Skip to content

Research at St Andrews

Spatial modeling with R-INLA: a review

Research output: Contribution to journalReview article

DOI

Open Access permissions

Open

Standard

Spatial modeling with R-INLA : a review. / Bakka, Haakon; Rue, Håvard; Fuglstad, Geir-arne; Riebler, Andrea; Bolin, David; Illian, Janine; Krainski, Elias; Simpson, Daniel; Lindgren, Finn.

In: Wiley Interdisciplinary Reviews: Computational Statistics, Vol. Early View, e1443, 05.07.2018.

Research output: Contribution to journalReview article

Harvard

Bakka, H, Rue, H, Fuglstad, G, Riebler, A, Bolin, D, Illian, J, Krainski, E, Simpson, D & Lindgren, F 2018, 'Spatial modeling with R-INLA: a review' Wiley Interdisciplinary Reviews: Computational Statistics, vol. Early View, e1443. https://doi.org/10.1002/wics.1443

APA

Bakka, H., Rue, H., Fuglstad, G., Riebler, A., Bolin, D., Illian, J., ... Lindgren, F. (2018). Spatial modeling with R-INLA: a review. Wiley Interdisciplinary Reviews: Computational Statistics, Early View, [e1443]. https://doi.org/10.1002/wics.1443

Vancouver

Bakka H, Rue H, Fuglstad G, Riebler A, Bolin D, Illian J et al. Spatial modeling with R-INLA: a review. Wiley Interdisciplinary Reviews: Computational Statistics. 2018 Jul 5;Early View. e1443. https://doi.org/10.1002/wics.1443

Author

Bakka, Haakon ; Rue, Håvard ; Fuglstad, Geir-arne ; Riebler, Andrea ; Bolin, David ; Illian, Janine ; Krainski, Elias ; Simpson, Daniel ; Lindgren, Finn. / Spatial modeling with R-INLA : a review. In: Wiley Interdisciplinary Reviews: Computational Statistics. 2018 ; Vol. Early View.

Bibtex - Download

@article{1de7b86c7da344e2ad4483c3358a9646,
title = "Spatial modeling with R-INLA: a review",
abstract = "Coming up with Bayesian models for spatial data is easy, but performing inference with them can be challenging. Writing fast inference code for a complex spatial model with realistically‐sized datasets from scratch is time‐consuming, and if changes are made to the model, there is little guarantee that the code performs well. The key advantages of R‐INLA are the ease with which complex models can be created and modified, without the need to write complex code, and the speed at which inference can be done even for spatial problems with hundreds of thousands of observations. R‐INLA handles latent Gaussian models, where fixed effects, structured and unstructured Gaussian random effects are combined linearly in a linear predictor, and the elements of the linear predictor are observed through one or more likelihoods. The structured random effects can be both standard areal model such as the Besag and the BYM models, and geostatistical models from a subset of the Mat{\'e}rn Gaussian random fields. In this review, we discuss the large success of spatial modeling with R‐INLA and the types of spatial models that can be fitted, we give an overview of recent developments for areal models, and we give an overview of the stochastic partial differential equation (SPDE) approach and some of the ways it can be extended beyond the assumptions of isotropy and separability. In particular, we describe how slight changes to the SPDE approach leads to straight‐forward approaches for nonstationary spatial models and nonseparable space–time models.",
keywords = "Approximate Bayesian inference, Gaussian Markov random fields, Laplace approximations, Sparse matrices, Stochastic partial differential equations",
author = "Haakon Bakka and H{\aa}vard Rue and Geir-arne Fuglstad and Andrea Riebler and David Bolin and Janine Illian and Elias Krainski and Daniel Simpson and Finn Lindgren",
year = "2018",
month = "7",
day = "5",
doi = "10.1002/wics.1443",
language = "English",
volume = "Early View",
journal = "Wiley Interdisciplinary Reviews: Computational Statistics",
issn = "1939-5108",
publisher = "John Wiley and Sons",

}

RIS (suitable for import to EndNote) - Download

TY - JOUR

T1 - Spatial modeling with R-INLA

T2 - Wiley Interdisciplinary Reviews: Computational Statistics

AU - Bakka, Haakon

AU - Rue, Håvard

AU - Fuglstad, Geir-arne

AU - Riebler, Andrea

AU - Bolin, David

AU - Illian, Janine

AU - Krainski, Elias

AU - Simpson, Daniel

AU - Lindgren, Finn

PY - 2018/7/5

Y1 - 2018/7/5

N2 - Coming up with Bayesian models for spatial data is easy, but performing inference with them can be challenging. Writing fast inference code for a complex spatial model with realistically‐sized datasets from scratch is time‐consuming, and if changes are made to the model, there is little guarantee that the code performs well. The key advantages of R‐INLA are the ease with which complex models can be created and modified, without the need to write complex code, and the speed at which inference can be done even for spatial problems with hundreds of thousands of observations. R‐INLA handles latent Gaussian models, where fixed effects, structured and unstructured Gaussian random effects are combined linearly in a linear predictor, and the elements of the linear predictor are observed through one or more likelihoods. The structured random effects can be both standard areal model such as the Besag and the BYM models, and geostatistical models from a subset of the Matérn Gaussian random fields. In this review, we discuss the large success of spatial modeling with R‐INLA and the types of spatial models that can be fitted, we give an overview of recent developments for areal models, and we give an overview of the stochastic partial differential equation (SPDE) approach and some of the ways it can be extended beyond the assumptions of isotropy and separability. In particular, we describe how slight changes to the SPDE approach leads to straight‐forward approaches for nonstationary spatial models and nonseparable space–time models.

AB - Coming up with Bayesian models for spatial data is easy, but performing inference with them can be challenging. Writing fast inference code for a complex spatial model with realistically‐sized datasets from scratch is time‐consuming, and if changes are made to the model, there is little guarantee that the code performs well. The key advantages of R‐INLA are the ease with which complex models can be created and modified, without the need to write complex code, and the speed at which inference can be done even for spatial problems with hundreds of thousands of observations. R‐INLA handles latent Gaussian models, where fixed effects, structured and unstructured Gaussian random effects are combined linearly in a linear predictor, and the elements of the linear predictor are observed through one or more likelihoods. The structured random effects can be both standard areal model such as the Besag and the BYM models, and geostatistical models from a subset of the Matérn Gaussian random fields. In this review, we discuss the large success of spatial modeling with R‐INLA and the types of spatial models that can be fitted, we give an overview of recent developments for areal models, and we give an overview of the stochastic partial differential equation (SPDE) approach and some of the ways it can be extended beyond the assumptions of isotropy and separability. In particular, we describe how slight changes to the SPDE approach leads to straight‐forward approaches for nonstationary spatial models and nonseparable space–time models.

KW - Approximate Bayesian inference

KW - Gaussian Markov random fields

KW - Laplace approximations

KW - Sparse matrices

KW - Stochastic partial differential equations

U2 - 10.1002/wics.1443

DO - 10.1002/wics.1443

M3 - Review article

VL - Early View

JO - Wiley Interdisciplinary Reviews: Computational Statistics

JF - Wiley Interdisciplinary Reviews: Computational Statistics

SN - 1939-5108

M1 - e1443

ER -

Related by author

  1. Careful prior specification avoids incautious inference for log-Gaussian Cox point processes

    Sørbye, S., Illian, J. B., Simpson, D. P., Burlsem, D. & Rue, H., Apr 2019, In : Journal of the Royal Statistical Society: Series C (Applied Statistics). 68, 3, p. 543-564

    Research output: Contribution to journalArticle

  2. inlabru: an R package for Bayesian spatial modelling from ecological survey data

    Bachl, F. E., Lindgren, F., Borchers, D. L. & Illian, J. B., 21 Mar 2019, In : Methods in Ecology and Evolution. Early View, 7 p.

    Research output: Contribution to journalArticle

  3. Understanding species distribution in dynamic populations: a new approach using spatio‐temporal point process models

    Soriano-Redondo, A., Jones-Todd, C. M., Bearhop, S., Hilton, G. M., Lock, L., Stanbury, A., Votier, S. C. & Illian, J. B., 4 Mar 2019, In : Ecography. Early View

    Research output: Contribution to journalArticle

  4. Non-stationary Gaussian models with physical barriers

    Bakka, H., Vanhatalo, J., Illian, J. B., Simpson, D. & Rue, H., 18 Jan 2019, In : Spatial Statistics. In press

    Research output: Contribution to journalArticle

  5. Accounting for preferential sampling in species distribution models

    Pennino, M. G., Paradinas, I., Illian, J. B., Muñoz, F., Bellido, J. M., López-Quílez, A. & Conesa, D., 1 Jan 2019, In : Ecology and Evolution. 9, 1, p. 653-663 11 p.

    Research output: Contribution to journalArticle

ID: 254603596