MLGL: Multi-Layer Group-Lasso

In MLGL: Multi-Layer Group-Lasso

Multi-Layer Group-Lasso

Description

Run hierarchical clustering following by a group-lasso on all the different partitions.

Usage

MLGL(X, ...)

## Default S3 method:
MLGL(
  X,
  y,
  hc = NULL,
  lambda = NULL,
  weightLevel = NULL,
  weightSizeGroup = NULL,
  intercept = TRUE,
  loss = c("ls", "logit"),
  sizeMaxGroup = NULL,
  verbose = FALSE,
  ...
)

## S3 method for class 'formula'
MLGL(
  formula,
  data,
  hc = NULL,
  lambda = NULL,
  weightLevel = NULL,
  weightSizeGroup = NULL,
  intercept = TRUE,
  loss = c("ls", "logit"),
  verbose = FALSE,
  ...
)

Arguments

X	matrix of size n*p
...	Others parameters for gglasso function
y	vector of size n. If loss = "logit", elements of y must be in -1,1
hc	output of hclust function. If not provided, hclust is run with ward.D2 method. User can also provide the desired method: "single", "complete", "average", "mcquitty", "ward.D", "ward.D2", "centroid", "median".
lambda	lambda values for group lasso. If not provided, the function generates its own values of lambda
weightLevel	a vector of size p for each level of the hierarchy. A zero indicates that the level will be ignored. If not provided, use 1/(height between 2 successive levels). Only if hc is provided
weightSizeGroup	a vector of size 2*p-1 containing the weight for each group. Default is the square root of the size of each group. Only if hc is provided
intercept	should an intercept be included in the model ?
loss	a character string specifying the loss function to use, valid options are: "ls" least squares loss (regression) and "logit" logistic loss (classification)
sizeMaxGroup	maximum size of selected groups. If NULL, no restriction
verbose	print some information
formula	an object of class "formula" (or one that can be coerced to that class): a symbolic description of the model to be fitted.
data	an optional data.frame, list or environment (or object coercible by as.data.frame to a data.frame) containing the variables in the model. If not found in data, the variables are taken from environment (formula)

Value

a MLGL object containing:

lambda

lambda values

b0

intercept values for lambda

beta

A list containing the values of estimated coefficients for each values of lambda

var

A list containing the index of selected variables for each values of lambda

group

A list containing the values index of selected groups for each values of lambda

nVar

A vector containing the number of non zero coefficients for each values of lambda

nGroup

A vector containing the number of non zero groups for each values of lambda

structure

A list containing 3 vectors. var: all variables used. group: associated groups. weight: weight associated with the different groups. level: for each group, the corresponding level of the hierarchy where it appears and disappears. 3 indicates the level with a partition of 3 groups.

time

computation time

dim

dimension of X

hc

Output of hierarchical clustering

call

Code executed by user

Author(s)

Quentin Grimonprez

See Also

cv.MLGL, stability.MLGL, listToMatrix, predict.MLGL, coef.MLGL, plot.cv.MLGL

Examples

set.seed(42)
# Simulate gaussian data with block-diagonal variance matrix containing 12 blocks of size 5
X <- simuBlockGaussian(50, 12, 5, 0.7)
# Generate a response variable
y <- X[, c(2, 7, 12)] %*% c(2, 2, -2) + rnorm(50, 0, 0.5)
# Apply MLGL method
res <- MLGL(X, y)

MLGL documentation built on March 31, 2023, 9:32 p.m.