Skip to contents

Diagnostic Plots for a Fitted SIR Model

plot.sir.Rd

Produces a selection of diagnostic plots for model assessment. By default, plots 1-4 (influence matrix heatmaps and distributions) are shown. Use the which argument to select specific plots. All plots use ggplot2 and are combined via patchwork when combine = TRUE.

Usage

# S3 method for class 'sir'
plot(
  x,
  which = 1:4,
  combine = TRUE,
  title = NULL,
  theme_base = theme_bw(),
  ...
)

Arguments

x

A fitted sir object from sir.

which

Integer vector selecting which plots to produce. Options:

1

Heatmap of sender influence matrix A. Shows how each node influences others' outgoing ties.

2

Heatmap of receiver influence matrix B. Shows how each node affects others' incoming ties.

3

Histogram and density of off-diagonal A values. Useful for assessing the overall strength and distribution of sender effects.

4

Histogram and density of off-diagonal B values. Same for receiver effects.

5

Convergence trace plot showing deviance across ALS iterations. Iteration history is always stored in the fitted model.

6

Coefficient plot with 95% confidence intervals. Requires standard errors (calc_se = TRUE). Parameters are grouped by type (exogenous, alpha, beta).

combine

Logical. If TRUE (default), combines selected plots into a single patchwork layout. If FALSE, returns a list of individual plots.

title

Optional character string for the combined plot title.

theme_base

A ggplot2 theme applied to all plots. Default is theme_bw().

...

Additional arguments (unused).

Value

When combine = TRUE and multiple plots are requested, a patchwork object. When a single plot is requested, a ggplot object. When combine = FALSE, a named list of ggplot objects.

Examples

if (FALSE) { # \dontrun{
model <- sir(Y, W, X, family = "poisson")

# default: influence heatmaps and distributions
plot(model)

# all plots combined with a title
plot(model, which = 1:6, title = "SIR Diagnostics")

# individual plots for custom arrangement
plots <- plot(model, which = c(1, 6), combine = FALSE)
plots$A_heatmap
plots$coef_plot
} # }