imp.lasso.tuning: Tuning of LASSO coefficients bounds

Description Usage Arguments Value

View source: R/imprecise_lasso.R

Description

Tuning of the model over lambda using lower and upper estimates of LASSO coefficients using optimization over weights

Usage

1
2
imp.lasso.tuning(x, y, wtl, wtu, ts = NULL, method = lasso_optim_cd,
  n_it = 10)

Arguments

x

predictors

y

response

wtl

lower bound of the weights for the coefficients of weighted LASSO

wtu

upper bound of the weights for the coefficients of weighted LASSO

ts

stepsize for proximal gradient and sub-gradient method (use opt_ts() to generate stepsize). Defaults to NULL

method

lasso optimization function. Three different methods are available to use. method = c(lasso_optim_cd, lasso_optim_sg, lasso_optim_pg). Defaults to lasso_optim_cd

n_it

number of iteration for lasso_optim_cd method. Default value is 10.

lambdas

values of the penalty parameter

Value

The function returns the lower and upper estimates of LASSO coefficients.


tathagatabasu/bootlasso documentation built on Aug. 9, 2019, 1:07 a.m.