BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:BSU Seminar: "Semiparametric posterior corrections" - Andrew Yiu\,
  University of Oxford
DTSTART:20240130T140000Z
DTEND:20240130T150000Z
UID:TALK209506@talks.cam.ac.uk
CONTACT:Alison Quenault
DESCRIPTION:Suppose we wish to estimate a finite-dimensional parameter but
  we don't want to restrict ourselves to a finite-dimensional model. This i
 s called semiparametric inference. An exciting aspect of this paradigm is 
 that we might be able to leverage state-of-the-art machine learning algori
 thms to estimate our high-dimensional nuisance parameters and still obtain
  statistical guarantees (e.g. a 95% confidence interval). To achieve this\
 , however\, we will generally need to carefully tailor our inference to th
 e target estimand. This can be problematic for nonparametric Bayesian infe
 rence\, which focuses on good performance for the whole data-generating di
 stribution\, possibly at the expense of low-dimensional parameters of inte
 rest. To remedy this\, we introduce a simple\, computationally efficient p
 rocedure that corrects the marginal posterior of our target estimand\, yie
 lding a new debiased and calibrated one-step posterior.
LOCATION:MRC Biostatistics Unit\, East Forvie Building\, Forvie Site Robin
 son Way Cambridge CB2 0SR.
END:VEVENT
END:VCALENDAR