BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
X-WR-CALNAME:Data Science and Statistics Seminar
X-WR-TIMEZONE:Eastern Time (US & Canada)
BEGIN:VEVENT
DTSTAMP:20261008T193035Z
UID:tag:localist.com\,2008:EventInstance_46005704595407
DTSTART:20240404T193000Z
DTEND:20240404T203000Z
DESCRIPTION:Speaker: Prof. Lihu Xu\, University of Macau and Harvard Univer
 sity \n\nTitle: Robust heavy tailed statistical estimations\n\nAbstract: W
 e extend the well known Catoni's robust estimator\, which was originally d
 esigned to estimate the mean of the data with finite second moment\, to th
 e situation in which the data are heavy tailed and only have finite $\\bet
 a$-th moment with $\\beta \\in (1\,2]$. Inspired by the generalized Taylor
  expansion developed in the Stein's method for stable approximation\, we p
 ropose a new influence truncation function and obtain a mean estimator for
  heavy tailed data. This estimator enjoys the nice properties of Caton's e
 stimator and recovers his classical result as $\\beta \\uparrow 2$. Based 
 on the new influence function\, we put forward a general robust estimation
  framework\, which covers classical statistical estimations such as quanti
 le regression\, generalized linear models\, elastic net\, etc. Simulations
  confirm that our robust estimation outperforms other related estimations 
 when the data are heavy tailed. We will also consider a medium of mean est
 imator via DNN-based GAN. \n\nShort Bio: Dr. Lihu Xu received his PhD at I
 mperial College in 2008\, and is currently an associate professor in the D
 epartment of Mathematics at the University of Macau. From January 2024  to
  July 2024\, Dr. Lihu Xu is a visiting faculty in the Department of Statis
 tics at Harvard University. His research interests lie in applied stochast
 ic analysis\, high dimensional statistics\, and stochastic algorithms rela
 ted to machine learning. He has published more than 50 papers on journals 
 such as  Annals of Applied Probability\, Annals of Statistics\, Journal of
  Machine Learning Research\, Probability Theory and Related Fields\, Mathe
 matics of Operations Research\, etc.
GEO:35.957579;-83.925751
LOCATION:Ayres Hall\, Room G013
SUMMARY:Data Science and Statistics Seminar
URL;VALUE=URI:https://calendar.utk.edu/event/data-science-and-statistics-se
 minar
CATEGORIES:Lectures & Presentations
END:VEVENT
END:VCALENDAR