glmnet.constr: Constrained lasso fitting.

In stephenbates19/logratiolasso: Tools For Large Log Ratio Models

Description

Run the constrained lasso procedure.

Usage

glmnet.constr(x, y, family = c("gaussian", "binomial"), alpha = 1,
  nlambda = 100, lambda.min.ratio = 0.01, lambda = NULL, ...)

Arguments

x	The matrix of features.
y	The response value. In the real-valued repsonse case, y should be centered.
family	The model family. Default is "gaussian" for standard lasso. "binomial" is available for lasso logistic regression.
alpha	The elastic net parameter. Default is 1 (the standard lasso).
nlambda	Numer of lambda values to fit
lambda.min.ratio	Smallest value for lambda as a fraction of lambda.max
lambda	Optional grid of lambda values
...	Additional arguments to internally pass to the "glmnet" function.

Value

beta The resulting coefficient vectors in matrix form.

stephenbates19/logratiolasso documentation built on May 18, 2019, 4:52 p.m.