imabc: Incremental Mixture Approximate Bayesian Computation (IMABC)

Provides functionality to perform a likelihood-free method for estimating the parameters of complex models that results in a simulated sample from the posterior distribution of model parameters given targets. The method begins with a accept/reject approximate bayes computation (ABC) step applied to a sample of points from the prior distribution of model parameters. Accepted points result in model predictions that are within the initially specified tolerance intervals around the target points. The sample is iteratively updated by drawing additional points from a mixture of multivariate normal distributions, accepting points within tolerance intervals. As the algorithm proceeds, the acceptance intervals are narrowed. The algorithm returns a set of points and sampling weights that account for the adaptive sampling scheme. For more details see Rutter, Ozik, DeYoreo, and Collier (2018) <arXiv:1804.02090>.

Getting started

Package details

AuthorChristopher, E. Maerzluft [aut, cre], Carolyn Rutter [aut, cph] (<>), Jonathan Ozik [aut] (<>), Nicholson Collier [aut] (<>)
Maintainer"Christopher, E. Maerzluft" <>
Package repositoryView on CRAN
Installation Install the latest version of this package by entering the following in R:

Try the imabc package in your browser

Any scripts or data that you put into this service are public.

imabc documentation built on April 12, 2021, 9:06 a.m.