Bayesian inference using the no-U-turn (NUTS) algorithm by Hoffman and Gelman (2014) < https://www.jmlr.org/papers/v15/hoffman14a.html>. Designed for 'AD Model Builder' ('ADMB') models, or when R functions for log-density and log-density gradient are available, such as 'Template Model Builder' models and other special cases. Functionality is similar to 'Stan', and the 'rstan' and 'shinystan' packages are used for diagnostics and inference.
The aim of 'adnuts' is to provide advanced MCMC sampling for 'ADMB' and 'TMB' models. It mimics 'Stan' in functionality and feel, specifically providing no-U-turn (NUTS) sampling with adaptive mass matrix and parallel execution.
The 'sample_admb' and 'sample_tmb' functions draw posterior samples using an MCMC algorithm (NUTS by default). The returned fitted object contains samples and other information. The function 'extract_samples' can be used to get posterior samples (post warmup and thinning) into a data frame for inference, while 'launch_shinyadmb' can be used for interactive diagnostics based on 'ShinyStan'.
A brief demonstration file is provided to help get you started, and there is also a user guide:
vignette('adnuts') for more detailed information
The adnuts R package can be installed from CRAN:
install.packages('adnuts'). To use the ADMB functionality you need to build your model with version 12.0 (released December 2017) or later, otherwise this functionality is not available.