HMT: Hierarchical Multiple Testing procedure

In MLGL: Multi-Layer Group-Lasso

Hierarchical Multiple Testing procedure

Description

Apply Hierarchical Multiple Testing procedure on a MLGL object

Usage

HMT(
  res,
  X,
  y,
  control = c("FWER", "FDR"),
  alpha = 0.05,
  test = partialFtest,
  addRoot = FALSE,
  Shaffer = FALSE,
  ...
)

Arguments

res	MLGL object
X	matrix of size n*p
y	vector of size n.
control	either "FDR" or "FWER"
alpha	control level for testing procedure
test	test used in the testing procedure. Default is partialFtest
addRoot	If TRUE, add a common root containing all the groups
Shaffer	If TRUE, a Shaffer correction is performed (only if control = "FWER")
...	extra parameters for selFDR

Value

a list containing:

lambdaOpt

lambda values maximizing the number of rejects

var

A vector containing the index of selected variables for the first lambdaOpt value

group

A vector containing the values index of selected groups for the first lambdaOpt value

selectedGroups

Selected groups for the first lambdaOpt value

indLambdaOpt

indices associated with optimal lambdas

reject

Selected groups for all lambda values

alpha

Control level

test

Test used in the testing procedure

control

"FDR" or "FWER"

time

Elapsed time

hierTest

list containing the output of the testing function for each lambda. Each element can be used with the selFWER or selFDR functions.

lambda

lambda path

nGroup

Number of groups before testing

nSelectedGroup

Numer of groups after testing

See Also

hierarchicalFWER hierarchicalFDR selFWER selFDR

Examples

set.seed(42)
X <- simuBlockGaussian(50, 12, 5, 0.7)
y <- X[, c(2, 7, 12)] %*% c(2, 2, -2) + rnorm(50, 0, 0.5)
res <- MLGL(X, y)

# perform hierarchical testing with FWER control
out <- HMT(res, X, y, alpha = 0.05)

# test a new value of alpha for a specific lambda
selFWER(out$hierTest[[60]], alpha = 0.1)

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