Synthetic Data and Risk Measures for Statistical Disclosure Control

> français

Collaborative Research Team Project 23: 2022–2025

Collaborative research and efficient data sharing have proven to be one of the cornerstones in our efforts to promote scientific discoveries. Many government and funding agencies have implemented new policies to encourage the practice of sharing research data. However, given the growing concern about disclosures and invasions of personal privacy from not only the research community but also public and private organizations, carrying out such policies can only happen when the subject’s identity can be well protected and the information in the data is faithfully preserved. 

Our CANSSI CRT proposal is a great opportunity to foster collaborations nationally and internationally, train HQP to satisfy the growing needs of Canadian organizations, and more generally raise awareness and interests about statistical research on data privacy in Canada. Our research proposal concentrates on the use of synthetic datasets for privacy protection. We tackle specifically three aspects of this endeavour, namely developing advanced methods for synthetic data generation, constructing sophisticated risk measures, and deriving novel inferential procedures for synthetic datasets that satisfy formal privacy guarantee.

Team leaders:

Bei Jiang – University of Alberta
Anne-Sophie Charest – Laval University

Collaborators:

Sébastien Gambs – Université du Québec à Montréal
Linglong Kong – University of Alberta
Jingchen (Monika) Hu – Vassar College
Adrian E. Raftery – University of Washington
Russell J. Steele – McGill University
Steven Thomas – Statistics Canada


Comments are closed.