Linear Latent Non-Gaussian Models with Flexible Distributions
Overview
Fits and analyzes linear latent non-Gaussian models for temporal, spatial, and space-time data. The package provides model components for autoregressive and Ornstein-Uhlenbeck processes, random walks, Matern fields based on stochastic partial differential equations, separable and non-separable space-time models, graph-based Matern models, bivariate type-G fields, and user-defined sparse operators. Latent fields and observation models can use Gaussian and non-Gaussian noise distributions, including normal inverse Gaussian, generalized asymmetric Laplace, and skew-t distributions. Functions are included for simulation, likelihood-based estimation, prediction, cross-validation, convergence diagnostics, stochastic gradient optimization, batch-means confidence intervals, and posterior-like sampling. The modeling framework is described in Bolin, Jin, Simas and Wallin (2026) "A Unified and Computationally Efficient Non-Gaussian Statistical Modeling Framework" doi:10.48550/arXiv.2602.23987.
- Limit parallel fits in automated tests to two threads without reducing the number of estimation chains.
- Explicitly discard unused random-number draws when advancing the generator; this removes the Clang/libc++ warning while preserving the random stream.
- Update CHOLMOD integration for Eigen 5.0.1 and declare Matrix in LinkingTo.
- Check the current source package on Fedora with both GCC and Clang, including all CRAN-hosted suggested packages and full test progress in the check logs.
From NEWS.md in the release
Install
Health
- OK13 OKNow
- OK9 OK2026-10-14
- ERROR3 OK, 2 ERROR2026-04-29
Documentation
- Help pages
- 116 (4 internal)includes a package overview page
- Examples that run
- 100%28 run in a plain check
- Exported functions with examples
- 29%
- Documented parameters
- 94%
- Return-value docs
- 100%
- Help pages citing references
- 7%
- ngme2 Documentation HTML
Downloads
Repository
- PRs merged (all time)
- 100%
- Median days to close, 50 most recently updated closed issues
- 471
- Median days to close, 50 most recently updated closed PRs
- 0
Forks over time
Releases over time
PRs over time
Commits over time
Commit identities over time
5 development-tooling and community-health practices detected across 3 families in the upstream repository
Checks run against github.com/davidbolin/ngme2 on 2026-10-21.
The R CMD check workflow runs on Linux, macOS and Windows
GitHub Pages site last deployed 2026-10-13
pkgdown site last built 2026-04-06, before release 1.0.1 (2026-10-14)
This repository shows signs of AI coding tools.
- Aiderfirst seen 2025-08-18Named in .gitignore: .aider*
- Claude Codefirst seen 2025-09-11Named in .Rbuildignore: .claudeCommits: at least 1 crediting it in the message
Dependencies
Code & Tests
Datasets
People & History
Show 4 more contributorsHide 4 contributors
Pick two releases to compare their code metrics.
- unarchivedReturned to CRAN2026-10-14
- 1.0.1Latest2026-10-14 · current release · diff ↗ ·
13 hours in CRAN's queue
Listed in CRAN's incoming queue from 2026-10-13 16:37 to 2026-10-14 05:40 UTC, in 4 readings. - archivedRemoved from CRAN2026-10-11issues were not corrected in time
- 1.0.0
- 0.9.8
- 0.9.7first seen 2026-04-28
- RR 4.6.0 released · 2026-04-24
Package metadata
- First seen
- 2026-04-28
- Total releases
- 4 / 1 yrs
- License
- GPL (>= 3) OSI
- Copyright
- The R package and code, and the main programs, were written by and are Copyright by David Bolin, Xiaotian Jin, Alexandre Simas, and Jonas Wallin, and are redistributable under the GNU Public License, version 3 or later. The package also includes bundled SuiteSparse components. For details see inst/COPYRIGHTS.
- Additional repositories
- inla.r-inla-download.org
- Minimum R
- ≥ 3.5.0
- Bundled data
- 52 KB / 4 files
- Download size
- 2.2 MB
Cite
Cite this package
Run in R for the authors' preferred citation:
citation("ngme2")This package has no citation file, so citation() returns this entry built from DESCRIPTION.
Cite the R Observatory
For a number measured here: a download total, a coverage figure, an archival date.
From data release v2026-10-25, which the citation names so these numbers can be found later. More on citing and the projects behind them.