BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:A robust and adaptive estimator for regression II - Yannick Baraud
 \, Laboratoire J.A. Dieudonné\, Nice
DTSTART:20140624T090000Z
DTEND:20140624T093000Z
UID:TALK53102@talks.cam.ac.uk
CONTACT:37296
DESCRIPTION:Our purpose is to present a new method for adaptively estimati
 ng a regression\nfunction when little is known about the shape and scale o
 f the errors. For instance\, it\ncan cope with error distributions as diff
 erent as Gaussian\, Uniform\, Cauchy or even\nwith unimodal unbounded dens
 ities. In favorable cases and when the true\ndistribution belongs to the m
 odel\, the estimator is asymptotically equivalent to the\nM.L.E. and\, nev
 ertheless\, still behaves reasonably well when the model is wrong\,\neven 
 in cases for which the least-squares do not work. The assumptions that are
 \nneeded to get our results are rather weak\, in particular no moment cond
 ition is\nrequired on the errors\, and this is why the method can adapt to
  both the regression\nfunction\, the shape of the errors and their scale. 
 Moreover\, it appears that the\npractical results obtained by simulation a
 re surprisingly good as compared to more\nspecific estimators. The corresp
 onding paper is available on arXiv at\nhttp://arxiv.org/abs/1403.6057\nJoi
 nt work with Mathieu Sart.
LOCATION:Centre for Mathematical Sciences\, Meeting Room 2
END:VEVENT
END:VCALENDAR