wrapper_weighted_stat_undir: An estimation framework for weighted static binary undirected...

In ergmclust: Exponential-Family Random Graph Models for Network Clustering

An estimation framework for weighted static binary undirected networks.

Description

The inner wrapper function wrapper_weighted_stat_undir() is an estimation framework for weighted binary static undirected network based on variational Expectation-Maximization algorithm.

Usage

wrapper_weighted_stat_undir(adjmat = NULL, wtmat = NULL, K, 
thresh = 10^(-6), iter.max, coef.init)

Arguments

adjmat	Array of the network adjacency matrices.
wtmat	Array of the network weight matrices.
K	Number of clusters for which the estimation must be performed.
thresh	Convergence threshold for estimation, the default being 1e-6.
iter.max	The maximum number of iterations after which the algorithm is terminated. The default value is set as 200.
coef.init	The initial value of the density, stability, transitivity parameters for which clustering is desired, the default being 1e-1.

Value

Returns a list of ergmclust object. Each object of class ergmclust is a list with the following components:

Converged Parameters	The first elements of the list in sequential order are gamma, alpha, pi, tau, theta for dynamic networks. For static networks, there is no pi or tau.
Estimated Cluster IDs	The second element of the list is a vector (matrix) of estimated cluster memberships for a static (dynamic) case. This is absent for all K=1 cases.
ICL Values	The third element of the list is the Integrated Classification Likelihood value that can be used for model selection, i.e. selecting the appropriate number of clusters.

ergmclust documentation built on March 16, 2026, 5:06 p.m.