main_noisySBM_GGM {noisysbmGGM}

GGM Inference from Noisy Data by Multiple Testing using SILGGM and Drton test statistics

Description

The main_noisySBM_GGM() function is a key feature of the noisysbmGGM package, dedicated to Gaussian Graphical Model (GGM) inference. This function takes an $n$-sample of a Gaussian vector of dimension $p$ and provides the GGM associated with the partial correlation structure of the vector. GGM inference is essential in capturing the underlying relationships between the vector's coefficients, helping users uncover meaningful interactions while controlling the number of false discoveries.

Usage

main_noisySBM_GGM(
  X,
  Meth = "Ren",
  NIG = NULL,
  threshold = 0.5,
  Nbrepet = 2,
  rho = NULL,
  tau = NULL,
  a = NULL,
  b = NULL,
  c = NULL,
  d = NULL,
  n0 = 1,
  eta0 = 1,
  zeta0 = 1,
  alpha = 0.1,
  Qup = NULL,
  nbCores = parallel::detectCores(),
  nbOfZ = 12,
  sigma0 = 1,
  sigma1 = 1,
  percentageOfPerturbation = 0.3,
  verbatim = TRUE
)

Arguments

X

A n by p matrix containing a n-sample of a p-vector

Meth

Choice of test statistics between "Ren", "Jankova_NW", "Jankova_GL", "Liu_SL", "Liu_L", and "zTransform" (warning "zTransform" only work if n>p)

NIG

A Boolean (automatically chosen according to the selected method : NIG=FALSE except for "Liu_SL" and "Liu_L" test statistics as input). If FALSE, the variance under the alternative hypothesis in assumed to be known. If TRUE, the variances under the alternatives are unknown and estimated with the NIG method.

threshold

Threshold use when updating the latent graphs structure from l-values (by default threshold=0.5)

Nbrepet

Number of times the algorithm is repeated (by default Nbrepet=2)

rho

Hyperparameter of the non-NIG method (by default rho=1)

tau

Hyperparameter of the non-NIG method (by default tau=1)

a

Hyperparameter of the NIG method (by default a=0)

b

Hyperparameter of the NIG method (by default b=1)

c

Hyperparameter of the NIG method (by default c=1)

d

Hyperparameter of the NIG method (by default d=1)

n0

Hyperparameter (by default n0=1)

eta0

Hyperparameter (by default eta0=1)

zeta0

Hyperparameter (by default zeta0=1)

alpha

Level of significance of the multiple testing procedure (by default alpha=0.1)

Qup

Maximal number of cluster (by default Qup =10)

nbCores

Nb of cores to be used during calculations (by default nbCores=parallel::detectCores())

nbOfZ

Nb of initialization (by default nbOfZ=12)

sigma0

standard deviation under the null hypothesis (by default sigma0=1)

sigma1

standard deviation under the alternative hypothesis in the non-NIG method (by default sigma1=1)

percentageOfPerturbation

perturbation during initialization (by default percentageOfPerturbation=0.3)

verbatim

print information messages

Value

A

the adjacency matrix of the inferred graph

Z

the inferred clustering

theta

the parameters of the noisySBM at the end

Q

the number of clusters at the end

#' @examples main_noisySBM_GGM(GGMtest$dataMatrix,Meth="Ren",NIG=TRUE,Qup=10,nbOfZ=1)

See Also

main_noisySBM