mrc-ide/dust: Iterate Multiple Realisations of Stochastic Models

An Engine for simulation of stochastic models. Includes support for running stochastic models in parallel, either with shared or varying parameters. Simulations are run efficiently in compiled code and can be run with a fraction of simulated states returned to R, allowing control over memory usage. Support is provided for building bootstrap particle filter for performing Sequential Monte Carlo (e.g., Gordon et al. 1993 <doi:10.1049/ip-f-2.1993.0015>). The core of the simulation engine is the 'xoshiro256**' algorithm (Blackman and Vigna <arXiv:1805.01407>), and the package is further described in FitzJohn et al 2021 <doi:10.12688/wellcomeopenres.16466.2>.

Getting started

Package details

Maintainer
LicenseMIT + file LICENSE
Version0.15.2
URL https://github.com/mrc-ide/dust
Package repositoryView on GitHub
Installation Install the latest version of this package by entering the following in R:
install.packages("remotes")
remotes::install_github("mrc-ide/dust")
mrc-ide/dust documentation built on May 11, 2024, 1:08 p.m.