Current browse context:
stat
Change to browse by:
References & Citations
Statistics > Methodology
Title: Transparent Privacy is Principled Privacy
(Submitted on 15 Jun 2020 (v1), last revised 18 Sep 2022 (this version, v3))
Abstract: In a technical treatment, this article establishes the necessity of transparent privacy for drawing unbiased statistical inference for a wide range of scientific questions. Transparency is a distinct feature enjoyed by differential privacy: the probabilistic mechanism with which the data are privatized can be made public without sabotaging the privacy guarantee. Uncertainty due to transparent privacy may be conceived as a dynamic and controllable component from the total survey error perspective. As the 2020 U.S. Decennial Census adopts differential privacy, constraints imposed on the privatized data products through optimization constitute a threat to transparency and result in limited statistical usability. Transparent privacy presents a viable path toward principled inference from privatized data releases, and shows great promise toward improved reproducibility, accountability, and public trust in modern data curation.
Submission history
From: Ruobin Gong [view email][v1] Mon, 15 Jun 2020 16:30:29 GMT (80kb,D)
[v2] Tue, 26 Apr 2022 16:59:44 GMT (71kb,D)
[v3] Sun, 18 Sep 2022 18:02:14 GMT (895kb)
Link back to: arXiv, form interface, contact.