boot_ProbitLasso: Title Bootstrap validation for probit regression with Lasso...

In EOgundimu300/HeckmanSelect: Regularization method for binary Heckman selection model

Description

Title Bootstrap validation for probit regression with Lasso penalty

Usage

boot_ProbitLasso(object, data, mboot = 50, seed, ...)

Arguments

object	a fitted object of class inheriting from "ProbitLasso"
data	data matrix containing both the outcome and selection variables
mboot	number of bootstrap samples. 50 bootstrap samples are used if not specified
seed	integer value set for reproducibility
...

Value

coefficients: same as in coef

nonconvergence: the number of samples that did not converge in the bootstrap sample

lambda: optimal shrinkage parameter

resu: data matrix containing apparent, bootstrap, test and optimism corrected performance measures

Vectors of performance measures for each bootstrap sample are also returned

Examples

pp2 <- ProbitLasso(formula, data=data, allowParallel = TRUE, penalty="ALASSO", crit="bic")
boot_ProbitLasso(pp2, data=data, mboot=2, seed=1)

EOgundimu300/HeckmanSelect documentation built on Feb. 5, 2022, 2:48 a.m.