Model diagnostics for infectious epidemics using data augmentation
Date: Thursday 28th May
Speaker: Dr. George Streftaris. Actuarial Mathematics and Statistics. School of Mathematical and Computer Sciences, Heriot-Watt University.
Time: 4pm
Location: Science East Romm E0.01
Abstract:
The dynamics of disease spread in epidemics, or system change in ecology, are often described through spatio-temporal compartmental models where important characteristics, such as the distribution of sojourn time in a particular infection state, or the spatial transmission kernel, can be represented using a range of different settings. This leads to issues of model assessment and choice within a certain class of models, which typically suffers from problems related to the incomplete nature of observed data. I will talk about model evaluation using methodology based on the properties of the so-called Bayesian latent residuals, which become available through data augmentation within Markov chain Monte Carlo estimation schemes. Comparison between candidate models is also considered using a latent likelihood ratio–type test that avoids common problems associated with Bayesian model choice. The methods are illustrated using simulations and an application involving ecological data.
Series: Statistics
Social Media Links