BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Assessing simulator uncertainty using evaluations from several dif
 ferent simulators - House\, L (Virginia Tech)
DTSTART:20110908T130000Z
DTEND:20110908T133000Z
UID:TALK32721@talks.cam.ac.uk
CONTACT:Mustapha Amrani
DESCRIPTION:Any simulator-based prediction must take account of the discre
 pancy between the simulator and the underlying system. In physical systems
 \, such as climate\, this discrepancy has a complex\, unknown structure th
 at makes direct elicitation very demanding. Here\, we propose a fundamenta
 lly different framework to that currently in use and consider information 
 in a collection of simulator-evaluations\, known as a Multi-Model Ensemble
  (MME). We justify our approach both in terms of its transparency\, tracta
 bility\, and consistency with standard practice in\, say\, Climate Science
 . The statistical modelling framework is that of second-order exchangeabil
 ity\, within a Bayes linear treatment. We apply our methods based on a rec
 onstruction of boreal winter surface temperature.\n
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR