glbin_lcd: Block coordinate gradient descent for logistic group lasso

In antiphon/glbinc:

Description

Like in Breheny and Huang 2009

Usage

glbin_lcd(X, y, offset, index, eps = 1e-04, lambda, lambda.min = 0.01,
  nlambda = 50, dfmax = ncol(X) + 1, verb = 1, std = TRUE, alpha = 1,
  maxit = 1000, stability_eps = 1e-05)

Arguments

X	covariate/design matrix. intercept is taken to be the first column!
y	response 0/1 vector
offset	offset terms
index	grouping. set 0 or NA for unpenalised terms. if not given, assumes X[,1] is intercept
eps	convergence criterion
lambda	vector of lambdas
lambda.min	fraction of max lambda to go down to
nlambda	number of penalties
dfmax	max df, stop if reached
verb	verbosity

antiphon/glbinc documentation built on July 31, 2019, 11 p.m.