ADtools: Automatic Differentiation Toolbox

Implements the forward-mode automatic differentiation for multivariate functions using the matrix-calculus notation from Magnus and Neudecker (2019) <doi:10.1002/9781119541219>. Two key features of the package are: (i) it incorporates various optimisation strategies to improve performance; this includes applying memoisation to cut down object construction time, using sparse matrix representation to speed up derivative calculation, and creating specialised matrix operations to reduce computation time; (ii) it supports differentiating random variates with respect to their parameters, targeting Markov chain Monte Carlo (MCMC) and general simulation-based applications.

Version: 0.5.4
Depends: R (≥ 3.6.0), methods, Matrix
Imports: purrr, dplyr, magrittr, assertthat, mvtnorm, Rcpp
LinkingTo: Rcpp, RcppArmadillo
Suggests: testthat, covr, knitr, rmarkdown, pryr, MCMCpack
Published: 2020-11-09
Author: Chun Fung Kwok ORCID iD [aut, cre], Dan Zhu ORCID iD [aut], Liana Jacobi ORCID iD [aut]
Maintainer: Chun Fung Kwok <kwokcf at>
License: MIT + file LICENSE
NeedsCompilation: yes
CRAN checks: ADtools results


Reference manual: ADtools.pdf
Vignettes: introduction-to-ADtools
Package source: ADtools_0.5.4.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel: not available
macOS binaries: r-release: ADtools_0.5.4.tgz, r-oldrel: ADtools_0.5.4.tgz


Please use the canonical form to link to this page.