proxymix: Kullback-Leibler Optimal Gaussian Mixture Proxies for Target Densities

Fits multivariate Gaussian-mixture proxies that are Kullback-Leibler optimal to user-supplied target densities on real Euclidean space. Three fitting regimes are unified under one verb: (i) closed-form moment matching for a single component, (ii) classical expectation-maximisation when independent samples are available, and (iii) importance-sampled expectation-maximisation that minimises the Kullback-Leibler divergence when the target can be evaluated point-wise but not (cheaply) sampled. Closed-form Gaussian-mixture operators (density, sampling, marginalisation, conditioning, divergence) round out the toolkit. The conditioning operator drives multiple imputation of data missing at random, covering the multimodal and heteroscedastic cases a single-Gaussian model cannot represent. Implements the regime hierarchy of van der Hoek and Elliott (2024) <doi:10.1080/07362994.2024.2372605>.

Version:	0.16.0
Depends:	R (≥ 4.3)
Imports:	S7, stats, mvnfast, cli, rlang, data.table, withr
Suggests:	mclust, mice, generics, ggplot2, viridis, patchwork, knitr, rmarkdown, testthat (≥ 3.0.0), vdiffr, spelling, covr
Published:	2026-10-08
DOI:	10.32614/CRAN.package.proxymix (may not be active yet)
Author:	Max Moldovan [aut, cre], John van der Hoek [aut]
Maintainer:	Max Moldovan <max.moldovan at gmail.com>
BugReports:	https://github.com/max578/proxymix/issues
License:	MIT + file LICENSE
URL:	https://github.com/max578/proxymix, https://max578.github.io/proxymix/
NeedsCompilation:	no
Language:	en-AU
Citation:	proxymix citation info
Materials:	README, NEWS
CRAN checks:	proxymix results

Documentation:

Reference manual:

proxymix.html , proxymix.pdf

Vignettes:

Mapping the optima of an objective (source, R code) How well a mixture proxies four awkward shapes (source, R code) Testing the last observation for instability (source, R code) Reading the entropy of a fitted mixture (source, R code) Compressing a kernel density estimate into a mixture (source, R code) One mixture, many methods (source, R code) Imputing missing data with a mixture (source, R code) Missing data that depends on the missing value (source, R code) The closed-form operator calculus on a mixture (source, R code) Compressing a Bayesian posterior you can evaluate but not sample (source, R code) Fitting a proxy to a density you cannot sample (source, R code) Choosing between the three fitting regimes (source, R code)

Downloads:

Package source:	proxymix_0.16.0.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): proxymix_0.16.0.tgz, r-oldrel (x86_64): proxymix_0.16.0.tgz

Linking:

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