Abstract
We present a general methodology to construct triplewise independent sequences of random variables having a common but arbitrary marginal distribution F (satisfying very mild conditions). For two specific sequences, we obtain in closed form the asymptotic distribution of the sample mean. It is non-Gaussian (and depends on the specific choice of F). This allows us to illustrate the extent of the ‘failure’ of the classical central limit theorem (CLT) under triplewise independence. Our methodology is simple and can also be used to create, for any integer K, new K-tuplewise independent sequences that are not mutually independent. For K [four.tf], it appears that the sequences created using our methodology do verify a CLT, and we explain heuristically why this is the case.
References
[1] Avanzi, B., G. Boglioni Beaulieu, P. Lafaye de Micheaux, F. Ouimet, and B. Wong (2021). A counterexample to the existence of a general central limit theorem for pairwise independent identically distributed random variables. J. Math. Anal. Appl. 499(1), 124982.10.1016/j.jmaa.2021.124982Search in Google Scholar
[2] Balbuena, C. (2008). Incidence matrices of projective planes and of some regular bipartite graphs of girth 6 with few vertices. SIAM J. Discrete Math. 22(4), 1351–1363.10.1137/070688225Search in Google Scholar
[3] Biggs, N. (1993). Algebraic Graph Theory. Second edition. Cambridge University Press.Search in Google Scholar
[4] Billingsley, P. (1995). Probability and measure. Third edition. Wiley, New York.Search in Google Scholar
[5] Böttcher, B., M. Keller-Ressel, and R. L. Schilling (2019). Distance multivariance: new dependence measures for random vectors. Ann. Statist. 47(5), 2757–2789.10.1214/18-AOS1764Search in Google Scholar
[6] Bradley, R. C. and A. R. Pruss (2009). A strictly stationary, N-tuplewise independent counterexample to the central limit theorem. Stochastic Process. Appl. 119(10), 3300–3318.10.1016/j.spa.2009.05.009Search in Google Scholar
[7] Chakraborty, S. and X. Zhang (2019). Distance metrics for measuring joint dependence with application to causal inference. J. Amer. Statist. Assoc. 114(528), 1638–1650.10.1080/01621459.2018.1513364Search in Google Scholar
[8] Diestel, R. (2005). Graph Theory. Third edition. Springer, Berlin.Search in Google Scholar
[9] Drton, M., F. Han, and H. Shi (2020). High-dimensional consistent independence testing with maxima of rank correlations. Ann. Statist. 48(6), 3206–3227.10.1214/19-AOS1926Search in Google Scholar
[10] Etemadi, N. (1981). An elementary proof of the strong law of large numbers. Z. Wahrsch. Verw. Gebiete 55(1), 119–122.10.1007/BF01013465Search in Google Scholar
[11] Fan, Y., P. Lafaye de Micheaux, S. Penev, and D. Salopek (2017). Multivariate nonparametric test of independence. J. Multivariate Anal. 153, 189–210.10.1016/j.jmva.2016.09.014Search in Google Scholar
[12] Gaunt, R. E. (2019). A note on the distribution of the product of zero-mean correlated normal random variables. Stat. Neerl. 73(2), 176–179.10.1111/stan.12152Search in Google Scholar
[13] Genest, C., J. G. Nešlehová, B. Rémillard, and O. A. Murphy (2019). Testing for independence in arbitrary distributions. Biometrika 106(1), 47–68.10.1093/biomet/asy059Search in Google Scholar
[14] Hall, P. and C. C. Heyde (1980). Martingale Limit Theory and its Application. Academic Press, New York.Search in Google Scholar
[15] Jin, Z. and D. S. Matteson (2018). Generalizing distance covariance to measure and test multivariate mutual dependence via complete and incomplete V-statistics. J. Multivariate Anal. 168, 304–322.10.1016/j.jmva.2018.08.006Search in Google Scholar
[16] Kantorovitz, M. R. (2007). An example of a stationary, triplewise independent triangular array for which the CLT fails. Statist. Probab. Lett. 77(5), 539–542.10.1016/j.spl.2006.09.002Search in Google Scholar
[17] Pollard, D. (2002). A User’s Guide to Measure Theoretic Probability. Cambridge University Press.Search in Google Scholar
[18] Pruss, A. R. (1998). A bounded N-tuplewise independent and identically distributed counterexample to the CLT. Probab. Theory Relat. Fields 111, 323–332.10.1007/s004400050170Search in Google Scholar
[19] Rosenblatt, M. (1956). A central limit theorem and a strong mixing condition. Proc. Nat. Acad. Sci. U.S.A. 42(1), 43–47.10.1073/pnas.42.1.43Search in Google Scholar PubMed PubMed Central
[20] Takeuchi, K. (2019). A family of counterexamples to the central limit theorem based on binary linear codes. IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences E102.A(5), 738–740.10.1587/transfun.E102.A.738Search in Google Scholar
[21] Tanabe, K. and M. Sagae (1992). An exact Cholesky decomposition and the generalized inverse of the variance-covariance matrix of the multinomial distribution, with applications. J. R. Stat. Soc. Ser. B Stat. Methodol. 54(1), 211–219.10.1111/j.2517-6161.1992.tb01875.xSearch in Google Scholar
[22] Weakley, L. M. (2013). Some Strictly Stationary, N-tuplewise Independent Counterexamples to the Central Limit Theorem. Ph.D. Thesis, Indiana University.Search in Google Scholar
[23] Yao, S., X. Zhang, and X. Shao (2018). Testing mutual independence in high dimension via distance covariance. J. R. Stat. Soc. Ser. B. Stat. Methodol. 80(3), 455–480.10.1111/rssb.12259Search in Google Scholar
© 2021 Guillaume Boglioni Beaulieu et al., published by De Gruyter
This work is licensed under the Creative Commons Attribution 4.0 International License.
