Posterior Summary of a BBNI MCMC Object

Prints a statistical summary of the MCMC run and returns a structured list of posterior quantities for further usage in downstream code.

Usage

# S3 method for class 'bbni'
summary(object, threshold = 0.5, n_top = 5, ...)

Arguments

object: A bbni object returned by run_bbni().
threshold: Numeric. Posterior probability threshold for counting strong edges.
n_top: Integer. Number of highest-posterior edges to display and include in top_edges.
...: Additional arguments (ignored).

Value

A list with components num_nodes, num_update, burn_in, final_log_posterior, n_strong_edges, and top_edges (data frame with columns child, parent, and posterior).

Examples

set.seed(123)
net <- GenerateNetwork(5)
data <- GenerateSample(net, 100)
res <- run_bbni(data, num_update = 50)
summary(res, threshold = 0.6, n_top = 3)
#> BBNI MCMC summary
#>   Nodes:                 5
#>   Outer iterations:      50
#>   Burn-in ratio:         0.70
#>   Final log-posterior:   -87.922
#>   Edges with P > 0.60:    5
#>   Highest-posterior edges (child <- parent):
#>     N3 <- N1 (P = 1.00)
#>     N4 <- N1 (P = 1.00)
#>     N5 <- N1 (P = 1.00)