post_lasso_cv: This function uses the 'glmnet' package to estimate the...

In MCKnaus/dmlmt: Double Machine Learning for Multiple Treatments

Description

This function uses the glmnet package to estimate the coefficient paths and cross-validates Lasso AND Post-Lasso

Usage

post_lasso_cv(x, y, w = NULL, kf = 10, family = "gaussian", seed = NULL,
  output = TRUE, d = NULL, parallel = FALSE, graph_file = NULL,
  se_rule = NULL, ...)

Arguments

x	Matrix of covariates (number of observations times number of covariates matrix)
y	vector of outcomes
w	vector of weights
kf	number of folds in k-fold CV
family	Outcome family, currenlty "gaussian" and "binomial" supported
seed	CV samples are drawn starting with a specified seed
output	If TRUE, output and graphs are printed
d	vector with binary treatment indicator, if not NULL, weights are normalized to sum to N in treated and control also in CV samples
parallel	If TRUE, cross-validation parallelized
graph_file	Option to define path to save graph
se_rule	If not NULL, define, e.g., c(-1,1) to get 1SE and 1SE+ rule
...	Pass glmnet options

Value

List with the names of selected variables at cross-validated minima for Lasso and Post-Lasso

MCKnaus/dmlmt documentation built on Dec. 4, 2020, 9:48 a.m.