BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Computer model calibration with large nonstationary spatial output
 s: application to the calibration of a climate model - Serge Guillas (Univ
 ersity College London)
DTSTART:20180413T090000Z
DTEND:20180413T093000Z
UID:TALK103765@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:Bayesian calibration of computer models tunes unknown input pa
 rameters by comparing outputs to observations. For model outputs distribut
 ed over space\, this becomes computationally expensive due to the output s
 ize. To overcome this challenge\, we employ a basis representations of the
  model outputs and observations: we match these decompositions to efficien
 tly carry out the calibration.  In a second step\, we incorporate the nons
 tationary behavior\, in terms of spatial variations of both variance and c
 orrelations\, into the calibration. We insert two INLA-SPDE parameters int
 o the calibration. A synthetic example and a climate model illustration hi
 ghlight the benefits of our approach.
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR