BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Estimation and variable selection with Lasso and Dantzig Selector 
 in the high-dimensional regression model - Karim Lounici\, Université Par
 is 7
DTSTART:20081119T170000Z
DTEND:20081119T180000Z
UID:TALK15303@talks.cam.ac.uk
CONTACT:8047
DESCRIPTION:We derive $l_{\\infty}$ convergence rate simultaneously for La
 sso and Dantzig\nestimators in a high-dimensional linear regression model 
 under a mutual\ncoherence assumption on the Gram matrix of the design and 
 two different\nassumptions on the noise: Gaussian noise and general noise 
 with finite\nvariance. Then we prove that simultaneously the thresholded L
 asso and\nDantzig estimators with a proper choice of the threshold enjoy a
  sign\nconcentration property provided that the non-zero components of the
  target\nvector are not too small.\n
LOCATION:MR11\, CMS\, Wilberforce Road\, Cambridge\, CB3 0WB
END:VEVENT
END:VCALENDAR