Is there a way to make a configuration model with the statnet ecosystem?

Question

I can make a configuration model with igraph like this:

set.seed(124) 
k_i <- rpois(10, 3)
## [1] 1 2 3 2 2 2 3 3 6 2
g <- igraph::sample_degseq(k_i, method = "vl")
plot(g)

I can recover the correct degree histogram for this network with statnet tools:

library(statnet)
G <- intergraph::asNetwork(g)
summary(G ~ degree(min(k_i):max(k_i)))

But going the other way results in several warnings and errors:

m <- ergm(G ~ degree(min(k_i):max(k_i)))
## e.g.
## In mple.existence(pl) : The MPLE does not exist!

Is there a function or series of steps in one of the statnet packages that produces a similar result to igraph's sample_degseq?

Answer 1

Is the goal to sample networks? I assume Yes. If you have already produced a graph that realizes a given degree sequence you can construct an ERGM distribution in order to sample from it. The key is to use sample space constraints, e.g. (disregard the warnings):

1. Sample networks with identical degree sequence

m <- ergm(G ~ edges, constraints = ~ degrees)
nets <- simulate(m, nsim = 4)
sapply(nets, sna::degree) / 2
#>       [,1] [,2] [,3] [,4]
#>  [1,]    1    1    1    1
#>  [2,]    2    2    2    2
#>  [3,]    3    3    3    3
#>  [4,]    2    2    2    2
#>  [5,]    2    2    2    2
#>  [6,]    2    2    2    2
#>  [7,]    3    3    3    3
#>  [8,]    3    3    3    3
#>  [9,]    6    6    6    6
#> [10,]    2    2    2    2

2. Sample networks with identical degree distribution

m <- ergm(G ~ edges, constraints = ~ degreedist)
nets <- simulate(m, nsim = 4)
sapply(nets, sna::degree) / 2
#>       [,1] [,2] [,3] [,4]
#>  [1,]    1    3    2    3
#>  [2,]    2    2    3    1
#>  [3,]    2    2    2    3
#>  [4,]    3    2    3    2
#>  [5,]    3    1    2    2
#>  [6,]    2    2    2    3
#>  [7,]    3    2    2    2
#>  [8,]    2    3    1    2
#>  [9,]    6    6    6    6
#> [10,]    2    3    3    2