Differentially private significance tests for regression coefficients
Preprint
Abstract
Many data producers seek to provide users access to confidential data without unduly compromising data subjects' privacy and confidentiality. When intense redaction is needed to do so, one general strategy is to require users to do analyses without seeing the confidential data, for example, by releasing fully synthetic data or by allowing users to query remote systems for disclosure-protected outputs of statistical models. With fully synthetic data or redacted outputs, the analyst never really knows how much to trust the resulting findings. In particular, if the user did the same analysis on the confidential data, would regression coefficients of interest be statistically significant or not? We present algorithms for assessing this question that satisfy differential privacy. We describe conditions under which the algorithms should give accurate answers about statistical significance. We illustrate the properties of the methods using artificial and genuine data.Keywords
All Related Versions
This publication has 14 references indexed in Scilit:
- A Course in Mathematical Statistics and Large Sample Theory2016
- Towards Unrestricted Public Use Business Microdata: The Synthetic Longitudinal Business DatabaseInternational Statistical Review, 2011
- Synthetic Datasets for Statistical Disclosure Control2011
- The Dataverse Network®: An Open-Source Application for Sharing, Discovering and Preserving DataD-Lib Magazine, 2011
- Verification servers: Enabling analysts to assess the quality of inferences from public use dataComputational Statistics & Data Analysis, 2009
- Calibrating Noise to Sensitivity in Private Data Analysis2006
- Categorical data regression diagnostics for remote access serversJournal of Statistical Computation and Simulation, 2005
- Releasing Multiply Imputed, Synthetic Public use Microdata: An Illustration and Empirical StudyJournal of the Royal Statistical Society Series A: Statistics in Society, 2004
- Model Diagnostics for Remote Access Regression ServersStatistics and Computing, 2003
- Weaving Technology and Policy Together to Maintain ConfidentialityJournal of Law, Medicine & Ethics, 1997