Generates networks from binary tree stochastic block model, with provided sequence of connection probability along the tree
BTSBM(n, d, a.seq, lambda, alpha = NULL, N = 1)A list of objections of
the generated network adjacency matrices
the connection probability matrix between K communities, where K = 2^d
the vector of community labels for n nodes
the connection probability matrix between the n nodes. It is the expectation of adjacency matrices, except on the diagonal
the binary string similarity matrix between communities
the binary string similarity matrix between nodes
number of nodes in the network
number of layers until leaves (excluding the root)
the connection probability sequence along the tree, a_r, see details in the paper
average node degree, only used when alpha is not provided
the common scaling of the a_r sequence. So at the end, essentially the a_r sequence is a.seq*alpha
the number of networks to generate from the same model
Tianxi Li, Lihua Lei, Sharmodeep Bhattacharyya, Purnamrita Sarkar, Peter Bickel, and Elizaveta Levina.
Maintainer: Tianxi Li <tianxili@umn.edu>
Li, T., Lei, L., Bhattacharyya, S., Van den Berge, K., Sarkar, P., Bickel, P.J. and Levina, E., 2022. Hierarchical community detection by recursive partitioning. Journal of the American Statistical Association, 117(538), pp.951-968.
dt <- BTSBM(n=1600,d=4,a.seq=0.2^seq(0,4),lambda=50)
A <- dt$A.list[[1]]
Run the code above in your browser using DataLab