BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Bayesian computation for partially observed S(P)DEs - Frank van de
 r Meulen (Vrije Universiteit Amsterdam)
DTSTART:20241128T133000Z
DTEND:20241128T142000Z
UID:TALK221554@talks.cam.ac.uk
DESCRIPTION:Let (X_t\, t\\ge 0) be defined as a solution to a stochastic (
 partial) differential equation. Suppose X is partially observed at n fixed
  time instances and that its dynamics are parametrised by a finite-dimensi
 onal parameter theta.&nbsp\;\nI will consider recent work on sampling late
 nt paths of the process and the parameter\, conditional on the data. The k
 ey idea behind the approach is to consider a change of measure on path spa
 ce that turns the forward process into the conditioned process. By approxi
 mating this change of measure we can obtain weighted samples from the cond
 itioned process. Making this idea precise turns out to be much more involv
 ed for SPDEs compared to the SDE case. Some numerical results will show st
 rengths and limitations of the methods proposed.\nJoint work with Thorben 
 Pieper-Sethmacher\, Moritz Schauer and Aad van der Vaart.&nbsp\;
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR