BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:High-dimensional variable selection - Richard Samworth\, Statistic
 al Laboratory\, University of Cambridge
DTSTART:20140220T191500Z
DTEND:20140220T213000Z
UID:TALK47524@talks.cam.ac.uk
CONTACT:Peter Watson
DESCRIPTION:We live in the era of Big Data\, where it is often relatively 
 cheap to collect and store vast quantities of data.  While this presents u
 nprecendented opportunities to improve our scientific understanding and qu
 ality of life\, there are also substantial statistical challenges.  One of
  the most important of these is variable selection\, and it is a topic tha
 t has received enormous attention among statisticians over the last 15 yea
 rs or so.\n\nI will describe some methods that have been proposed\, and ou
 tline some of my own work\, in collaboration with Rajen Shah\, that aims t
 o improve the performance of any existing variable selection algorithm thr
 ough Complementary Pairs Stability Selection.\n
LOCATION:Department of Plant Sciences\, Downing Site\, Cambridge 
END:VEVENT
END:VCALENDAR