BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:The Dantzig selector for high dimensional statistical problems - V
 incent Rivoirard (Université Paris-Dauphine)
DTSTART:20110322T160000Z
DTEND:20110322T170000Z
UID:TALK29557@talks.cam.ac.uk
CONTACT:6743
DESCRIPTION:The Dantzig selector has been introduced by Emmanuel Candés a
 nd Terence Tao  in an outstanding paper that deals with prediction and var
 iable selection in the setting of the curse of dimensionality extensively 
 considered in statistics recently. Using sparsity assumptions\, variable s
 election performed by the Dantzig selector can improve estimation accuracy
  by effectively identifying the subset of important predictors\, and then 
 enhance model interpretability allowed by parsimonious representations. Th
 e goal of this talk is to present the main ideas of the paper by Candés a
 nd Tao and the remarkable results they obtained. We also wish to emphasize
  some of the extensions proposed in different settings and in particular f
 or density estimation considered in the dictionary approach. Finally\, con
 nections between the Dantzig selector and the popular lasso procedure will
  be also highlighted. 
LOCATION:Seminar Room 1\, Isaac Newton Institute for Mathematical Sciences
END:VEVENT
END:VCALENDAR