Using confidential data to publish statistics based on small samples is challenging due to privacy loss. This column introduces a simple method for dealing with this issue which adds noise to each statistic in proportion to its sensitivity to the addition or removal of a single observation from the data. The method generally outperforms widely used methods of disclosure limitation such as count-based cell suppression both in terms of privacy loss and statistical bias. As an illustration, the method is used to release estimates of social mobility by Census tract in the Opportunity Atlas.
Raj Chetty, John Friedman, 18 April 2019
Related
-
Andrew Bernard, Andreas Moxnes, Karen-Helene Ulltveit-Moe
Most Read
-
Lalive, Magesan, Staubli
-
Chronopoulos, Kampanelis, Oto-Peralías, Wilson
-
Cagé, Dagorret, Grosjean, Jha
-
Escobari, Levy Yeyati
-
Caffarra, Scott Morton
-
Eichengreen, O'Rourke
-
Burgess, Sievertsen
-
Mitze, Kosfeld, Rode, Wälde
-
Heldring, Robinson
-
Eichengreen
Blogs&Reviews
-
Sapir
-
Bouchaud, Farmer
-
Baldwin
-
Kende
-
Angeloni
Vox eBooks

Bartsch, Bénassy-Quéré, Corsetti, Debrun, 15 December 2020
Don't Miss
Bartsch, Bénassy-Quéré, Corsetti, Debrun
Scheuer
Bozio, Garbinti, Goupille-Lebret, Guillot, Piketty
Events
-
28 - 29 January 2021 / Online - Zoom / Timberlake Consultants
-
29 - 29 January 2021 / Online /
-
5 - 5 February 2021 / Online /
-
12 - 12 February 2021 / Online / Erasmus University Rotterdam, Tilburg University and the University of Amsterdam
-
12 - 12 February 2021 / Online /
CEPR Policy Research
-
Gobillon, Solignac
-
Giglio, Maggiori, Stroebel, Weber
-
Summers, Fatás
-
Favero, Galasso
-
Butt, Churm, McMahon, Morotz, Schanz
-
Eichengreen, Avgouleas, Poiares Maduro, Panizza, Portes, Weder di Mauro, Wyplosz, Zettelmeyer
-
Baldwin, Beck, Bénassy-Quéré, Blanchard, Corsetti, De Grauwe, den Haan, Giavazzi, Gros, Kalemli-Ozcan, Micossi, Papaioannou, Pesenti, Pissarides , Tabellini, Weder di Mauro
-
Baldwin, Nakatomi
-
Thimann
-
Goodhart, Perotti