Estimate a within-block network model. — est

Function to estimate the within-block model. Both pseudo-maximum likelihood and monte carlo approximate maximum likelihood estimators are implemented.

Usage

est_within(
  formula,
  network,
  seed = NULL,
  method = "MPLE",
  add_intercepts = TRUE,
  clustering_with_features = FALSE,
  return_network = FALSE,
  ...
)

Arguments

formula: An R formula object of the form y ~ <model terms>, where y is a network object. The network object must contain block information as a vertex attribute with the name 'block'. For the details on the possible <model terms>, see ergmTerm and Morris, Handcock and Hunter (2008). The L-ergmTerm is supported to enable size-dependent coefficients.
network: a network object with one vertex attribute called 'block' representing which node belongs to which block
seed: seed value (integer) for the random number generator
method: If "MPLE" (the default), then the maximum pseudolikelihood estimator is returned. If "MLE", then an approximate maximum likelihood estimator is returned.
add_intercepts: Boolean value to indicate whether adequate intercepts should be added to the provided formula so that the model in the first stage of the estimation is a nested model of the estimated model in the second stage of the estimation
clustering_with_features: Boolean value to indicate if the clustering was carried out making use of the covariates or not (only important if add_intercepts = TRUE)
return_network: Boolean value to indicate if the network object should be returned in the output. This is needed if the user wants to use, e.g., the gof function as opposed to the gof.bigergm function.
...: Additional arguments, to be passed to the ergm function

Value

'ergm' object of the estimated model.

References

Morris M, Handcock MS, Hunter DR (2008). Specification of Exponential-Family Random Graph Models: Terms and Computational Aspects. Journal of Statistical Software, 24.

Examples

adj <- c(
c(0, 1, 0, 0, 1, 0),
c(1, 0, 1, 0, 0, 1),
c(0, 1, 0, 1, 1, 0),
c(0, 0, 1, 0, 1, 1),
c(1, 0, 1, 1, 0, 1),
c(0, 1, 0, 1, 1, 0)
)
adj <- matrix(data = adj, nrow = 6, ncol = 6)
rownames(adj) <- as.character(1001:1006)
colnames(adj) <- as.character(1001:1006)

# Use non-consecutive block names
block <- c(70, 70, 70, 70, 95, 95)
g <- network::network(adj, matrix.type = "adjacency", directed = FALSE)
g %v% "block" <- block
g %v% "vertex.names" <- 1:length(g %v% "vertex.names")
est <- est_within(
formula = g ~ edges,
  network = g,
  parallel = FALSE,
  verbose = 0,
  initial_estimate = NULL,
  seed = NULL,
  method = "MPLE", 
  add_intercepts = FALSE,
  clustering_with_features = FALSE
  )
#> Starting maximum pseudolikelihood estimation (MPLE):
#> Obtaining the responsible dyads.
#> Evaluating the predictor and response matrix.
#> Maximizing the pseudolikelihood.
#> Finished MPLE.
#> Evaluating log-likelihood at the estimate. 
#>