Differentially Private Joint Independence Test
This work addresses privacy concerns in statistical testing for joint dependence, which is incremental by extending existing methods to private settings.
The paper tackles the problem of detecting joint dependence among multiple random vectors while preserving differential privacy, proposing a test based on the dHSIC metric with private permutation that achieves minimax optimal power across privacy regimes.
Identification of joint dependence among more than two random vectors plays an important role in many statistical applications, where the data may contain sensitive or confidential information. In this paper, we consider the the $d$-variable Hilbert-Schmidt independence criterion (dHSIC) in the context of differential privacy. Given the limiting distribution of the empirical estimate of dHSIC is complicated Gaussian chaos, constructing tests in the non-privacy regime is typically based on permutation and bootstrap. To detect joint dependence in privacy, we propose a dHSIC-based testing procedure by employing a differentially private permutation methodology. Our method enjoys privacy guarantee, valid level and pointwise consistency, while the bootstrap counterpart suffers inconsistent power. We further investigate the uniform power of the proposed test in dHSIC metric and $L_2$ metric, indicating that the proposed test attains the minimax optimal power across different privacy regimes. As a byproduct, our results also contain the pointwise and uniform power of the non-private permutation dHSIC, addressing an unsolved question remained in Pfister et al. (2018). Both numerical simulations and real data analysis on causal inference suggest our proposed test performs well empirically.