Figure 3.

(A) Shows the true directed causal structure and the estimated undirected network for v = 10 nodes, a network density of d = 0.3, and n = 1000 observations. Node size is a function of the (z-standardized) causal effect (left) or betweenness centrality (right). (B) Shows the distributions of the causal effect scores CE_KL and betweenness centrality C_B for the ten nodes in the example. Nodes are ordered by causal effect. In this example, the rank-correlation between the causal effects and betweenness centrality is 0.32. As our simulations show, this is a prototypical case. An interactive version of this plot is available as a R Shiny app at https://fdabl.shinyapps.io/causality-centrality-app/.