|
Activity Number:
|
313
- Statistical Models in Survey Sampling and Analysis
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2018 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Survey Research Methods Section
|
|
Abstract #329773
|
|
|
Title:
|
Cluster-Level Inference Under Element Sampling
|
|
Author(s):
|
Danhyang Lee* and Jae-kwang Kim and Chris Skinner
|
|
Companies:
|
Iowa State University and Iowa State University and London School of Economics and Political Science
|
|
Keywords:
|
Two-level model;
Element sampling;
EM algorithm
|
|
Abstract:
|
A two-level model is a useful tool for analyzing clustered data. When the sampling design for the clustered population is element sampling, then the classical estimation method for a two-level model can be biased. We propose a general estimation method for two-level models by incorporating some induced cluster-level weights due to sampling design, when sampling weights are only available at the element-level. It also uses an EM algorithm based on the approximate predictive distribution of the cluster-specific random effects. Some limited simulation studies demonstrate that the proposed method performs well for sufficiently large samples.
|
Authors who are presenting talks have a * after their name.
