BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Bayesian inference for sparsely observed diffusions - Golightly\, 
 A (Newcastle University)
DTSTART:20140424T100500Z
DTEND:20140424T104000Z
UID:TALK52169@talks.cam.ac.uk
CONTACT:Mustapha Amrani
DESCRIPTION:Co-authors: Chris Sherlock (Lancaster University)\n\nWe consid
 er Bayesian inference for parameters governing nonlinear multivariate diff
 usion processes using data that may be incomplete\, subject to measurement
  error and observed sparsely in time. We adopt a high frequency imputation
  approach to inference\, by introducing additional time points between obs
 ervations and working with the Euler-Maruyama approximation over the induc
 ed discretisation. We assume that interest lies in the marginal parameter 
 posterior and sample this target via particle MCMC. A vanilla implementati
 on based on a bootstrap filter is eschewed in favour of an auxiliary parti
 cle filter where the latent path is extended by sampling a discretisation 
 of a conditioned diffusion. This conditioned diffusion should be carefully
  constructed to allow for nonlinear dynamics between observations.\n\n
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR