skip to content
 

Inference pipelines for nonlinear time series analysis applied to an emerging childhood infection

Date: 
Tuesday 20th August 2013 - 15:00 to 15:30
Venue: 
INI Seminar Room 1
Session Title: 
Data & Statistics: New Methods
Abstract: 
Using the case study of Hand Foot and Mouth Disease (HFMD) in Japan, we distill inference down to atomic, pipe-able plug-and-play methods for time series analysis with state space models. For HFMD, we find evidence for cross protection between two causative agents (EV71 and CA16) and herd immunity, as well as some evidence for reinfection. The inference pipeline method, coupled with cloud computing, provides the opportunity for full Bayesian model comparison to be conducted over a set of more than 3000 of mechanistic models.
The video for this talk should appear here if JavaScript is enabled.
If it doesn't, something may have gone wrong with our embedded player.
We'll get it fixed as soon as possible.
Presentation Material: 
University of Cambridge Research Councils UK
    Clay Mathematics Institute London Mathematical Society NM Rothschild and Sons