Comparison of Bayesian models for production efficiency

In this paper, we use Markov Chain Monte Carlo (MCMC) methods in order to estimate and compare stochastic production frontier models from a Bayesian perspective. We consider a number of competing models in terms of different production functions and the distribution of the asymmetric error term. All MCMC simulations are done using the package <monospace>JAGS</monospace> (Just Another Gibbs Sampler), a clone of the classic <monospace>BUGS</monospace> package which works closely with the <monospace>R</monospace> package where all the statistical computations and graphics are done.