|
Activity Number:
|
561
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 3, 2016 : 10:30 AM to 11:15 AM
|
|
Sponsor:
|
Survey Research Methods Section
|
|
Abstract #321702
|
|
|
Title:
|
Calibration of Design Weights Using a Power Transformation
|
|
Author(s):
|
Sarjinder Singh* and Veronica Salinas and Stephen Sedory
|
|
Companies:
|
Texas A&M University - Kingsville and Texas A&M University - Kingsville and Texas A&M University - Kingsville
|
|
Keywords:
|
Empirical log-likelihood ;
Estimation of population total ;
Auxiliary information ;
Two-Step calibration ;
Searls' shrinkage
|
|
Abstract:
|
In this note, a new method of calibration of design weights has been suggested by following the two-step calibration approach of Singh and Sedory (2013, 2015). The traditional empirical log- likelihood estimator due to Owen (1998) is shown to be a special case of it. In addition the resultant new estimator is shown to have properties similar to the well known shrinkage estimator due to Searls (1964). The minimum mean squared error of the new proposed estimator is shown to be always less than that of the traditional empirical log likelihood estimator.
|
Authors who are presenting talks have a * after their name.
