The plotBayes package illustrates Bayesian inference, showing how the prior distribution and the likelihood of the data combine to produce a posterior distribution.

This is mostly a toy package useful for teaching.

Installation

You need the devtools package to install this from github.

install.packages("devtools")

Then install plotBayes.

devtools::install_github("mcbeem/plotBayes")

And then load it.

library(plotBayes)
library(plotBayes)

Example

Normal prior with $\mu=0$, $\sigma=0.5$:

set.seed(1)
data <- rnorm(n=10, mean=1, sd=1)
plotBayes(data, prior.type="normal", prior.parameters=c(0, .5), min=-2, max=2)

You can request a different credible interval with with argument credible=.

plotBayes(data, prior.parameters=c(.0, .5), prior.type="normal",
          min=-2, max=2, credible=.68)

Uniform prior with $a=.7$, $b=1.5$:

plotBayes(data, prior.type="uniform", prior.parameters=c(.7, 1.5), min=-2, max=2)


mcbeem/plotBayes documentation built on Nov. 19, 2019, 8:15 a.m.