|
Activity Number:
|
125
|
|
Type:
|
Luncheons
|
|
Date/Time:
|
Monday, August 9, 2004 : 12:30 PM to 1:50 PM
|
|
Sponsor:
|
Section on Government Statistics
|
| Abstract - #300606 |
|
Title:
|
Protecting Confidentiality by Releasing Simulated Microdata: A Discussion of Benefits, Limitations, and Practical Experiences - SOLD OUT
|
|
Author(s):
|
Jerome Reiter*+
|
|
Companies:
|
Duke University
|
|
Address:
|
ISDS, Durham, NC, 27708,
|
|
Keywords:
|
confidentiality ; disclosure ; multiple imputation ; synthetic data
|
|
Abstract:
|
Given the proliferation of information on individuals available in external databases, and the improvements in record linkage technologies, it is not inconceivable that statistical agencies and data collectors will be unable or unwilling to release genuine microdata in the near future. One solution is to release simulated data constructed so as to preserve many of the relationships in the real data but not reveal any confidential values. We will discuss at a nontechnical level the foundations of this approach, focusing on its benefits and limitations as compared to other disclosure limitation methods. We also will discuss practical experiences and issues with releasing simulated microdata. The goal of the discussions is to introduce statisticians in federal agencies or organizations that release public use data to this promising new approach to disclosure limitation. The lunch organizer, Jerry Reiter, is an assistant professor at Duke University. He has consulted with the U.S. Bureau of the Census on methods for generating and analyzing simulated microdata. He has published several articles on the topic in Journal of Official Statistics
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2004 program |
