BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:MRC Biostatistics Unit Seminars
SUMMARY:BSU Seminar: &quot\;Semiparametric posterior corre
 ctions&quot\; - Andrew Yiu\, University of Oxford
DTSTART;TZID=Europe/London:20240130T140000
DTEND;TZID=Europe/London:20240130T150000
UID:TALK209506AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/209506
DESCRIPTION:Suppose we wish to estimate a finite-dimensional p
 arameter but we don't want to restrict ourselves t
 o a finite-dimensional model. This is called semip
 arametric inference. An exciting aspect of this pa
 radigm is that we might be able to leverage state-
 of-the-art machine learning algorithms to estimate
  our high-dimensional nuisance parameters and stil
 l obtain statistical guarantees (e.g. a 95% confid
 ence interval). To achieve this\, however\, we wil
 l generally need to carefully tailor our inference
  to the target estimand. This can be problematic f
 or nonparametric Bayesian inference\, which focuse
 s on good performance for the whole data-generatin
 g distribution\, possibly at the expense of low-di
 mensional parameters of interest. To remedy this\,
  we introduce a simple\, computationally efficient
  procedure that corrects the marginal posterior of
  our target estimand\, yielding a new debiased and
  calibrated one-step posterior.
LOCATION:MRC Biostatistics Unit\, East Forvie Building\, Fo
 rvie Site Robinson Way Cambridge CB2 0SR.
CONTACT:Alison Quenault
END:VEVENT
END:VCALENDAR