BEGIN:VCALENDAR
PRODID:-//University of Illinois//Web Services Calendar//EN
VERSION:2.0
CALSCALE:GREGORIAN
BEGIN:VEVENT
DTSTAMP:20120214T133946Z
DTSTART;TZID=America/Chicago:20091112T160000
DTEND;TZID=America/Chicago:20091112T160000
SUMMARY:Statistics seminar: ``Improved HorvitzThompson Estimation of Mode
 l Parameters from Stratified Case-Control/Cohort Data" by Norman Breslow
 \, University of Washington
CREATED:20090401T120000Z
DESCRIPTION:Abstract: Stratified case-control and case-cohort studies inv
 olve simple random samplingfrom an infinite superpopulation at phase one
  and stratified sampling without replacement from a finite cohort at pha
 se two. The asymptotic variance of the Horvitz-Thompson (HT) estimator\,
  of parameters in a variety of (semi)-parametric models\, is the sum of 
 two components: the model-based variance of the MLE that would be calcul
 ated from complete data for the entire cohort\; and the design-based var
 iance from HT estimation of the unknown cohort total of the (efficient) 
 influence function contributions. The second component may be reduced by
  adjusting the sampling weights\, e.g. by calibration to known cohort to
 tals of auxiliary variables correlated with the influence function. Asym
 ptotic theory suggests that further reduction may be possible by calibra
 ting to within stratum totals of these same variables. We compare standa
 rd and adjusted HT estimates of log hazard ratios for coronary heart dis
 ease using case-cohort data from the Atherosclerosis Risk in Communities
  Study and software available in Lumley's R survey package.Similar compa
 risons are made using simulated case-cohort samples from the National Wi
 lms Tumor Study. The results suggest that standard case-cohort analyses\
 , as published today in the medical literature\, may entail a substantia
 l waste of information in comparison with the adjusted analyses. This is
  particularly true for hazard ratios associated with covariates known fo
 r the entire cohort. There are practical limits\, however\, on the numbe
 r of adjustment variables that may be accommodated.
LAST-MODIFIED:20091106T130000Z
LOCATION:Everitt Lab 165 (SPECIAL LOCATION)
CATEGORIES:Seminar
CONTACT:3-2167
ORGANIZER:office@illinois.edu
URL:http://illinois.edu/calendar/detail/1439?key=2000010120000101115396
UID:115396@illinois.edu
END:VEVENT
END:VCALENDAR