Network Graph Visualization of Influence Matrices

Draws the estimated influence matrix as a directed network graph, where edges represent influence weights between nodes. Edge color and width encode the sign and magnitude of influence. Requires the igraph and ggraph packages to be installed.

Usage

plot_sir_network(
  x,
  matrix = c("A", "B"),
  threshold = 0.1,
  node_labels = NULL,
  layout = "fr"
)

Arguments

x

A fitted sir object from sir.

matrix

Character string: "A" (default) for sender effects or "B" for receiver effects.

threshold

Numeric. Edges with absolute influence below this value are hidden. Default is 0.1. Increase for cleaner plots with dense networks.

node_labels

Optional character vector of node names. If NULL, nodes are labeled 1 through m.

layout

Character string specifying the graph layout algorithm. Default is "fr" (Fruchterman-Reingold). Other options include "kk" (Kamada-Kawai) and "circle".

Value

A ggplot object produced by ggraph. Returns NULL if igraph or ggraph are not installed.