grouplasso_linreg_slower: Minimize the objective function of the group lasso problem...

In gregorkb/semipadd2pop: Combine two data sets to fit semiparametric additive models

Minimize the objective function of the group lasso problem with a continuous response

Description

Minimize the objective function of the group lasso problem with a continuous response

Usage

grouplasso_linreg_slower(
  rY,
  rX,
  groups,
  lambda,
  w,
  tol,
  maxiter,
  beta_init = as.numeric(c())
)

Arguments

groups	a vector of integers indicating to which group each covariate belongs
lambda	the level of sparsity penalization
w	vector of group-specific weights for different penalization of groups
tol	a convergence criterion
beta_init	optional starting value for beta
Y	the response vector
X	matrix containing the design matrices
eigen	a list of eigen info on groups
max.iter	the maximum allowed number of iterations
return_obj	a logical indicating whether the value of the objection function should be recorded after every step of the algorithm

Value

Returns the minimizer of the group lasso objective function

Examples

data <- get_grouplasso_data(n = 500,response = "continuous")

grouplasso_linreg.out <- grouplasso_linreg(rY = data$Y,
                                           rX = data$X,
                                           groups = data$groups,
                                           lambda = 10,
                                           w = data$w,
                                           tol = 1e-4,
                                           maxiter = 500)

gregorkb/semipadd2pop documentation built on Oct. 2, 2022, 1:37 p.m.