A New Class of Private Chi-Square Tests

2016/10/24 by Kifer, Daniel, Rogers, Ryan · 1 citation
#Cryptography and Security (cs.CR) #FOS: Computer and information sciences #FOS: Mathematics #Statistics Theory (math.ST)

paper · doi:10.48550/arxiv.1610.07662

Abstract

In this paper, we develop new test statistics for private hypothesis testing. These statistics are designed specifically so that their asymptotic distributions, after accounting for noise added for privacy concerns, match the asymptotics of the classical (non-private) chi-square tests for testing if the multinomial data parameters lie in lower dimensional manifolds (examples include goodness of fit and independence testing). Empirically, these new test statistics outperform prior work, which focused on noisy versions of existing statistics.

Cited by

• Fiducial Matching: Differentially Private Inference for Categorical Data

Related

• Differentially Private Chi-Squared Hypothesis Testing: Goodness of Fit\n and Independence Testing
• Local Private Hypothesis Testing: Chi-Square Tests
• Revisiting Differentially Private Hypothesis Tests for Categorical Data
• Privacy Aware Experimentation over Sensitive Groups: A General Chi Square Approach
• The Test of Tests: A Framework For Differentially Private Hypothesis Testing