PANEL: Where Are Statistical Agencies Going?
Panelat Data Privacy: From Foundations to Applications (Simons Institute workshop, 2019) · 2019-03-04..03-08·
summary
Panel reports that traditional suppression and record‑swapping are revealed as vulnerable to reconstruction, prompting adoption of formal methods: Census applies differential privacy to the 2020 census with released code and built‑in verification tests. Administrative data access runs at scale—~1,200 users, ~700 projects, ~1,000 outputs/month with ~80% automated review—and legal use‑case limits block broader release. Demonstrations against other agencies’ simulated data expose risks, and agencies face tradeoffs on utility, privacy budget allocation and whether perturbation will be accepted.
AI-written summary · about
chapters (24)
- 0:00Introduction
- 2:50European Perspective
- 8:04Differential Privacy
- 15:25perturbative methods
- 28:20best practices
- 30:16data protection
- 30:59data privacy
- 31:23how to convince people
- 33:05target
- 33:43data collection
- 34:56reporting
- 35:23administrative data
- 36:24privacy
- 37:50survey data
- 38:20data systems
- 39:59EU table builder
- 41:44Microdata sharing
- 43:37Open data
- 44:07Collecting too much data
- 45:57Identifying government data
- 49:26Scandinavian countries
- 50:09Clinical researchers
- 52:00Data linkage
- 53:17Data sharing
related talks
01h 00mPractical Difficulties in the Application of Differential Privacy for the Release of Large-Scale...
01h 06mPanel on Interpretability in the Law
01h 00mPANEL: Statistical Theory, Privacy and Data Analysis
43mAnalyzing Administrative Data with Privacy Protection in Place
01h 20mUnlocking societal benefits of AI with privacy protective technologies.
29mPrivacy-preserving methods: Building secure projects
01h 04mData Hemorrhage, Inequality, and You...
01h 19mAcademic sessions 1.