Adaptive Gauss Hermite Quadrature for Bayesian inference.
The AGHQ method for normalizing posterior distributions
and making Bayesian inferences based on them. Functions are provided for doing
quadrature and marginal Laplace approximations, and summary methods are provided
for making inferences based on the results.
See Stringer (2021). "Implementing Adaptive Quadrature for Bayesian Inference:
the aghq Package"
This package implements Bayesian inference using Adaptive Gauss-Hermite Quadrature, as specified in our paper Stochastic Convergence Rates and Applications of Adaptive Quadrature in Bayesian Inference with Yanbo Tang and Blair Bilodeau. See also the accompanying vignette, available on arXiv, and the complete code for the examples from that vignette.
You can install the development version from Github:
install.packages('devtools')
devtools::install_github('awstringer1/aghq')
You can also install the stable version from CRAN:
install.packages('aghq')
The two papers linked above give a comprehensive overview of the method, application, and theory.