MTB Group Seminar Series: Bayesian Modelling of Complicated Systems

Abstract

Complicated systems can be fruitfully described using a combination of mathematical and statistical models in a Bayesian framework. In this presentation, I will describe a number of real-world challenges that can be addressed in this way. These examples include estimation of the within-field disease dynamics of a biosecurity plant pest using a novel stochastic model; informing management decisions for ecological networks using dynamic models calibrated to noisy time-series data; fitting covid trajectories using nonparametic Hawkes processes; and understanding the intrinsic dimension of basketball games based on video data. The computational methods used to implement these approaches will also be discussed. These include variations of Markov chain Monte Carlo (MCMC), Approximate Bayesian Computation (ABC) and Sequential Monte Carlo (SMC).
This work is joint with a range of authors who will be acknowledged in the presentation, with special acknowledgement to Abhishek Varghese, Matthew Adams, Raiha Browning and Edgar Santos-Fernandez.

Link to the session: https://zoom.us/j/91749344611?pwd=MEc2MXFDODdEWVFkZ1Nsd1FKZGZ2dz09