Raj Chetty, John Friedman, 18 April 2019

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. 

Vox Talks

Events

CEPR Policy Research