stat.mboost_diff_Rsq: Importance statistics based on R-square via boosting

In hanfu-bios/varsel: Model-Free Knockoffs Filter for Controlled Variable Selection

Description

This function provides importance scores for variables (including the knockoffs) in order to compute statistics. The variable importance is measured by R-squares obtained from boosting.

Usage

stat.mboost_diff_Rsq(Xaug, y, max.mstop = 100, bl = c("bbs", "bols", "btree"), cv.fold = 5, family = Gaussian())

Arguments

Xaug	augmented design matrix combining original predictors and knockoff variables
y	response vector, or a survival object with two columns
max.mstop	maximum number of boosting iteration
bl	base-learners when fitting models using mboost. 'bols' means linear base-learners, 'bbs' penalized regression splines with a B-spline basis, and 'btree' boosts stumps.
cv.fold	number of folds in cross-validation to choose number of iteration
family	Binomial(), Binomial(link = <e2><80><9c>logit<e2><80><9d>, type=<e2><80><9d>glm<e2><80><9d>), Gaussian(), Poisson(), CoxPH(), Cindex(), GammaReg(), NBinomial(), Weibull(), Loglog(), Lognormal(), etc. See mboost documentation for details.

Value

2p vector containing varible importance for both orginal variables and knockoff variables

hanfu-bios/varsel documentation built on May 27, 2019, 4:50 a.m.