Skip to contents

Plot MCMC Diagnostics

Source: R/plot-diagnostics.R
plot_diagnostic.Rd

plot_diagnostic creates visual MCMC diagnostics for a fitted brma object. Convenience wrappers are available for trace, density, and autocorrelation plots.

Usage

plot_diagnostic(x, ...)

# S3 method for class 'brma'
plot_diagnostic(
  x,
  parameter = NULL,
  parameter_mods = NULL,
  parameter_scale = NULL,
  type,
  plot_type = "base",
  lags = 30,
  component = "auto",
  ...
)

plot_diagnostic_autocorrelation(x, ...)

# S3 method for class 'brma'
plot_diagnostic_autocorrelation(
  x,
  parameter = NULL,
  parameter_mods = NULL,
  parameter_scale = NULL,
  type = "autocorrelation",
  plot_type = "base",
  lags = 30,
  component = "auto",
  ...
)

plot_diagnostic_trace(x, ...)

# S3 method for class 'brma'
plot_diagnostic_trace(
  x,
  parameter = NULL,
  parameter_mods = NULL,
  parameter_scale = NULL,
  type = "trace",
  plot_type = "base",
  lags = 30,
  component = "auto",
  ...
)

plot_diagnostic_density(x, ...)

# S3 method for class 'brma'
plot_diagnostic_density(
  x,
  parameter = NULL,
  parameter_mods = NULL,
  parameter_scale = NULL,
  type = "density",
  plot_type = "base",
  lags = 30,
  component = "auto",
  ...
)

Arguments

x

a fitted brma object

...

additional graphical arguments passed through RoBMA's diagnostic setup to BayesTools::JAGS_diagnostics().

parameter

base parameter to plot. Defaults to NULL, which uses "mu" or the meta-regression intercept. Valid values include "mu", "tau", "rho", "PET", "PEESE", and "omega" or "weightfunction" when present. A quantity of the fitted model that is no model parameter (e.g. an inclusion indicator or a weight-function coordinate such as "omega[1]") stops with the classes BayesTools_hypothesis_target and BayesTools_hypothesis_unavailable, as in plot.brma().

parameter_mods

legacy moderator selector. Prefer parameter with component = "mods".

parameter_scale

legacy scale-regression selector. Prefer parameter with component = "scale".

type

diagnostic plot type. Convenience wrappers set a type-specific default but still forward this argument to plot_diagnostic.brma().

plot_type

whether to use a base plot "base" or ggplot2 "ggplot" for plotting. Defaults to "base".

lags

number of lags for autocorrelation plots. Defaults to 30.

component

parameter component. Defaults to "auto", which infers the component when possible. Use "mods" (alias "location"), "scale", "random", or "bias" to disambiguate terms used in multiple model components. The random component selects semantic standard deviation, correlation, and allocation parameters from brma.mv() random formulas. A parameter_mods or parameter_scale selection with another component stops with an error of class RoBMA_component_mismatch.

Value

plot_diagnostic returns the object returned by BayesTools::JAGS_diagnostics(), including a list when several coordinate plots are produced, invisibly for base graphics.

See also

summary.brma()