In kenkellner/jagsUI: A Wrapper Around 'rjags' to Streamline 'JAGS' Analyses

knitr::opts_chunk$set(
  fig.path = "README_figs/README-"
)

jagsUI: Run JAGS from R

This package runs JAGS (Just Another Gibbs Sampler) analyses from within R. It acts as a wrapper and alternative interface for the functions in the rjags package and adds some custom output and graphical options. It also makes running chains in parallel quick and easy.

Installation

You can install the package from CRAN, or get the development version from Github:

devtools::install_github('kenkellner/jagsUI')

You will also need to separately install JAGS, which you can download here.

Example

library(jagsUI)

Format data:

jags_data <- list(
  gnp = longley$GNP,
  employed = longley$Employed,
  n = nrow(longley)
)

Write BUGS model file:

modfile <- tempfile()
writeLines("
model{

  # Likelihood
  for (i in 1:n){ 
    # Model data
    employed[i] ~ dnorm(mu[i], tau)
    # Calculate linear predictor
    mu[i] <- alpha + beta*gnp[i]
  }

  # Priors
  alpha ~ dnorm(0, 0.00001)
  beta ~ dnorm(0, 0.00001)
  sigma ~ dunif(0,1000)
  tau <- pow(sigma,-2)

}
", con=modfile)

Set initial values and parameters to save:

inits <- function(){  
  list(alpha=rnorm(1,0,1),
       beta=rnorm(1,0,1),
       sigma=runif(1,0,3)
  )  
}

params <- c('alpha','beta','sigma')

Run JAGS:

out <- jags(data = jags_data,
            inits = inits,
            parameters.to.save = params,
            model.file = modfile,
            n.chains = 3,
            n.adapt = 100,
            n.iter = 1000,
            n.burnin = 500,
            n.thin = 2)

View output:

out

Acknowledgments

• Martyn Plummer, developer of the excellent JAGS software package and the rjags R package.
• Andrew Gelman, Sibylle Sturtz, Uwe Ligges, Yu-Sung Su, and Masanao Yajima, developers of the R2WinBUGS and R2jags packages on which the package was originally based.
• Robert Swihart, Marc Kery, Jerome Guelat, Michael Schaub, and Mike Meredith who tested and provided helpful suggestions and improvements for the package.

kenkellner/jagsUI documentation built on Feb. 4, 2024, 5:20 a.m.