missSBM (version 0.3.0)

missSBM_fit: An R6 class to represent an SBM fit with missing data

Description

The function estimateMissSBM() fits a collection of SBM for varying number of block. Each fitted SBM is an instance of an R6 object with class missSBM_fit, described here.

Fields are accessed via active binding and cannot be changed by the user.

This class comes with a set of R6 methods, some of them being useful for the user and exported as S3 methods. See the documentation for show(), print(), fitted(), predict(), plot().

Active bindings

fittedSBM

the fitted SBM with class SimpleSBM_fit_missSBM, inheriting from class sbm::SimpleSBM_fit

fittedSampling

the fitted sampling, inheriting from class networkSampling and corresponding fits

imputedNetwork

The network data as a matrix with NAs values imputed with the current model

monitoring

a list carrying information about the optimization process

entropyImputed

the entropy of the distribution of the imputed dyads

entropy

the entropy due to the distribution of the imputed dyads and of the clustering

vExpec

double: variational expectation of the complete log-likelihood

penalty

double, value of the penalty term in ICL

loglik

double: approximation of the log-likelihood (variational lower bound) reached

ICL

double: value of the integrated classification log-likelihood

Methods

Public methods

• missSBM_fit$new()

• missSBM_fit$doVEM()

• missSBM_fit$show()

• missSBM_fit$print()

• missSBM_fit$clone()

Method new()

constructor for networkSampling

Usage

missSBM_fit$new(partlyObservedNet, nbBlocks, netSampling, clusterInit, useCov)

Arguments

partlyObservedNet

An object with class partlyObservedNetwork.

nbBlocks

integer, the number of blocks in the SBM

netSampling

The sampling design for the modelling of missing data: MAR designs ("dyad", "node") and NMAR designs ("double-standard", "block-dyad", "block-node" ,"degree")

clusterInit

Initial clustering: either a character in "hierarchical", "spectral" or "kmeans", or a vector with size ncol(adjacencyMatrix), providing a user-defined clustering with nbBlocks levels. Default is "hierarchical".

useCov

logical. If covariates are present in partlyObservedNet, should they be used for the inference or of the network sampling design, or just for the SBM inference? default is TRUE.

Method doVEM()

a method to perform inference of the current missSBM fit with variational EM

Usage

missSBM_fit$doVEM(
  control = list(threshold = 0.001, maxIter = 100, fixPointIter = 5, trace = 1)
)

Arguments

control

a list of parameters controlling the variational EM algorithm. See details of function estimateMissSBM()

Method show()

show method for missSBM_fit

Usage

missSBM_fit$show()

Method print()

User friendly print method

Usage

missSBM_fit$print()

Method clone()

The objects of this class are cloneable with this method.

Usage

missSBM_fit$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Examples

# NOT RUN {
## Sample 75% of dyads in  French political Blogosphere's network data
adjMatrix <- missSBM::frenchblog2007 %>%
  igraph::as_adj (sparse = FALSE) %>%
  missSBM::observeNetwork(sampling = "dyad", parameters = 0.25)
collection <- estimateMissSBM(adjMatrix, 3:5, sampling = "dyad")
my_missSBM_fit <- collection$bestModel
class(my_missSBM_fit)
plot(my_missSBM_fit, "connectivity")

# }