main_noisySBM {noisysbmGGM}

Graph Inference from Noisy Data by Multiple Testing

Description

The main_noisySBM() function is a core component of the noisysbmGGM package, responsible for applying the greedy algorithm to estimate model parameters, perform node clustering, and conduct a multiple testing procedure to infer the underlying graph. This function is versatile, offering various options and providing useful outputs for further analysis

Usage

main_noisySBM(
  X,
  NIG = FALSE,
  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 p-square matrix containing the data

NIG

A Boolean. If FALSE (by default), 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(NSBMtest$dataMatrix,NIG=TRUE,Qup=10,nbOfZ=1,nbCores=1)