Fixed Rank Kriging
Overview
A tool for spatial/spatio-temporal modelling and prediction with large datasets. The approach models the field, and hence the covariance function, using a set of basis functions. This fixed-rank basis-function representation facilitates the modelling of big data, and the method naturally allows for non-stationary, anisotropic covariance functions. Discretisation of the spatial domain into so-called basic areal units (BAUs) facilitates the use of observations with varying support (i.e., both point-referenced and areal supports, potentially simultaneously), and prediction over arbitrary user-specified regions. `FRK` also supports inference over various manifolds, including the 2D plane and 3D sphere, and it provides helper functions to model, fit, predict, and plot with relative ease. Version 2.0.0 and above also supports the modelling of non-Gaussian data (e.g., Poisson, binomial, negative-binomial, gamma, and inverse-Gaussian) by employing a generalised linear mixed model (GLMM) framework. Zammit-Mangion and Cressie doi:10.18637/jss.v098.i04 describe `FRK` in a Gaussian setting, and detail its use of basis functions and BAUs, while Sainsbury-Dale, Zammit-Mangion, and Cressie doi:10.18637/jss.v108.i10 describe `FRK` in a non-Gaussian setting; two vignettes are available that summarise these papers and provide additional examples.
Install
Health
ERRORre-building of vignette outputsr-devel-linux-x86_64-fedora-clang, r-devel-linux-x86_64-fedora-gcc
r-devel-linux-x86_64-fedora-clang
Error(s) in re-building vignettes: --- re-building ‘FRK_intro.Rnw’ using knitr OMP: Warning #96: Cannot form a team with 28 threads, using 2 instead. OMP: Hint Consider unsetting KMP_DEVICE_THREAD_LIMIT (KMP_ALL_THREADS), KMP_TEAMS_THREAD_LIMIT, and OMP_THREAD_LIMIT (if any are set). --- re-building ‘FRK_non-Gaussian.Rnw’ using knitr --- finished re-building ‘FRK_non-Gaussian.Rnw’ SUMMARY: processing the following file failed: ‘FRK_intro.Rnw’ Error: Vignette re-building failed. Execution halted
r-devel-linux-x86_64-fedora-gcc
Error(s) in re-building vignettes: --- re-building ‘FRK_intro.Rnw’ using knitr --- re-building ‘FRK_non-Gaussian.Rnw’ using knitr --- finished re-building ‘FRK_non-Gaussian.Rnw’ SUMMARY: processing the following file failed: ‘FRK_intro.Rnw’ Error: Vignette re-building failed. Execution halted
ERRORtestsr-devel-linux-x86_64-fedora-clang, r-devel-linux-x86_64-fedora-gcc
r-devel-linux-x86_64-fedora-clang
Running ‘testthat.R’ [90m/53m]
Running the tests in ‘tests/testthat.R’ failed.
Complete output:
> ops <- options("crayon.enabled" = FALSE)
> library(testthat)
> require("covr")
Loading required package: covr
>
> test_check("FRK")
Loading required package: FRK
Attaching package: 'FRK'
The following object is masked from 'package:stats':
simulate
NOTE: In FRK >2.0 simple_kriging_fixed = FALSE by default, and hence
universal kriging is done by default. However this is only the case when
method = 'TMB'. When method = 'EM', simple kriging is done, irrespective of
what the argument simple_kriging_fixed is set to.
|
| | 0%
|
|=======================
...[truncated]...
|======================= | 33%
|
|=============================================== | 67%
|
|======================================================================| 100%
Maximum EM iterations reached
Averaging over polygons...
NOTE: In FRK >2.0 simple_kriging_fixed = FALSE by default, and hence
universal kriging is done by default. However this is only the case when
method = 'TMB'. When method = 'EM', simple kriging is done, irrespective of
what the argument simple_kriging_fixed is set to.
|
| | 0%r-devel-linux-x86_64-fedora-gcc
Running ‘testthat.R’ [90m/48m]
Running the tests in ‘tests/testthat.R’ failed.
Complete output:
> ops <- options("crayon.enabled" = FALSE)
> library(testthat)
> require("covr")
Loading required package: covr
>
> test_check("FRK")
Loading required package: FRK
Attaching package: 'FRK'
The following object is masked from 'package:stats':
simulate
NOTE: In FRK >2.0 simple_kriging_fixed = FALSE by default, and hence
universal kriging is done by default. However this is only the case when
method = 'TMB'. When method = 'EM', simple kriging is done, irrespective of
what the argument simple_kriging_fixed is set to.
|
| | 0%
|
|=======================
...[truncated]...
|======================= | 33%
|
|=============================================== | 67%
|
|======================================================================| 100%
Maximum EM iterations reached
Averaging over polygons...
NOTE: In FRK >2.0 simple_kriging_fixed = FALSE by default, and hence
universal kriging is done by default. However this is only the case when
method = 'TMB'. When method = 'EM', simple kriging is done, irrespective of
what the argument simple_kriging_fixed is set to.
|
| | 0%NOTEdependencies in R coder-devel-linux-x86_64-debian-clang, r-devel-linux-x86_64-debian-gcc, r-devel-linux-x86_64-fedora-clang, r-devel-linux-x86_64-fedora-gcc, r-devel-windows-x86_64
r-devel-linux-x86_64-debian-clang, r-devel-linux-x86_64-debian-gcc, r-devel-linux-x86_64-fedora-clang, r-devel-linux-x86_64-fedora-gcc
Base package in Suggests/Enhances imported in NAMESPACE: ‘parallel’
r-devel-windows-x86_64
Base package in Suggests/Enhances imported in NAMESPACE: 'parallel'
- ERROR8 OK, 3 NOTE, 2 ERRORNow
- ERROR9 OK, 3 NOTE, 2 ERROR2026-03-10
Documentation
- Help pages
- 47
- Examples that run
- 79%26 run in a plain check · 3 never run · 4 hold no code
- Exported functions with examples
- 91%
- Documented parameters
- 94%
- Return-value docs
- 11%
- Help pages citing references
- 17%
Also includes 2 demo scripts (run with demo())
Downloads
Repository
- PRs merged (all time)
- 50%
- Median days to close, 50 most recently updated closed issues
- 2
- Median days to close, 50 most recently updated closed PRs
- 4
Stars over time
Forks over time
Releases over time
Issues over time
PRs over time
8 development-tooling and community-health practices detected across 4 families in the upstream repository
Checks run against github.com/andrewzm/frk on 2026-10-18.
The R CMD check workflow runs on Linux, macOS and Windows, and on R-devel
GitHub Pages site last deployed 2026-01-15 · The repository has a site deploy workflow
Dependencies
Installing FRK adds 116 packages (67 with compiled code, 117 MB of source tarballs). Heaviest: stringi 11 MB, sf 9.3 MB, ggplot2 6.0 MB.
Code & Tests
Datasets
People & History
Show 4 peopleHide 4 people
Pick two releases to compare their code metrics.
- RR 4.6.0 released · 2026-04-24
- 2.3.2Latest
- RR 4.5.0 released · 2025-04-11
- 2.3.12024-07-16 · diff ↗
- 2.3.02024-06-17 · diff ↗
- RR 4.4.0 released · 2024-04-24
- 2.2.32024-04-03 · diff ↗
- 2.2.22024-02-26 · diff ↗
- 2.2.12024-01-09 · diff ↗
- 2.2.0
- RR 4.3.0 released · 2023-04-21
- 2.1.52023-02-01 · diff ↗
- 2.1.32022-12-20 · diff ↗
- 2.1.22022-12-16 · diff ↗
- 2.1.0
- 2.0.62022-09-01 · diff ↗
Show 26 earlier eventsHide 26 earlier events
- RR 4.2.0 released · 2022-04-22
- 2.0.52022-03-25 · diff ↗
- 2.0.42022-02-18 · diff ↗
- 2.0.32022-01-10 · diff ↗
- 2.0.22021-11-16 · diff ↗
- 2.0.1
- 2.0.0
- RR 4.1.0 released · 2021-05-18
- 1.0.02021-02-22 · diff ↗
- RR 4.0.0 released · 2020-04-24
- 0.2.2.12020-04-01 · diff ↗
- RR 3.6.0 released · 2019-04-26
- 0.2.22018-07-13 · diff ↗
- RR 3.5.0 released · 2018-04-23
- 0.2.1
- 0.1.82018-01-13 · diff ↗
- 0.1.7
- 0.1.62017-10-13 · diff ↗
- 0.1.5
- 0.1.4
- 0.1.3
- 0.1.22017-04-21 · diff ↗ ·
removed 2 exports
sp_to_ST_basis, timelineadded 2 exports
Basis_as_data.frame, TensorP_Basis_as_data.frame - RR 3.4.0 released · 2017-04-21
- 0.1.12017-03-07
- removedRemoved from CRAN2017-02-05correct copyright information was missing. Re-added 2017-03-07
- RR 3.3.0 released · 2016-05-03
Package metadata
- First published
- 2017-03-07
- Total releases
- 30 / 9 yrs
- License
- GPL (>= 2) OSI
- Additional repositories
- andrewzm.github.io
- Minimum R
- ≥ 3.5.0
- Bundled data
- 2.5 MB / 6 files
- Download size
- 4.4 MB
Cite
Cite this package
Run in R for the authors' entry:
citation("FRK")The package provides a citation file, so citation() returns the authors' own entry.
It points to 10.18637/jss.v098.i04, 10.18637/jss.v108.i10.
Cite the R Observatory
For a number measured here: a download total, a coverage figure, an archival date.
From data release v2026-10-21, which the citation names so these numbers can be found later. More on citing and the projects behind them.