extended.randomlasso: An implementation of random lasso regression.

In snelson89/extendedglmnetGroup7: Extension to glmnet package

An implementation of random lasso regression.

Description

This function takes a matrix of candidate predictors X and a vector of response variables Y and performs random lasso regression.

Usage

extended.randomlasso(
  X,
  Y,
  lambda = NULL,
  B = 200,
  q1 = length(Y),
  q2 = length(Y),
  ytype = "continuous"
)

Arguments

X	an n by p matrix of candidate predictors.
Y	an n by 1 vector of responses
lambda	an optional lambda value. If no lambda is provided, an optimal lambda is automatically determined using glmnet::cv.glmnet(alpha=1).
B	the number of Bootstrap samples. Default is 200.
q1	the number of random predictors to use in step 1 of the algorithm. Default is length(Y). Must be less than or equal to this value.
q2	the number of random predictors to use in step 2 of the algorithm. Default is length(Y). Must be less than or equal to this value.
ytype	description as to whether or not the response variable y is binary or continuous. Defaults to 'continuous'.

Examples

extended.randomlasso(X,Y)

snelson89/extendedglmnetGroup7 documentation built on May 12, 2022, 7:38 p.m.