BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:BSU Seminar: &quot\;A Regression Tree Approach to Missing Data&quo
 t\; - Professor Wei-Yin Loh\, Department of Statistics\, University of Wis
 consin\, Madison\, USA
DTSTART:20250513T130000Z
DTEND:20250513T140000Z
UID:TALK231928@talks.cam.ac.uk
CONTACT:Alison Quenault
DESCRIPTION:Analysis of data with missing values is arguably the hardest p
 roblem in statistics.  Statistical methods are often designed for complete
 ly observed data and are inapplicable if some values are missing.  Althoug
 h there are many techniques for imputation of missing values\, the statist
 ical properties of the resulting fitted models are unknown\, except in spe
 cial situations that require unverifiable and likely unjustifiable assumpt
 ions\, such as "missing at random" (MAR) and "no unobserved confounding".\
 n\nWe use a large dataset of electronic health records of Covid-19 patient
 s and a national consumer expenditure survey to show that (1) routine impu
 tation of missing data is inadvisable and even illogical\, as missingness 
 itself can contain useful information that imputation destroys and (2) pop
 ular imputation algorithms such as MICE are impractical when the amount of
  missing data is large. We also show how the GUIDE classification and regr
 ession tree method easily overcomes these difficulties. GUIDE is unique am
 ong tree algorithms in many respects\, including its ability to completely
  avoid imputation of missing data in predictor variables and to explicitly
  display the effects of missing values in its decision tree diagrams. Lite
 rature on GUIDE and its accompanying software may be obtained at https://p
 ages.stat.wisc.edu/~loh/guide.html.
LOCATION:Large Seminar Room\, East Forvie Building\, Forvie Site Robinson 
 Way Cambridge CB2 0SR.
END:VEVENT
END:VCALENDAR