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 0.7.0 provides a variety of new features, as well as various bug fixes.
New features include:
- greatly improved efficiency of sampling for Bayesian nonparametric (BNP) mixture models that use the dCRP (Chinese Restaurant process) distribution;
- addition of the double exponential (Laplace) distribution for use in models and nimbleFunctions;
- a new “RW_wishart” MCMC sampler, for sampling non-conjugate Wishart and inverse-Wishart nodes;
- handling of the normal-inverse gamma conjugacy for BNP mixture models using the dCRP distribution;
- enhanced functionality of the getSamplesDPmeasure function for posterior sampling from BNP random measures with Dirichlet process priors.
- handling of five-dimensional arrays in models;
- enhanced warning messages; and
- an HTML version of the NIMBLE manual.
Please see the NEWS file in the installed package for more details.