|
Activity Number:
|
215
- Evolving Survey Inference in the Big Data Era: Challenges and Opportunities
|
|
Type:
|
Invited
|
|
Date/Time:
|
Monday, July 29, 2019 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Survey Research Methods Section
|
|
Abstract #300223
|
Presentation
|
|
Title:
|
Small Area Estimation to Correct for Measurement Errors in Big Population Registers
|
|
Author(s):
|
Dano Ben-Hur* and Danny Pfeffermann
|
|
Companies:
|
Central Bureau of Statistics, Israel and Central Bureau of Statistics and Hebrew Unversity, Israel, University of Southampton, UK
|
|
Keywords:
|
Fay-Herriot model;
NMAR nonresponse;
Sample selection
|
|
Abstract:
|
As in many countries, Israel has a fairly accurate population register at the national level, which consists of about 9 million persons. However, the register is much less accurate for small domains, with an average domain enumeration error of about 13%. The main reason for the inaccuracy at the domain level is that people moving in or out an area are often late in reporting their change of address. In the presentation we shall discuss ways of correcting the estimates obtained from the register for small domains by mean of a survey taken from the register. In particular, we shall address the following questions: A- How to best draw the sample, B- How to combine the sample information with the register data, C- How to correct for informative non-response and D- How to assess the variance and bias of the resulting census estimators. Empirical illustrations based on data from the 2008 census will be shown.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2019 program
|
