nabla: Exact Derivatives via Automatic Differentiation
Exact automatic differentiation for R functions. Provides a composable derivative operator D that computes gradients, Hessians, Jacobians, and arbitrary-order derivative tensors at machine precision. D(D(f)) gives Hessians, D(D(D(f))) gives third-order tensors for skewness of maximum likelihood estimators, and so on to any order. Works through any R code including loops, branches, and control flow.
| Version: | 0.7.1 |
| Depends: | R (≥ 4.0.0), methods |
| Imports: | stats |
| Suggests: | testthat (≥ 3.0.0), knitr, rmarkdown |
| Published: | 2026-02-10 |
| DOI: | 10.32614/CRAN.package.nabla |
| Author: | Alexander Towell 
[aut, cre] |
| Maintainer: | Alexander Towell <queelius at gmail.com> |
| BugReports: | https://github.com/queelius/nabla/issues |
| License: | MIT + file LICENSE |
| URL: | https://github.com/queelius/nabla, https://metafunctor.com |
| NeedsCompilation: | no |
| Citation: | nabla citation info |
| Materials: | README, NEWS |
| CRAN checks: | nabla results |
Documentation:
| Reference manual: | nabla.html , nabla.pdf |
| Vignettes: |
Higher-Order Derivatives (source, R code) Introduction to nabla (source, R code) Higher-Order MLE Analysis (source, R code) Gradient and Hessian Computation with nabla (source, R code) Optimizer Integration (source, R code) |
Downloads:
| Package source: | nabla_0.7.1.tar.gz |
| Windows binaries: | r-devel: nabla_0.7.1.zip, r-release: nabla_0.7.1.zip, r-oldrel: nabla_0.7.1.zip |
| macOS binaries: | r-release (arm64): nabla_0.7.1.tgz, r-oldrel (arm64): nabla_0.7.1.tgz, r-release (x86_64): nabla_0.7.1.tgz, r-oldrel (x86_64): nabla_0.7.1.tgz |
Linking:
Please use the canonical form https://CRAN.R-project.org/package=nabla to link to this page.