SBM_sampler: An R6 Class to represent a sampler for a SBM

Description

The function simulate produces an instance of an object with class SBM_sampler.

Usage

SBM_sampler

Arguments

Format

An object of class R6ClassGenerator of length 24.

Fields

nNodes: The number of nodes

nBlocks

The number of blocks

nCovariates

The number of covariates

nDyads

The number of possible dyad in the network (depends on the direction)

direction

A character indicating if the network is directed or undirected

hasCovariates

a boolean indicating if the model has covariates

mixtureParam

the vector of mixture parameters

connectParam

the matrix of connectivity: inter/intra probabilities of connection when the network does not have covariates, or a logit scaled version of it.

covarParam

the vector of parameters associated with the covariates

covarArray

the array of covariates

Details

All fields of this class are only accessible for reading. This class comes with a set of methods, some of them being useful for the user (see examples)

R6 methods:$rBlocks(), $rAdjancencyMatrix()
S3 methodsprint(), plot()

Examples

Run this code

# NOT RUN {
## SBM parameters
directed <- FALSE
N <- 300 # number of nodes
Q <- 3   # number of clusters
alpha <- rep(1,Q)/Q     # mixture parameter
pi <- diag(.45,Q) + .05 # connectivity matrix
gamma <- log(pi/(1-pi)) # logit transform fo the model with covariates
M <- 2 # two Gaussian covariates
covarMatrix <- matrix(rnorm(N*M,mean = 0, sd = 1), N, M)
covarParam  <- rnorm(M, -1, 1)

## draw a SBM without covariates through simulateSBM
sbm <- missSBM::simulate(N, alpha, pi, directed)

## equivalent construction from the SBM_sampler class itslef
sbm_s <- SBM_sampler$new(directed, N, alpha, pi)
sbm_s$rBlocks() # draw some blocks
sbm_s$rAdjMatrix() # draw some edges

# }

Run the code above in your browser using DataCamp Workspace