Statistical analysis of spatial and spatiotemporal data has become an indispensable component in modern day statistical research. Methodological developments are being pursued to address their manifestation in diverse areas of application including but not limited to public health, environment and ecology, climatology, computer vision. This session aims to bring in a cohort of five leading research minds in the field to showcase cutting edge developments in the field.

The boom of modern data recording systems has strewn spatially and temporally referenced data into the scientific research domain. Such data presents challenges that include—complex structure and dimensionality, dependence and computation. Solutions require novel methodological and computational frameworks to be developed. Research pursuits espouse computationally efficient Bayesian methodologies strengthening the connection with the real-world applications. This session is dedicated to bringing some of the best researchers working in the frontiers of such development within the domain of spatial and spatiotemporal modeling and analysis to the audience. The talks presented include Bayesian modeling and inference for large scale spatiotemporal  data leveraging mechanistic systems within state space models, data sketching methods to overcome challenges faced in devising Bayesian computation for spatial data, boundary analysis frameworks that leverage local geometric properties for spatial surfaces, statistical inference for Gaussian processes on compact Riemannian manifolds and Bayesian analysis and inferential frameworks for age-specific event data to obtain better posterior inference regarding age-standardized estimates.

Organiser: Dr Aritra Halder 

Chair: Dr Aritra Halder  

Speaker: Rajarshi Guhaniyogi 

Speaker: Didong Li 

Speaker: Aritra Halder 

Speaker:  Harrison Quick