BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20250709T181751EDT-7188TxKfhU@132.216.98.100
DTSTAMP:20250709T221751Z
DESCRIPTION:CoCoLasso for High-dimensional Error-in-variables Regression\n
 \nMuch theoretical and applied work has been devoted to high-dimensional r
 egression with clean data. However\, we often face corrupted data in many 
 applications where missing data and measurement errors cannot be ignored. 
 Loh and Wainwright (2012) proposed a non-convex modification of the Lasso 
 for doing high-dimensional regression with noisy and missing data. It is g
 enerally agreed that the virtues of convexity contribute fundamentally the
  success and popularity of the Lasso. In light of this\, we propose a new 
 method named CoCoLasso that is convex and can handle a general class of co
 rrupted datasets including the cases of additive measurement error and ran
 dom missing data. We establish the estimation error bounds of CoCoLasso an
 d its asymptotic sign-consistent selection property. We further elucidate 
 how the standard cross validation techniques can be misleading in presence
  of measurement error and develop a novel corrected cross-validation techn
 ique by using the basic idea in CoCoLasso. The corrected cross-validation 
 has its own importance. We demonstrate the superior performance of our met
 hod over the non-convex approach by simulation studies.\n\n \n\n \n
DTSTART:20160930T193000Z
DTEND:20160930T203000Z
LOCATION:room 1205\, Burnside Hall\, CA\, QC\, Montreal\, H3A 0B9\, 805 rue
  Sherbrooke Ouest
SUMMARY:Hui Zou\, University of Minnesota
URL:/mathstat/channels/event/hui-zou-university-minnes
 ota-263154
END:VEVENT
END:VCALENDAR