The fortnightly BRAG meeting will be held this Thursday 04/05 at 1 pm via Zoom/GP-Y801. This week we will have presentations by @Dilishiya and Harry.
Zoom Link: https://qut.zoom.us/j/86982060024?pwd=dHFHdm44R2NjYWNEbWhwTmFWQ09PZz09
Password: brag@QUT (if prompted)
Dilishiya’s Talk
Title: Model-robust Bayesian design through Generalised Additive Models for monitoring submerged shoals
Abstract: Bayesian designs are generally found by maximising the expectation of a utility function with respect to the joint distribution of the parameters and the response conditional on an assumed model. Unfortunately, specifying such a model a priori can be difficult. To address this, we propose to find designs based on a class of models as defined by a Generalised Additive Model (GAM). The additive component of a GAM enables capturing discrepancies between what is assumed and the underlying data-generating process, thus providing robustness to model misspecification. Our approach is demonstrated initially on an exemplar design problem where a theoretic result is derived and used to explore the properties of optimal designs. Then we apply our approach to design future monitoring of a submerged shoal off the Western coast of Australia, and explore the robustness properties of the optimal designs. Further, we discuss how to use parallel processing architecture to implement the optimisation algorithm such that optimal designs can be obtained within a feasible time. The presentation concludes with a discussion of the results and provides some limitations which lead to suggestions for future research.
Harry’s Talk
Title: Bayesian Modelling of Cohort Time Series Data: An analysis of training and racing in sprint kayak
Abstract: Utilising Bayesian modelling techniques to predict an athlete’s true performance in training and to discern the effect of weather from the day-to-day athlete variability to better estimate how an athlete performance changes. By incorporating time series and latent variable elements, the Bayesian model will be used to identify whether an athlete undergoes a significant change in performance.
Bayesian Research & Applications Group 