A Martingale Kernel Two-Sample Test
By: Anirban Chatterjee, Aaditya Ramdas
Potential Business Impact:
Finds differences between groups faster.
The Maximum Mean Discrepancy (MMD) is a widely used multivariate distance metric for two-sample testing. The standard MMD test statistic has an intractable null distribution typically requiring costly resampling or permutation approaches for calibration. In this work we leverage a martingale interpretation of the estimated squared MMD to propose martingale MMD (mMMD), a quadratic-time statistic which has a limiting standard Gaussian distribution under the null. Moreover we show that the test is consistent against any fixed alternative and for large sample sizes, mMMD offers substantial computational savings over the standard MMD test, with only a minor loss in power.
Similar Papers
Signature Maximum Mean Discrepancy Two-Sample Statistical Tests
Machine Learning (Stat)
Compares movement patterns to find fake data.
Integral-Operator-Based Spectral Algorithms for Goodness-of-Fit Tests
Methodology
Makes computer tests better at spotting fake data.
Kernel Two-Sample Testing via Directional Components Analysis
Methodology
Finds differences between groups of data.