BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:BSU Seminar: &quot\;Graphical  models for missing data not at rand
 om: identification\, inference\, and imputation&quot\; - Ilya Shpitser\, J
 ohns Hopkins School University 
DTSTART:20260113T140000Z
DTEND:20260113T150000Z
UID:TALK242368@talks.cam.ac.uk
CONTACT:Alison Quenault
DESCRIPTION:Missing data is a pervasive problem in data analyses\, resulti
 ng in datasets that contain censored realizations of a target distribution
 . Many approaches to inference on the target distribution using censored o
 bserved data rely on missing data\n models represented as a factorization 
 with respect to a graph.  I describe a simple characterization of all ide
 ntified missing data models where the full data distribution factorizes wi
 th respect to a directed acyclic graph (DAG).  We show how statistical\n 
 inference may be performed within a maximum likelihood and semi-parametric
  frameworks in this class of models.  In addition\, we discuss how Markov
  restrictions in his model class naturally lead to an imputation procedure
  analogues to Gibbs sampling procedures\n for the missing at random model\
 , such as MICE and Amelia\, while allowing imputation even in high dimensi
 onal settings where many missingness patterns have no support.\n\nThis is 
 joint work with Rohit Bhattacharya\, Razieh Nabi\, Trung Phung\, and Kyle 
 Reese.
LOCATION:Large Downstairs Teaching Room\, East Forvie Building\, Forvie Si
 te Robinson Way Cambridge CB2 0SR.
END:VEVENT
END:VCALENDAR