The Statistical Computing and Visualisation section of the Statistical Society of Australia proudly presents the Venables Award seminar. This award is to encourage new open source software development from the Australian community with a view to support efforts to develop and share data science and statistics methodology.
Venables Award seminar
-
-
-
Wollongong Campus
Building 6 Room 210
This year's winners are UOW student Matthew Sainsbury-Dale and UOW staff Andrew Zammit-Mangion, with their software package FRK: Fixed Rank Kriging and the runner up, Rex Parsons, Robin Blythe, Adrian Barnett, Susannna Cramb, Steven McPhail, with predictNMB.
Presenters
Matthew Sainsbury-Dale
PhD Student, National Institute for Applied Statistics Research Australia, School of Mathematics and Applied Statistics, University of Wollongong
Andrew Zammit-Mangion
Associate Professor, National Institute for Applied Statistics Research Australia, School of Mathematics and Applied Statistics, University of Wollongong
Seminar Title: Modelling Big, Heterogeneous, Non-Gaussian Spatial and Spatio-Temporal Data using FRK
Abstract: FRK is a framework for spatial/spatiotemporal modelling and prediction in which a set of basis functions is used to model the underlying (latent) process of interest. The 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 integration 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 (2021) describe ‘FRK’ in a Gaussian setting, and detail its use of basis functions and BAUs, while Sainsbury-Dale et al. (2023) describe ‘FRK’ in a non-Gaussian setting; two vignettes are available that summarise these papers and provide additional examples.