Repository files navigation

Provides functions for the Bayesian analysis of some simplecommonly-used models, without using Markov Chain Monte Carlo (MCMC)methods such as Gibbs sampling. The ‘rust’ packagehttps://cran.r-project.org/package=rust is used to simulate a randomsample from the required posterior distribution, using theratio-of-uniforms method. Currently three conjugate hierarchical modelsare available: beta-binomial, gamma-Poisson and a 1-way Analysis ofVariance (ANOVA). Advantages of the ratio-of-uniforms method over MCMCin this context are that the user is not required to set tuningparameters nor to monitor convergence and a random posterior sample isproduced.

Thehef function samples from the posterior distribution of theparameters of certain hierarchical exponential family models. Thefollowing code performs essentially the same analysis of the rat tumordata using a beta-binomial hierarchical model that appears in Section5.3 of Gelman, A., Carlin, J. B., Stern, H. S. Dunson, D. B., Vehtari,A. and Rubin, D. B. (2014) Bayesian Data Analysis. Chapman & Hall / CRC.http://www.stat.columbia.edu/~gelman/book/.

library(bang)rat_res<- hef(model="beta_binom",data=rat)plot(rat_res)

To get the current released version from CRAN:

install.packages("bang")