Network-level posterior summary

Compute network-level statistics at each time point across posterior draws (e.g., mean edge weight, density of positive edges)

Usage

network_summary(
  fit,
  stat = c("mean", "density", "strength"),
  rel = 1,
  time = NULL
)

Arguments

fit

A dbn model fit object

stat

Statistic to compute: "mean" (average theta), "density" (fraction of positive values), or "strength" (mean absolute theta)

rel

Relation index (default: 1)

time

Time indices (default: all)

Value

Data frame with columns: time, mean, lower, upper

See also

edge_prob, theta_summary, param_summary

Examples

# \donttest{
sim <- simulate_dynamic_dbn(n = 6, time = 5, seed = 1)
fit <- dbn(sim$Y, model = "dynamic", nscan = 200, burn = 100, verbose = FALSE)
ns <- network_summary(fit)
# }