Plots the posterior marginal density of a parameter. Users may specify whether they want a histogram, a density, or both.
plot_marginal(parameter, percent.burnin = 0, thinning = 1, histogram = TRUE,
density = TRUE, population.names = NULL, param.name = deparse(substitute(parameter)))
The parameter for which the marginal plot is being generated.
The percent of the sampled MCMC generations to be discarded as "burn-in." If the
MCMC is run for 1,000,000 generations, and sampled every 1,000 generations, there
will be 1,000 sampled generations. A percent.burnin of 20 will discard
the first 200 sampled parameter values from that sample.
The multiple by which the sampled MCMC generations are thinned. A thinning of
5 will sample every 5th MCMC generation.
A switch that controls whether or not the plot contains a histogram of the values
estimated for the parameter over the course of the MCMC. Default is TRUE.
A switch that controls whether or not the plot shows the density of the values
estimated for the parameter over the course of the MCMC. Default is TRUE.
A vector of length k, where k is the number of populations/individuals
(i.e. k = nrow(counts)), giving the name or identifier of each
population/individual included in the analysis. These will be used to title the
k marginal plots of the phi parameters estimated for each
population/individual in the beta-binomial model. If the binomial model is used,
population.names will not be used by this function.
The name of the parameter for which the trace plot is being displayed.
Gideon Bradburd
The marginal plot is another basic visual tool for MCMC diagnosis. Users should look for marginal plots that are "smooth as eggs" (indicating that the chain has been run long enough) and unimodal (indicating a single peak in the likelihood surface).