plot.latsoc: Plotting latent space models

In MNWeiss/aninet: Statistical Models for Animal Social Networks

Plotting latent space models

Description

Plots the fitted network object with node positions determined by latent space modelling. 1 and 2 dimensional fits can be plotted directly, while > 3 dimensional fits use multidimensional scaling to project distances into 2 dimensions.

Usage

## S3 method for class 'latsoc'
plot(
  object,
  post.method = "mean",
  edge.col = NULL,
  edge.lwd = NULL,
  vertex.pch = 21,
  vertex.col = "black",
  vertex.cex = 1,
  labels = F,
  label.col = NULL,
  label.cex = NULL,
  xlim = NULL,
  ylim = NULL
)

Arguments

object	A model returned by latent_space
post.method	Character specifying the method for estimating the posterior latent positions. One of "mean", "median", or "mds", the latter of which uses multidimensional scaling.
edge.col	Square matrix indicating the colors to plot edges. If NULL, edges are plotted in black, with transparency determined by their edge weight.
edge.lwd	Square matrix indicating edge line widths. If NULL, edge widths are proportional to edge weight.
vertex.pch	Plotting characters for vertices/nodes.
vertex.col	A vector indicating the colors for plotting nodes.
vertex.cex	Expansion factor for vertices/nodes.
labels	Logical, should node labels be plotted?
label.col	Colors for vertex labels.
label.cex	Numeric, character expansion factor for labels.
xlim	Optional specification of x limits of plot
ylim	Optional specifical of y limits of plot

Details

This function uses base R plotting methods. More advanced network plotting methods are available (e.g. through the ggraph package), but this interface provides an easy way to visualize the results of the model quickly.

MNWeiss/aninet documentation built on Jan. 31, 2023, 3:55 a.m.