BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Causal Inference Reading Group
SUMMARY:A powerful methodology for analyzing correlated hi
 gh dimensional data with factor models - Hongyuan 
 Cao (Florida State University)
DTSTART;TZID=Europe/London:20230609T153000
DTEND;TZID=Europe/London:20230609T170000
UID:TALK201949AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/201949
DESCRIPTION:Multiple testing under dependence is a fundamental
  problem in high-dimensional statistical inference
 . We use a factor model to capture the dependence.
  Existing literature with factor models imposes jo
 int normality on the data or requires tuning param
 eters to obtain robust inference. This paper looks
  at the problem differently by transposing approxi
 mate factor models. This allows heteroscedasticity
  and a more accurate estimation of the covariance 
 matrix of idiosyncratic errors by projections. We 
 construct factor-adjusted one-sample and two-sampl
 e test statistics of high-dimensional data. Extens
 ive simulation studies demonstrate the favorable p
 erformance of the proposed method over state-of-th
 e-art methods while controlling the false discover
 y rate\, even for heavy-tailed data. The robustnes
 s and tuning parameter-free features make the prop
 osed method attractive to practitioners.
LOCATION:MR1\, Centre for Mathematical Sciences\, Wilberfor
 ce Road\, Cambridge
CONTACT:
END:VEVENT
END:VCALENDAR