BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Bayesian Inference for Diffusions using Data Augmentation MCMC - D
 r Kostas Kalogeropoulos\, CUED
DTSTART:20080826T131500Z
DTEND:20080826T141500Z
UID:TALK13110@talks.cam.ac.uk
CONTACT:Rachel Fogg
DESCRIPTION:Diffusion driven models provide natural models for phenomena e
 volving continuously in time. They appear quite often in various applicati
 ons across a wide range of diverse scientific areas. The task of likelihoo
 d based inference for their parameters is particularly challenging as thei
 r infinite dimensional paths can only be observed at a finite number of po
 ints and the marginal likelihood for these observations is generally intra
 ctable. Things are further complicated by the presence of various observat
 ion regimes: the observations may be noisy\, partial\, from multiple poten
 tially conflicting sources of information\, they may correspond to functio
 nals of the diffusion etc. Data augmentation schemes\, implemented through
  MCMC\, provide a general unified framework that may in theory handle all 
 such cases. However\, extra care is required because usually a suitable re
 parametrisation is needed to avoid degenerate MCMC algorithms. This talk p
 resents the main features of this framework and illustrates the methodolog
 y through various examples.\n
LOCATION:LR5\, Engineering\, Department of
END:VEVENT
END:VCALENDAR