JSM 2015 Preliminary Program

Online Program Home
My Program

Abstract Details

Activity Number: 299
Type: Topic Contributed
Date/Time: Tuesday, August 11, 2015 : 8:30 AM to 10:20 AM
Sponsor: Survey Research Methods Section
Abstract #315734 View Presentation
Title: Calibration and Model-Robustness: How Close Can We Get to Full Efficiency?
Author(s): Thomas Lumley*
Companies: The University of Auckland
Keywords: efficiency ; case-control ; raking ; contiguity
Abstract:

Calibration of weights allows auxiliary information available for a whole population or cohort to be used in estimation for a subsample, without any additional model assumptions: the resulting estimators are design-consistent when the sampling weights are correctly specified even if a misspecified model is fitted. If the model is also correctly specified, more efficient estimators are typically available, but these are not typically design consistent. I will show that the efficiency gain of these estimators is not robust to minor model misspecification, and argue that calibration estimators are preferable when the sampling probabilities are known, with double-robust estimators being preferable when the sampling mechanism must be modelled.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2015 program





For program information, contact the JSM Registration Department or phone (888) 231-3473.

For Professional Development information, contact the Education Department.

The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.

2015 JSM Online Program Home