sensobol: Computation of Variance-Based Sensitivity Indices

It allows to rapidly compute, bootstrap and plot up to fourth-order Sobol'-based sensitivity indices using several state-of-the-art first and total-order estimators. Sobol' indices can be computed either for models that yield a scalar as a model output or for systems of differential equations. The package also provides a suit of benchmark tests functions and several options to obtain publication-ready figures of the model output uncertainty and sensitivity-related analysis. An overview of the package can be found in Puy et al. (2022) <doi:10.18637/jss.v102.i05>.

Version: 1.1.5
Depends: R (≥ 3.5.0)
Imports: boot (≥ 1.3.20), data.table (≥ 1.12.0), ggplot2 (≥ 3.1.0), lhs (≥ 1.0.2), magrittr (≥ 1.5), matrixStats (≥ 0.54.0), randtoolbox (≥ 1.17.1), deSolve (≥ 1.27.1), Rdpack (≥ 2.1.2), Rfast (≥ 2.0.1), rlang (≥ 0.3.1), scales (≥ 1.0.0), stats, stringr (≥ 1.4.0), utils, Rcpp
LinkingTo: Rfast, Rcpp, RcppArmadillo
Suggests: knitr, rmarkdown, testthat (≥ 2.1.0), covr
Published: 2024-02-21
DOI: 10.32614/CRAN.package.sensobol
Author: Arnald Puy ORCID iD [aut, cre], Bertrand Ioos [ctb] (Author of included 'sensitivity' fragments), Gilles Pujol [ctb] (Author of included 'sensitivity' fragments), RStudio [cph] (Copyright holder of included 'sensitivity' fragments)
Maintainer: Arnald Puy <arnald.puy at>
License: GPL-3
NeedsCompilation: yes
Citation: sensobol citation info
Materials: README NEWS
CRAN checks: sensobol results


Reference manual: sensobol.pdf
Vignettes: author2019mypaper


Package source: sensobol_1.1.5.tar.gz
Windows binaries: r-devel:, r-release:, r-oldrel:
macOS binaries: r-release (arm64): sensobol_1.1.5.tgz, r-oldrel (arm64): sensobol_1.1.5.tgz, r-release (x86_64): sensobol_1.1.5.tgz, r-oldrel (x86_64): sensobol_1.1.5.tgz
Old sources: sensobol archive


Please use the canonical form to link to this page.