BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20260407T093057EDT-4133Uv6LWb@132.216.98.100
DTSTAMP:20260407T133057Z
DESCRIPTION:Sparse Penalized Quantile Regression: Method\, Theory\, and Alg
 orithm\n\nSparse penalized quantile regression is a useful tool for variab
 le selection\, robust estimation\, and heteroscedasticity detection in hig
 h-dimensional data analysis. We discuss the variable selection and estimat
 ion properties of the lasso and folded concave penalized quantile regressi
 on via non-asymptotic arguments. We also consider consistent parameter tun
 ing therein. The computational issue of the sparse penalized quantile regr
 ession has not yet been fully resolved in the literature\, due to non-smoo
 thness of the quantile regression loss function. We introduce fast alterna
 ting direction method of multipliers (ADMM) algorithms for computing the s
 parse penalized quantile regression. Numerical examples demonstrate the co
 mpetitive performance of our algorithm: it significantly outperforms sever
 al other fast solvers for high-dimensional penalized quantile regression.
 \n
DTSTART:20180223T203000Z
DTEND:20180223T213000Z
LOCATION:Room 1205\, Burnside Hall\, CA\, QC\, Montreal\, H3A 0B9\, 805 rue
  Sherbrooke Ouest
SUMMARY:Prof. Yuwen Gu Department of Statistics University of Connecticut
URL:https://www.mcgill.ca/mathstat/channels/event/prof-yuwen-gu-department-
 statistics-university-connecticut-285283
END:VEVENT
END:VCALENDAR