BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:BSU Seminar: "The Lifebelt Particle Filter: a novel robust SMC sch
 eme" - Alice Corbella\, University of Warwick
DTSTART:20250211T140000Z
DTEND:20250211T150000Z
UID:TALK228301@talks.cam.ac.uk
CONTACT:Alison Quenault
DESCRIPTION:Sequential Monte Carlo (SMC) methods can be applied to discret
 e State-Space Models on bounded domains\, to sample from and marginalise o
 ver unknown random variables. Similarly to continuous settings\, problems 
 such as particle degradation can arise: proposed particles can be incompat
 ible with the data\, lying in low probability regions or outside the bound
 ary constraints\, and the discrete system could result in all particles ha
 ving weights of zero. In this talk I will introduce the Lifebelt Particle 
 Filter (LBPF)\, a novel SMC method for robust likelihood estimation in low
 -valued count problems. The LBPF combines a standard particle filter with 
 one (or more) lifebelt particles which\, by construction\, lie within the 
 boundaries of the discrete random variables\, and therefore are compatible
  with the data. The main benefit of the LBPF is that only one or few\, wis
 ely chosen\, particles are sufficient to prevent particle collapse. The L
 BPF can be used within a pseudo-marginal scheme to draw inferences on stat
 ic parameters\, θ\, governing the system. In the talk I will also present
  an example of the use of the LBPF for the estimation of the parameters go
 verning the death and recovery process of hospitalised patients during an 
 epidemic.
LOCATION:Large Seminar Room\, East Forvie Building\, Forvie Site Robinson 
 Way Cambridge CB2 0SR.
END:VEVENT
END:VCALENDAR