Inference pipelines for nonlinear time series analysis applied to an emerging childhood infection
Seminar Room 1, Newton Institute
AbstractUsing 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.
If it doesn't, something may have gone wrong with our embedded player.
We'll get it fixed as soon as possible.