BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Towards a turnkey approach to unbiased Monte Carlo estimation of s
 mooth functions of expectations - Francesca Romana Crucinio (King's Colleg
 e London)
DTSTART:20240719T123000Z
DTEND:20240719T133000Z
UID:TALK219073@talks.cam.ac.uk
DESCRIPTION:Given a smooth function $f$\, we develop a general approach to
  turn Monte Carlo samples with expectation $m$ into an unbiased estimate o
 f $f(m)$. Specifically\, we develop estimators that are based on randomly 
 truncating the Taylor series expansion of $f$ and estimating the coefficie
 nts of the truncated series. We derive their properties and propose a stra
 tegy to set their tuning parameters -- which depend on $m$ -- automaticall
 y\, with a view to make the whole approach simple to use. We develop our m
 ethods for the specific functions $f(x)=\\log x$ and $f(x)=1/x$\, as they 
 arise in several statistical applications such as maximum likelihood estim
 ation of latent variable models and Bayesian inference for un-normalised m
 odels.Detailed numerical studies are performed for a range of applications
  to determine how competitive and reliable the proposed approach is.\n&nbs
 p\;\n&nbsp\;
LOCATION:External
END:VEVENT
END:VCALENDAR