Short Course Description
The course will cover methods for fitting geostatistical and spatial point process models to data obtained from surveys on which the whole region or population of interest is observed, as well as surveys on which observations are available only in a spatial sample from the region of interest, and surveys on which members of the population of interest are missed with some unknown probability. Inference for this last class of surveys will focus on distance sampling surveys. The workshop covers methods for spatial modelling problems in general, although the examples and exercises in the workshop will focus on ecological surveys.
Participants will be instructed in the use of the R package “inlabru”, developed by researchers at the Universities of St Andrews and Edinburgh, which is available here:
http://www.inlabru.org/
This package provides a flexible and convenient interface for spatial inference with the R-INLA package, without having to be familiar with the details of R-INLA syntax and structures, as well as a means of doing spatial inference when detection probability is not known.
Participants will be instructed in the use of the software on example datasets, with some time being devoted to practical computer exercises in which participants use the software.
Participants should bring their own laptop to the workshop.
Participants should have some experience using the R statistical software package, and be familiar with basic statistical methods to the extent of being comfortable at least with linear regression models and preferably be familiar with fitting generalized linear models, and have a basic understanding of Bayesian inference.
The course will provide an introduction to the analysis of spatial data with the R statistical software.
- Spatial data types
- Bayesian Inference and spatial modelling
- Introduction to INLA and inlabru
- Prior specification and pc-priors
- Continuous domain random field models
- Finite element meshes for approximating continuous space
- Log Gaussian Cox processes
- Model choice and model validation
- Spatial sampling: thinned point processes
- Distance sampling as a thinned Poisson process
- Multiple likelihoods
The course will include hands-on practicals. Practicals assume that the attendants will bring their own computers. Practicals will be done with the R software and will be based on the analysis of real datasets with R, R-INLA and inlabru.
All course materials (slides, R code and datasets) will be available on-line so that attendants can reproduce the examples by themselves.
References
- Yuan, Y, Bachl, FE, Lindgren, F, Borchers, DL, Illian, JB, Buckland, ST, Rue, H & Gerrodette, T 2017, Point process models for spatio-temporal distance sampling data from a large-scale survey of blue whales. Annals of Applied Statistics 11: 2270-2297. DOI: 10.1214/17-AOAS1078
- Illian, JB & Burslem, D 2017, Improving the usability of spatial point processes methodology: an interdisciplinary dialogue between statistics and ecology Advances in Statistical Analysis 101: 495-520. DOI: 10.1007/s10182-017-0301-8
- Rue, H, Riebler, A, Sørbye, SH, Illian, JB, Simpson, DP & Lindgren, FK 2017, Bayesian computing with INLA: a review. Annual Review of Statistics and its Application 4: 395-421. DOI: 10.1146/annurev-statistics-060116-054045
- Rue H., Martino S., Chopin N. (2009). Approximate Bayesian inference for latent Gaussian models using integrated nested Laplace approximations (with discussion). Journal of the Royal Statistical Society, Series B 71(2), 319–392.
- Blangiardo, M and Cameletti, M. (2015) Spatial and Spatio-temporal Bayesian Models with R–INLA. Wiley.
- Roger S. Bivand, Edzer J. Pebesma and Virgilio Gómez-Rubio. (2013) Applied Spatial Data Analysis with R. Springer.
Lecturer Background Information
|
Janine Illian
Dr Janine B Illian is a senior lecturer in statistics at the University of St Andrews. She held a Professor II position at the Norwegian University of Science and Technology, Trondheim, 2013-2016. Her work focuses on spatial point process methodology and she is the author of “Statistical Analysis and Modelling of Spatial Point Patterns” (Wiley, 2008), which has become a standard work on point process modeling since its publication. Her research profile focuses on the development of modern, realistically complex, spatial statistical methodology that is both computationally feasible and relevant to end-users. She aims at taking spatial point processes from the theoretical literature into the real world and encouraging statistical development by fostering strong relationships with the user community. Her research has impacted on spatial modeling and biodiversity research in the context of ecology, and she has diversified to applications in crime modeling, earthquake forecasting, environmental modeling and terrorism studies. |