BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Continuous inference for aggregated point process data - Dr Ben Ta
 ylor\, University of Lancaster
DTSTART:20171121T143000Z
DTEND:20171121T153000Z
UID:TALK81641@talks.cam.ac.uk
CONTACT:Alison Quenault
DESCRIPTION:We introduce new methods for inference with count data registe
 red on a set of aggregation units. Such data are omnipresent in epidemiolo
 gy due to confidentiality issues: it is much more common to know the count
 y in which an individual resides\, say\, than know their exact location in
  space. Inference for aggregated data has traditionally made use of models
  for discrete spatial variation\, for example conditional autoregressive m
 odels (CAR). We argue that such discrete models can be improved from both 
 a scientific and inferential perspective by using spatiotemporally continu
 ous models to directly model the aggregated counts. We introduce methods f
 or delivering (limiting) continuous inference with spatitemporal aggregate
 d count data in which the aggregation units might change over time and are
  subject to uncertainty. We illustrate our methods using real world exampl
 es and discuss the implementation of the methods on graphics processing un
 its\, which yields massive computational benefits.
LOCATION:Large Seminar Room\, 1st Floor\, Institute of Public Health\, Uni
 versity Forvie Site\, Robinson Way\, Cambridge
END:VEVENT
END:VCALENDAR