Create posterior plots for Bayesian meta-analysis

Produces density plots of the pooled effect ($\mu$), heterogeneity ($\tau$), and/or prediction distribution from a fitted bayesma model.

Usage

overall_plot(
  model,
  estimand,
  add_tau = FALSE,
  add_mu_prior = FALSE,
  add_tau_prior = FALSE,
  add_pred = FALSE,
  color_pred_posterior = "orange",
  color_pred_posterior_outline = "darkorange2",
  null_value = NULL,
  null_range = NULL,
  add_null_range = FALSE,
  color_null_range = "#77bb41",
  label_control = "Control",
  label_intervention = "Intervention",
  title = NULL,
  subtitle = NULL,
  title_align = "left",
  mu_xlim = NULL,
  tau_xlim = NULL,
  pred_xlim = NULL,
  x_breaks = NULL,
  tau_breaks = NULL,
  pred_breaks = NULL,
  color_overall_posterior = "dodgerblue",
  color_overall_posterior_outline = "blue",
  split_color_by_null = FALSE,
  color_favours_control = "firebrick",
  color_favours_intervention = "dodgerblue",
  tau_posterior_color = "#77bb41",
  tau_posterior_outline = NULL,
  tau_slab_scale = 0.7,
  font = NULL,
  plot_arrangement = "vertical"
)

Arguments

model: A bayesma object.
estimand: Character string for the effect measure or marginal estimand. Relative-effect: "OR", "HR", "RR", "IRR", "MD", "SMD". Marginal: "RD"/"ARR", "ATE", "ATT", "CATE". For marginal estimands the pooled posterior is drawn from model$marginal$draws.
add_tau: Logical, whether to include the heterogeneity (tau) plot.
add_mu_prior: Logical, whether to include the prior distribution for mu.
add_tau_prior: Logical, whether to include the prior distribution for tau.
add_pred: Logical, whether to include the prediction distribution plot.
color_pred_posterior: Color for the prediction posterior distribution.
color_pred_posterior_outline: Outline color for the prediction posterior distribution.
null_value: Numeric, the null value for the effect (default from get_measure_properties).
null_range: Numeric vector of length 2, range for null region (e.g., c(0.9, 1.1)).
add_null_range: Logical, whether to add shaded null range region.
color_null_range: Color for the null range shading.
label_control: Character string for control group label.
label_intervention: Character string for intervention group label.
title: Plot title.
subtitle: Plot subtitle.
title_align: Alignment for title ("left", "center", "right").
mu_xlim: Numeric vector of length 2 for mu plot x-axis limits.
tau_xlim: Numeric vector of length 2 for tau plot x-axis limits.
pred_xlim: Numeric vector of length 2 for prediction plot x-axis limits.
x_breaks: Numeric vector for x-axis breaks on mu plot.
tau_breaks: Numeric vector for x-axis breaks on tau plot.
pred_breaks: Numeric vector for x-axis breaks on prediction plot.
color_overall_posterior: Color for the overall posterior distribution.
color_overall_posterior_outline: Outline color for the posterior distribution.
split_color_by_null: Logical, whether to split posterior color by null value.
color_favours_control: Color for region favoring control.
color_favours_intervention: Color for region favoring intervention.
tau_posterior_color: Base color for tau posterior.
tau_posterior_outline: Outline color for tau posterior.
tau_slab_scale: Numeric, scaling factor for tau slab height relative to mu (default 0.7).
font: Font family for text elements.
plot_arrangement: Character, "vertical" (stacked) or "horizontal" (side by side).

Value

A ggplot object (or combined plot if add_tau = TRUE or add_pred = TRUE).