|
Activity Number:
|
33
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 2, 2009 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Survey Research Methods
|
| Abstract - #305696 |
|
Title:
|
An Investigation of Stratified Jackknife Estimators Using Simulated Establishment Data Under an Unequal Probability Sample Design
|
|
Author(s):
|
Philip Steel*+ and Victoria McNerney and John Slanta
|
|
Companies:
|
U.S. Census Bureau and U.S. Census Bureau and U.S. Census Bureau
|
|
Address:
|
5457 Nebraska Ave NW, Washington, DC, 20015,
|
|
Keywords:
|
variance estimation ; jackknife ; Tille ; Pareto ; fpc ; unequal probability sampling
|
|
Abstract:
|
Considering two different sampling schemes (Tille and Pareto), we present the results of a Monte Carlo simulation studying the statistical properties of several variance estimators on a synthetic data set, modeled on establishment data. We test the validity of including a varying finite population correction in the formulation of the stratified jackknife (as done with the Yates-Grundy-Sen estimator), and we compare the effects of direct replication of a rate to a Taylor linearization formulation.
|
