(feta))
y=rbinom(n, 1, fprob)

nsis=floor(n/log(n)/4)
xtune=matrix(rnorm(n*p, mean=0, sd=1), n, p)
xtune=xtune%*%cholmat
feta=xtune[, 1:4]%*%b
fprob=exp(feta)/(1+exp(feta))
ytune=rbinom(n, 1, fprob)

ISIScoef = INDEPgetfinalSCADcoef(x = x, y = y, 
                pickind = 1:4, xtune = xtune, ytune = ytune, 
                family = binomial())

[Package SIS version 0.2 Index]