On strict sub-Gaussianity, optimal proxy variance and symmetry for bounded random variables

Arbel, Julyan; Marchal, Olivier; Nguyen, Hien D.

ESAIM: Probability and Statistics, Tome 24 (2020), pp. 39-55.

Suite au passage du modèle économique de la revue en S20, le texte intégral des articles des années 2019 et 2020 est accessible uniquement sur le site de l'éditeur et est réservé aux abonnés.

We investigate the sub-Gaussian property for almost surely bounded random variables. If sub-Gaussianity per se is de facto ensured by the bounded support of said random variables, then exciting research avenues remain open. Among these questions is how to characterize the optimal sub-Gaussian proxy variance? Another question is how to characterize strict sub-Gaussianity, defined by a proxy variance equal to the (standard) variance? We address the questions in proposing conditions based on the study of functions variations. A particular focus is given to the relationship between strict sub-Gaussianity and symmetry of the distribution. In particular, we demonstrate that symmetry is neither sufficient nor necessary for strict sub-Gaussianity. In contrast, simple necessary conditions on the one hand, and simple sufficient conditions on the other hand, for strict sub-Gaussianity are provided. These results are illustrated via various applications to a number of bounded random variables, including Bernoulli, beta, binomial, Kumaraswamy, triangular, and uniform distributions.

Classification : 97K50
Mots-clés : Sub-Gaussian, beta distribution, Kumaraswamy distribution, triangular distribution

[1] A. Ben-Hamou, S. Boucheron and M.I. Ohannessian, Concentration inequalities in the infinite urn scheme for occupancy counts and the missing mass, with applications. Bernoulli 23 (2017) 249–287. | DOI | MR | Zbl

[2] D. Berend and A. Kontorovich, On the concentration of the missing mass. Electr. Commun. Prob. 18 (2013) 1–7. | MR | Zbl

[3] S.G. Bobkov and F. Götze, Exponential integrability and transportation cost related to logarithmic Sobolev inequalities. J. Funct. Anal. 163 (1999) 1–28. | DOI | MR | Zbl

[4] S. Boucheron, G. Lugosi and P. Massart, Concentration inequalities: a nonasymptotic theory of independence. Oxford University Press, Oxford (2013). | DOI | MR | Zbl

[5] S. Bubeck and N. Cesa-Bianchi, Regret analysis of stochastic and nonstochastic multi-armed bandit problems. Found. Trends Mach. Learn. 5 (2012) 1–122. | DOI | Zbl

[6] V.V. Buldygin and I.V. Kozachenko, volume 188 of Metric characterization of random variables and random processes. American Mathematical Society, Providence, Rhode Island (2000). | DOI | MR | Zbl

[7] O. Catoni, PAC-Bayesian supervised classification: the thermodynamics of statistical learning. Vol. 56 of Monograph Series. Institute of Mathematical Statistics Lecture Notes (2007). | MR | Zbl

[8] M.C. Jones, Kumaraswamy’s distribution: A beta-type distribution with some tractability advantages. Stat. Methodol. 6 (2009) 70–81. | DOI | MR | Zbl

[9] M.J. Kearns and L.K. Saul, Large deviation methods for approximate probabilistic inference. In Proceedings of the Fourteenth Conference on Uncertainty in Artificial Intelligence UAI1 (1998).

[10] S. Kotz and J.R. Van Dorp, Beyond Beta: Other Continuous Families of Distributions with Bounded Support and Applications. World Scientific, Singapore (2004). | DOI | MR | Zbl

[11] P. Kumaraswamy, A generalized probability density function for double-bounded random processes. J. Hydrol. 46 (1980) 79–88. | DOI

[12] O. Marchal and J. Arbel, On the sub-Gaussianity of the beta and Dirichlet distributions. Electr. Commun. Probab. 22 (2017). | MR | Zbl

[13] D.A. Mcallester and L. Ortiz, Concentration inequalities for the missing mass and for histogram rule error. J. Mach. Learn. Res. 4 (2003) 895–911. | MR | Zbl

[14] H.D. Nguyen and G.J. Mclachlan, Progress on a conjecture regarding the triangular distribution. Commun. Stat. Theory Methods 46 (2017) 11261–11271. | DOI | MR | Zbl

[15] M. Raginsky and I. Sason, Concentration of measure inequalities in information theory, communications, and coding. Found. Trends Commun. Inf. Theory 10 (2013) 1–246. | DOI | Zbl

[16] M. Rudelson and R. Vershynin, Non-asymptotic theory of random matrices: extreme singular values. In Proceedings of the International Congress of Mathematicians 2010 (ICM 2010) (In 4 Volumes) Vol. I: Plenary Lectures and Ceremonies Vols. II–IV: Invited Lectures. World Scientific, Singapore (2010) 1576–1602. | MR | Zbl

[17] E. Schlemm, The Kearns-Saul inequality for Bernoulli and Poisson-binomial distributions. J. Theoret. Probab. 29 (2016) 48–62. | DOI | MR | Zbl

[18] R. Van Handel, Probability in high dimension. Technical report, Princeton University, 2014. | DOI