The Statistical Society is pleased to invite you to the first episode of our new webinar series. Kicking off the series will be Noel Cressie, Distinguished Professor and Director, Centre for Environmental Informatics, University of Wollongong and 2014 Pitman Medal winner, speaking about 

Inference for Spatio-Temporal Changes of Arctic Sea Ice

on Tuesday, 17 September 2019 at 1pm EST.

Arctic sea-ice extent has been of considerable interest to scientists recently, mainly due to its decreasing trend over the past 20 years. In this talk, we propose a hierarchical spatio-temporal generalised linear model (GLM) for binary Arctic-sea-ice data, where data dependencies are introduced through a latent dynamic spatio-temporal linear mixed-effects model. By using a fixed number of spatial basis functions, the resulting model achieves both dimension-reduction and nonstationarity for spatial fields at different time points. An EM algorithm is proposed to estimate model parameters, and a (empirical) hierarchical-statistical-modelling approach is used to obtain the predictive distribution of latent spatio-temporal processes on the real line (y-scale) and, equivalently, on (0,1) (p-scale). This approach is applied to spatial binary Arctic-sea-ice data for each September over 20 years in the recent past, and several posterior summaries are computed in order to detect changes of Arctic-sea-ice extent. This is joint work with Dr Bohai Zhang of Nankai University, China.
Reference: Zhang, B. and Cressie, N. (2019). Estimating spatial changes over time of Arctic sea ice using hidden 2 x 2 tables, Journal of Time Series Analysis, 40, 288-311.

Professor Cressie's talk will be followed by the opportunity for participants to ask questions.

Attendance is free, but you will need to register here.

Kind regards

Marie-Louise Rankin 
SSA Executive Officer