Block-diagonal models using ergm

These two functions are used to go back and forth from a pooled ergm vs a blockdiagonal model, the latter to be fitted using ergm::ergm.

blockdiagonalize(x, attrname = "block")

splitnetwork(x, attrname)

ergm_blockdiag(formula, ...)

Arguments

x	In the case of blockdiagonalize, a list of networks or matrices. For splitnetwork a single network object with a vertex attribute that can be used to split the data.
attrname	Name of the attribute that holds the block ids.
formula	An ergm model which networks' will be wrapped with blockdiagonalize (see details).
...	Further arguments passed to the method.

Value

An object of class ergm::ergm.

Details

The function ergm_blockdiag is a wrapper function that takes the model's network, stacks the networks into a single block diagonal net, and calls ergm::ergm with the option constraints = blockdiag("block").

One side effect of this function is that it loads the ergm package via requireNamespace, so after executing the function ergm the package will be loaded.

Examples

library(ergm)
#> Loading required package: network
#> network: Classes for Relational Data
#> Version 1.16.0 created on 2019-11-30.
#> copyright (c) 2005, Carter T. Butts, University of California-Irvine
#>                     Mark S. Handcock, University of California -- Los Angeles
#>                     David R. Hunter, Penn State University
#>                     Martina Morris, University of Washington
#>                     Skye Bender-deMoll, University of Washington
#>  For citation information, type citation("network").
#>  Type help("network-package") to get started.
#> 
#> ergm: version 3.10.4, created on 2019-06-10
#> Copyright (c) 2019, Mark S. Handcock, University of California -- Los Angeles
#>                     David R. Hunter, Penn State University
#>                     Carter T. Butts, University of California -- Irvine
#>                     Steven M. Goodreau, University of Washington
#>                     Pavel N. Krivitsky, University of Wollongong
#>                     Martina Morris, University of Washington
#>                     with contributions from
#>                     Li Wang
#>                     Kirk Li, University of Washington
#>                     Skye Bender-deMoll, University of Washington
#>                     Chad Klumb
#> Based on "statnet" project software (statnet.org).
#> For license and citation information see statnet.org/attribution
#> or type citation("ergm").
#> NOTE: Versions before 3.6.1 had a bug in the implementation of the bd()
#> constriant which distorted the sampled distribution somewhat. In
#> addition, Sampson's Monks datasets had mislabeled vertices. See the
#> NEWS and the documentation for more details.
#> NOTE: Some common term arguments pertaining to vertex attribute and
#> level selection have changed in 3.10.0. See terms help for more
#> details. Use ‘options(ergm.term=list(version="3.9.4"))’ to use old
#> behavior.
data(fivenets)

fivenets2 <- blockdiagonalize(fivenets, attrname = "block") # A network with
ans0 <- ergm(
  fivenets2 ~ edges + nodematch("female"),
  constraints = ~blockdiag("block")
  )
#> Warning: `set_attrs()` is deprecated as of rlang 0.3.0
#> This warning is displayed once per session.
#> Starting maximum pseudolikelihood estimation (MPLE):
#> Evaluating the predictor and response matrix.
#> Maximizing the pseudolikelihood.
#> Finished MPLE.
#> Stopping at the initial estimate.
#> Evaluating log-likelihood at the estimate. 
#> 
ans1 <- ergmito(fivenets ~ edges + nodematch("female"))
#> Warning: The observed statistics (target.statistics) are near or at the boundary of its support, i.e. the Maximum Likelihood Estimates maynot exist or be hard to be estimated. In particular, the statistic(s) "edges", "nodematch.female".

# This is equivalent
ans2 <- ergm_blockdiag(fivenets ~ edges + nodematch("female"))
#> Starting maximum pseudolikelihood estimation (MPLE):
#> Evaluating the predictor and response matrix.
#> Maximizing the pseudolikelihood.
#> Finished MPLE.
#> Stopping at the initial estimate.
#> Evaluating log-likelihood at the estimate. 
#>