BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:A Concentration Inequality based methodology for Sparse Covariance
  Estimation - Adam Kashlak (University of Cambridge)
DTSTART:20170517T150000Z
DTEND:20170517T160000Z
UID:TALK72592@talks.cam.ac.uk
CONTACT:Nicolai Baldin
DESCRIPTION:In this talk\, we propose a general framework for covariance m
 atrix estimation making use of concentration inequality-based confidence s
 ets\, and we specify this framework for the estimation of large sparse cov
 ariance matrices. The usage of nonasymptotic dimension-free confidence set
 s yields good theoretical performance for such sparse estimators given rea
 sonable distributional assumptions. The proposed method merges past ideas 
 including shrinkage\, penalized\, and threshold estimators. Through extens
 ive simulations\, it is demonstrated to have superior performance when com
 pared with other such methods.
LOCATION:MR14\, Centre for Mathematical Sciences
END:VEVENT
END:VCALENDAR