|
Activity Number:
|
219
- Expanding Access to Administrative and Survey Data for Public Policy Decision-Making
|
|
Type:
|
Invited
|
|
Date/Time:
|
Tuesday, August 9, 2022 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Government Statistics Section
|
|
Abstract #320510
|
|
|
Title:
|
An Assessment of Privacy-Preserving Regression Analyses Within the Context of Validation Servers for Administrative Tax Data
|
|
Author(s):
|
Andres F. Barrientos* and Aaron R. Williams and Joshua Snoke and Claire McKay Bowen
|
|
Companies:
|
Florida State University and Urban Institute and RAND Corporation and Urban Institute
|
|
Keywords:
|
differential privacy;
validation server;
administrative data;
regression
|
|
Abstract:
|
In this talk, we will discuss the challenges of implementing differentially private algorithms for regression analyses within the context of validation servers for administrative tax data. We will present new methodological adaptations to existing differentially private algorithms to provide full inference. The talk will also include a study conducted to assess the accuracy of these new adaptations using several utility metrics and real administrative tax data obtained from the Internal Revenue Service Statistics of Income (SOI) Division.
|
Authors who are presenting talks have a * after their name.
