Revisiting Differentially Private Hypothesis Tests for Categorical Data

Abstract

In this paper, we consider methods for performing hypothesis tests on data protected by a statistical disclosure control technology known as differential privacy. Previous approaches to differentially private hypothesis testing either perturbed the test statistic with random noise having large variance (and resulted in a significant loss of power) or added smaller amounts of noise directly to the data but failed to adjust the test in response to the added noise (resulting in biased, unreliable $p$-values). In this paper, we develop a variety of practical hypothesis tests that address these problems. Using a different asymptotic regime that is more suited to hypothesis testing with privacy, we show a modified equivalence between chi-squared tests and likelihood ratio tests. We then develop differentially private likelihood ratio and chi-squared tests for a variety of applications on tabular data (i.e., independence, sample proportions, and goodness-of-fit tests). Experimental evaluations on small and large datasets using a wide variety of privacy settings demonstrate the practicality and reliability of our methods.

References in corpus (3)

• Differentially Private Chi-Squared Hypothesis Testing: Goodness of Fit and Independence Testing
• Convergence Rates for Differentially Private Statistical Estimation
• Private False Discovery Rate Control

Cited by in corpus (18)

• Privacy-preserving parametric inference: a case for robust statistics
• Privately Learning High-Dimensional Distributions
• Differentially Private Testing of Identity and Closeness of Discrete Distributions
• A Primer on Private Statistics
• Local Private Hypothesis Testing: Chi-Square Tests
• Locally Private Hypothesis Selection
• Locally private non-asymptotic testing of discrete distributions is faster using interactive mechanisms
• Priv'IT: Private and Sample Efficient Identity Testing
• Private Identity Testing for High-Dimensional Distributions
• Structure and Sensitivity in Differential Privacy: Comparing K-Norm Mechanisms
• A Differentially Private Wilcoxon Signed-Rank Test
• Differentially private methods for managing model uncertainty in linear regression models
• Differentially private significance tests for regression coefficients
• INSPECTRE: Privately Estimating the Unseen
• Unbiased Statistical Estimation and Valid Confidence Intervals Under Differential Privacy
• Statistical Inference in the Differential Privacy Model
• Differentially Private Multivariate Statistics with an Application to Contingency Table Analysis
• Inference under Information Constraints III: Local Privacy Constraints