BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Adaptive delayed-acceptance pseudo-marginal random walk Metropolis
  - Sherlock\, C (Lancaster University)
DTSTART:20140423T135500Z
DTEND:20140423T143000Z
UID:TALK52127@talks.cam.ac.uk
CONTACT:Mustapha Amrani
DESCRIPTION:Co-authors: Andrew Golightly (Newcastle University) and Daniel
  Henderson (Newcastle University)\n\nDelayed-acceptance (DA) pseudo-margin
 al Metropolis-Hastings (MH) algorithms can be applied when it is computati
 onally expensive to calculate an unbiased estimate of the true posterior\,
  but a computationally cheap approximation is available. A first accept-re
 ject stage is applied\, with the cheap approximation substituted for the t
 rue posterior in the MH acceptance ratio. Only for those proposals which p
 ass through the first stage is the computationally expensive true posterio
 r (or unbiased estimate thereof) evaluated\, with a second accept-reject s
 tage ensuring that detailed balance is satisfied with respect to the inten
 ded posterior. A weighted average of all previous unbiased estimates of th
 e true posterior provides a generic approximation to the true posterior. I
 f only the k-nearest neighbours are used in the average then evaluation of
  the approximate posterior can be made computationally cheap provided that
  the points at which the posterior has been estimated unbiasedly are store
 d in a multi-dimensional binary tree\, similar to a KD-tree. \n
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR