A package with general-purpose MCMC and SMC samplers, as well as plots and diagnostic functions for Bayesian statistics
A package with general-purpose MCMC and SMC samplers, as well as plots and diagnostic functions for Bayesian statistics, particularly for process-based models.
The package contains 2 central functions, createBayesianSetup, which creates a standardized Bayesian setup with likelihood and priors, and runMCMC, which allows to run various MCMC and SMC samplers.
The package can of course also be used for general (non-Bayesian) target functions.
To use the package, a first step to use createBayesianSetup to create a BayesianSetup, which usually contains prior and likelihood densities, or in general a target function.
Those can be sampled with runMCMC, which can call a number of general purpose Metropolis sampler, including the Metropolis that allows to specify various popular Metropolis variants such as adaptive and/or delayed rejection Metropolis; two variants of differential evolution MCMC DE, DEzs, two variants of DREAM DREAM and DREAMzs, the Twalk MCMC, and a Sequential Monte Carlo sampler smcSampler.
The output of runMCMC is of class mcmcSampler / smcSampler if one run is performed, or mcmcSamplerList / smcSamplerList if several sampler are run. Various functions are available for plotting, model comparison (DIC, marginal likelihood), or to use the output as a new prior.
For details on how to use the packgage, run vignette("BayesianTools", package="BayesianTools").
To get the suggested citation, run citation("BayesianTools")
To report bugs or ask for help, post a reproducible example via the BayesianTools issue tracker on GitHub.
Acknowledgements: The creation of this package was facilicated through meetings of Cost Action FP 1304 Profound.