BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Nonparametric estimation under shape constraints - Piet Groeneboom
 \, Delft University
DTSTART:20131101T160000Z
DTEND:20131101T170000Z
UID:TALK47608@talks.cam.ac.uk
CONTACT:20082
DESCRIPTION:After pioneering work of (among others) Brunk\, Chernoff and P
 rakasa Rao\, summarized in the book of Barlow\, Bartholomew\, Bremner and 
 Brunk\, the field of isotonic regression and shape constrained inference t
 emporarily received less attention. But there was a revival of interest in
  the nineties of the preceding century because of several reasons.\nFirst\
 , there was analytic progress when it became clear how to compute the "Che
 rnoffian distribution"\, first studied by Chernoff in a study of an estima
 tor of the mode of a distribution. This arose from a study of the connecti
 on between Brownian motion with a parabolic drift and Airy functions. Seco
 nd\, the relevance of the theory for nonparametric estimates of distributi
 on functions in inverse problems became apparent\, in particular for decon
 volution and interval censoring models. And finally\, fast algorithms beca
 me available for computing the shape-constrained estimates. I will discuss
  all three angles to these problems\, with an emphasis on recent results a
 nd open problems.
LOCATION:MR12\,  Centre for Mathematical Sciences\, Wilberforce Road\, Cam
 bridge
END:VEVENT
END:VCALENDAR