Likelihood-Based Methods for Fitting Stochastic Epidemic Models to Noisy Data
Due to noisy observational data and nonlinear dynamics, even simple stochastic epidemic models such as the Susceptible-Infectious-Removed (SIR) present significant challenges to inference. In particular, evaluating the marginal likelihood of the data under such stochastic processes is a notoriously difficult task. As a result, likelihood-based inference is typically considered out of reach in the presence of missing data, and practitioners often resort to simulation methods or approximations that may bias conclusions and reduce interpretability of estimates. In this talk, Dr. Xu will discuss some recent contributions that enable "exact" inference under the model via the likelihood of observed data, focusing our attention on a perspective that makes use of latent variables to explore configurations of the missing data within a Markov chain Monte Carlo framework. Motivated both by count data from large outbreaks and high-resolution contact data from mobile health studies, this talk will show how our data-augmented approach successfully learns the interpretable epidemic parameters and scales to handle large realistic data settings efficiently.
Dr. Jason Xu is an Assistant Professor in the Department of Statistical Science at Duke University, with a secondary appointment by courtesy in Biostatistics and Bioinformatics. Prior to joining the faculty at Duke, Jason worked at the University of California Los Angeles with support from the NSF Mathematical Sciences Postdoctoral Research Fellowship. He completed hisPhD in Statistics at the University of Washington where his work was funded by an NDSEG Fellowship. Dr. Xu grew up in Tucson, where he received a BS in Mathematics from the University of Arizona in 2012.