facetPost: Plot Multiple Posterior Distributions

In abnormally-distributed/Bayezilla: Bayesian Models With JAGS and Several Utilities for Data Exploration and Model Evaluation.

Description

This lets you plot several different posterior histograms at once in a faceted plot. The number of columns, rows, and several other features can be customized. It is also easy to drop or keep certain parameters so you can view only the ones of interest.

Usage

facetPost(fit, keeppars = NULL, droppars = c("ySim", "log_lik",
  "lp__"), col = "blue", nrow = 4, ncol = 2, breaks = "dhist",
  method = "QI", est = "mean", cred.level = 0.9)

Arguments

fit	a stanfit or runjags object. This can be used as an alternative to paramSampleVec, but you must specify which parameter you would like to plot. If using this argument be sure to type fit = "yourmodel" so that the function knows it is not intended to be a vector.
keeppars	variables to keep
droppars	variables to drop. Defaults to c("ySim", "log_lik", "lp__")
col	the color scheme. One of "blue", "green" "red", or "purple".
nrow	number of rows in the layout
ncol	number of columns in the layout
breaks	a number, function name, or function.
method	Quantile intervals "QI" (the default), or highest density intervals "HDI"
est	One of "mean" (the default), "median", or "mode". The mode gives the value corresponding to the highest density of a kernel density estimate of the marginal posterior.
cred.level	The credibility level. Defaults to 90% (.90).
comparison	value, the second being the lower limit of the ROPE and the third being the upper limit of the ROPE.

Details

An example of output:

Value

A plot

Examples

facetPost()

abnormally-distributed/Bayezilla documentation built on Oct. 31, 2019, 1:57 a.m.