2000 character limit reached
Testing multivariate normality by testing independence (2311.11575v2)
Published 20 Nov 2023 in stat.ME and cs.LG
Abstract: We propose a simple multivariate normality test based on Kac-Bernstein's characterization, which can be conducted by utilising existing statistical independence tests for sums and differences of data samples. We also perform its empirical investigation, which reveals that for high-dimensional data, the proposed approach may be more efficient than the alternative ones. The accompanying code repository is provided at \url{https://shorturl.at/rtuy5}.
- Tests for multivariate normality—a critical review with emphasis on weighted l2superscript𝑙2l^{2}italic_l start_POSTSUPERSCRIPT 2 end_POSTSUPERSCRIPT statistics. TEST, 29(4):845–892, Dec 2020.
- N. Henze and B. Zirkler. A class of invariant consistent tests for multivariate normality. Communications in Statistics - Theory and Methods, 19(10):3595–3617, 1990.
- A new goodness of fit test for multivariate normality and comparative simulation study. Mathematics, 9(23):3003, Nov 2021.
- A kernel statistical test of independence. In NeurIPS, volume 20. Curran Associates, Inc., 2007.
- Kernels: Support Vector Machines, Regularization, Optimization, and Beyond. The MIT Press, 2018.
- A normality criterion for random vectors based on independence. Statistics & Probability Letters, 33(2):159–165, 1997.
- V.P. Skitovič. Linear combinations of independent random variables and the normal distribution law. Sel. Transl. Math. Stat. Probab., 2:211–228, 1962.
- Karl Pearson. Liii. on lines and planes of closest fit to systems of points in space. The London, Edinburgh, and Dublin philosophical magazine and journal of science, 2(11):559–572, 1901.
- M. Kac. On a characterization of the normal distribution. American Journal of Mathematics, 61(3):726–728, 1939.
- An Information-Theoretic Proof of the Kac-Bernstein Theorem, 2022.
- Measuring and testing dependence by correlation of distances. The Annals of Statistics, 35(6):2769 – 2794, 2007.
- Large sample analysis of the median heuristic. arXiv: Statistics Theory, 2017.
- Measuring statistical dependencies via maximum norm and characteristic functions. arXiv preprint arXiv:2208.07934, 2022.
- D. H.-Lobato, P. M.-Mombiela and D.L.-Paz and A. Suárez, Non-linear causal inference using gaussianity measures. Journal of Machine Learning Research, 17(28):1–39, 2016.