DCSBM: DCSBM

In jdwilson4/NetSurv: Network Surveillance via the Degree Corrected Stochastic Block Model

Description

Simulate an undirected graph realization from the degree corrected stochastic block random graph model. Edge weights are discrete valued and are generated independently where e_ij ~ Poisson(theta_i*theta_j*P_c_i, c_j)

Usage

DCSBM(n, k = 2, P, sizes = c(round(n/2), n - round(n/2)),
  random.community.assignment = c(FALSE, TRUE), community.labels = NULL,
  delta = rep(0, k), edge.list = c(FALSE, TRUE))

Arguments

n:	number of nodes in the graph
k:	number of communities in the graph
P:	the k x k matrix of probabilities whose i,jth entry specifies the probability of connection between nodes in community i and community j
sizes:	a numeric vector of length k whose ith entry specifies the size of the ith community. The entries must add to n.
random.community.assignment:	a logical that specifies whether or not community labels are determined at random. Default is FALSE.
community.labels:	an integer vector of length n whose ith entry is the community label of the ith vertex. Default is NULL. If provided, community labels are no longer assigned.
delta:	a numeric vector of length k whose values must be between 0 and 1. Theta parameters for community r are generated as an iid sample from a U(0 + delta, 1 - delta) distribution
edge.list:	a logical that specifies whether or not the adjacency matrix should be returned as an edge list.

Value

a list containing the objects

• Adjacency: the adjacency matrix (or edge list if edge.list == TRUE) of the generated network

• Thetas: an n x 1 vector with values of each theta

• Membership: an n x 1 vector specifying community membership of each node

Author(s)

James D. Wilson and Nathaniel T. Stevens

References

• Wilson, James D., Stevens, Nathaniel T., and Woodall, William H. (2016). <e2><80><9c>Modeling and estimating change in temporal networks via a dynamic degree corrected stochastic block model.<e2><80><9d> arXiv Preprint: http://arxiv.org/abs/1605.04049

Examples

net <- DCSBM(n = 500, k = 2, P = cbind(c(0.10, 0.01), c(0.02, 0.075)),
             sizes = c(200, 300), random.community.assignment = FALSE,
             delta = c(0.2, 0.7), edge.list = FALSE)
image(Matrix(net$Adjacency))

jdwilson4/NetSurv documentation built on May 18, 2019, 11:40 p.m.