elasticnet: Wrapper to use the glmnet package that Elastic Net model can...

In ToshihiroIguchi/elasticnet: Wrapper to use the glmnet package that Elastic Net model can be analyzed by the formula format.

Description

The glmnet package that Elastic Net can use is nice, but you can not use the formula and you need to type it in matrix. So we created an elasticnet package that Elastic Net can use in formula. This is a wrapper for the glmnet package, which requires the glmnet package to work. Also, the elasticnet package can automatically tune the best alpha and lambda.

Usage

elasticnet(formula, data, family = NULL,
offset = NULL, nfolds = 5, lambda = "lambda.1se",
alpha = NULL, alpha_step = 0.05, seed = NULL)

Arguments

formula	an object of class "formula" : a symbolic description of the model to be fitted.
data	an optional data frame containing the variables in the model.
family	Response type. Default is NULL. If NULL, family is automatically selected.
lambda	"lambda.1se" or "lambda.min".
alpha	The elasticnet mixing parameter, 0 or more and 1 or less. alpha=1 is the lasso penalty, and alpha=0 the ridge penalty.
alpha_step	Step in searching for parameter alpha.
seed	Random number generation.

Author(s)

Toshihiro Iguchi

Examples

result <- elasticnet(Sepal.Length ~., iris)
plot(result)
summary(result)

ToshihiroIguchi/elasticnet documentation built on May 20, 2019, 2:20 p.m.