We’ve released the newest version of NIMBLE on CRAN and on our website. NIMBLE is a system for building and sharing analysis methods for statistical models, especially for hierarchical models and computationally-intensive methods (such as MCMC and SMC). Version 0.8.0 provides some new features, speed improvements, and a variety of bug fixes and better error/warning messages.
New features include:
- a reversible jump MCMC sampler for variable selection via configureRJ();
- greatly improved speed of MCMC sampling for Bayesian nonparametric models with a dCRP distribution by not sampling parameters for empty clusters;
- experimental faster MCMC configuration, available by setting nimbleOptions(oldConjugacyChecking = FALSE) and nimbleOptions(useNewConfigureMCMC = TRUE);
- and improved warning and error messages for MCEM and slice sampling.
Please see the release notes on our website for more details.