gSAM: Group-Sparse Additive Machine A group version of Sparse...

In sambiostat/WAPL: What the Package Does (One Line, Title Case)

Description

Group-Sparse Additive Machine A group version of Sparse Additive Machine @param X, n by p Covariate matrix @param y, Treatment assigned, length n vector @param w, instance weight @param p: integer, total number of basis @param lambda: A user supplied lambda sequence(ordered in decreasing value). Normally set it to null so the computing algorithm calculate it automatically. @param nlambda: number of lambda, default is 50 @param lambda.min.ratio: the ration between max lambda and minimal lambda @param group: group information, should be consective, default is NULL i.e. no group information presents @param thol: Stopping precision. The default value is 1e-5. @param mu: Smoothing parameter used in approximate the Hinge Loss. The default value is 0.05. @param max.ite: The number of maximum iterations. The default value is 1e5 @import gglasso @import glmnet @import SAM @export @return fitted ITR, an object of "gSAM" class @examples train.data <- gSim(N=200, sigma=0, scenario=1) H <- train.data[[1]] A <- train.data[[2]] R2 <- train.data[[3]] group=rep(1:20, each=3) tst = gSAM(X=H, y=A, w=R2, p=3, prop=rep(1,200), pentype = "lasso",lambda.min.ratio=0.2, m=7, group= group, plist=c(3:5))

Usage

gSAM(
  X,
  y,
  w = NULL,
  p = 3,
  lambda = NULL,
  nlambda = 50,
  lambda.min.ratio = 0.2,
  group = NULL,
  thol = 1e-05,
  mu = 0.05,
  max.ite = 1e+05
)

sambiostat/WAPL documentation built on May 26, 2020, 12:17 a.m.