cvxoptdes: Convex Optimal Experiment Design

Convex optimal experiment design across discrete candidate grids using open-source optimization solvers. Specifies and optimizes designs using an object-oriented interface (via 'R6') for linear models, generalized linear/additive models, nonlinear models, and random intercept linear mixed models. Optimization and augmentation of weighted, penalized and convex constrained approximate designs is supported using the Splitting Cone Solver (SCS). Optimal rounding to integer-valued designs is supported using point-exchange methods or mixed-integer programming with the 'HiGHS' solver. For details, see Chau et al. (2026) <doi:10.48550/arXiv.2610.14288>.

Version: 1.0.0
Depends: R (≥ 4.1.0)
Imports: methods, Matrix, stats, R6, reformulas, progress, mgcv, mirai
Suggests: highs, knitr, rmarkdown
Published: 2026-11-19
DOI: 10.32614/CRAN.package.cvxoptdes (may not be active yet)
Author: Joris Chau [aut, cre, cph]
Maintainer: Joris Chau <joris.chau at openanalytics.eu>
BugReports: https://github.com/openanalytics/cvxoptdes/issues
License: Apache License (≥ 2.0)
URL: https://repos.openanalytics.eu/docs/cvxoptdes/, https://github.com/openanalytics/cvxoptdes
NeedsCompilation: yes
SystemRequirements: GNU make, CMake
Language: en-US
Citation: cvxoptdes citation info
Materials: README, NEWS
CRAN checks: cvxoptdes results

Documentation:

Reference manual: cvxoptdes.html , cvxoptdes.pdf
Vignettes: Optimal design augmentation (source)
Example gallery (source)
Generalized linear/additive model optimal design (source)
Linear model optimal design (source)
Nonlinear/Linear mixed model optimal design (source)

Downloads:

Package source: cvxoptdes_1.0.0.tar.gz
Windows binaries: r-devel: not available, r-release: not available, r-oldrel: not available
macOS binaries: r-release (arm64): not available, r-oldrel (arm64): not available, r-release (x86_64): not available, r-oldrel (x86_64): not available

Linking:

Please use the canonical form https://CRAN.R-project.org/package=cvxoptdes to link to this page.