BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Post-selection confidence intervals and confidence curves - Gerda 
 Claeskens\, Katholieke Universiteit Leuven 
DTSTART:20190208T160000Z
DTEND:20190208T170000Z
UID:TALK115915@talks.cam.ac.uk
CONTACT:Dr Sergio Bacallado
DESCRIPTION:By selecting variables or models via information criteria or o
 ther formal methods a single selected model is the winner. Often\, this wi
 nning model is wrongly treated as if it was known before the start of the 
 analysis that precisely this model would get selected. Randomness is invol
 ved with the selection. Indeed\, with a different sample of data another m
 odel could have been selected. For the popular method of the Akaike inform
 ation criterion (AIC)\, the asymptotic distribution of parameter estimator
 s after model selection is studied. The overselection property of this cri
 terion is exploited to construct a selection region\, and to obtain the as
 ymptotic distribution of parameter estimators and linear combinations ther
 eof in the selected model. The proposed method does not require the true m
 odel to be in the model set. We investigate the method in linear and gener
 alized linear models. Confidence curves provide a broader picture of the s
 election methods post-selection.\n\nThis is joint work with A. Charkhi and
  A. Garcia Angulo.
LOCATION:MR12
END:VEVENT
END:VCALENDAR