A Lyapunov-type bound in $R^d$

V. Yu. Bentkus

Teoriâ veroâtnostej i ee primeneniâ, Tome 49 (2004) no. 2, pp. 400-410

Résumé

Let $\fs X1n$ be independent random vectors taking values in $R^d$ such that ${E X_k =0}$ for all $k$. Write ${S=\fsu X1n}$. Assume that the covariance operator, say $C^2$, of $S$ is invertible. Let $Z$ be a centered Gaussian random vector such that covariances of $S$ and $Z$ are equal. Let $\mathscr{C}$ stand for the class of all convex subsets of $R^d$. We prove a Lyapunov-type bound for $\Delta =\sup_{A\in\mathscr{C}}|P\{S\in A\}-P\{Z\in A\}|$. Namely, ${\Delta \le c d^{1/4} \beta}$ with ${\beta =\fsu \beta 1n}$ and ${\beta_k= E |C^{-1}X_k|^3}$, where $c$ is an absolute constant. If the random variables ${\fs X1n}$ are independent and identically distributed and $X_k$ has identity covariance, then the bound specifies to ${\Delta \le c d^{1/4} E |X_1|^3/\sqrt{n}}$. Whether one can remove the factor $d^{1/4}$ or replace it with a better one (eventually by $1$), remains an open question.

Détail

Keywords: multidimensional, central limit theorem, Berry–Esseen bound, Lyapunov, dependence on dimension, nonidentically distributed.

Bibliographie

[1] Ball K., “The reverse isoperimetric problem for Gaussian measure”, Discrete Comput. Geom., 10:4 (1993), 411–420 | DOI | MR | Zbl

[2] Bentkus V., “On normal approximations, approximations of semigroups of operators, and approximations by accompanying laws”, Lithuanian Math. J. (to appear) http://www.mathematik.uni-bielefeld.de/fgweb/

[3] Bentkus V., “On the dependence of the Berry–Esseen bound on dimension”, J. Statist. Planning and Inference, 113:2 (2003), 385–402 | DOI | MR | Zbl

[4] Bentkus V. Yu., “O zavisimosti otsenki Berri–Esseena ot razmernosti”, Litov. matem. sb., 26:2 (1986), 205–210 | MR | Zbl

[5] Bhattacharya R. N., Ranga Rao R., Normal Approximation and Asymptotic Expansions, Krieger, Melbourne, FL, 1986, 291 pp. | MR | Zbl

[6] Götze F., “On the rate of convergence in the multivariate CLT”, Ann. Probab., 19:2 (1991), 724–739 | DOI | MR | Zbl

[7] Nagaev S. V., “O veroyatnostnykh i momentnykh neravenstvakh dlya zavisimykh sluchainykh velichin”, Teoriya veroyatn. i ee primen., 45:1 (2000), 194–202 | MR

[8] Nagaev S. V., “An estimate of the remainder term in the multidimensional central limit theorem”, Lecture Notes in Math., 550, 1976, 419–438 | MR | Zbl

[9] Nazarov F., On the asymptotic behavior of the maximal perimeter of convex set in $\mathbb R^n$ with respect to the Gaussian measure, A manuscript, 2001, 1–5 pp.

[10] Nazarov F., On the maximal perimeter of a convex set in $\mathbb R^n$ with respect to the Gaussian measure II (anisotropic case), A manuscript, 2001, 1–11 pp.

[11] Paulauskas V., Račkauskas A., Approximation Theory in the Central Limit Theorem. Exact Results in Banach Spaces, Kluwer, Dordrecht, 1989, 156 pp. | MR

[12] Sazonov V. V., Normal Approximations – Some Recent Advances, Springer, Berlin, New York, 1981, 105 pp. | MR

[13] Senatov V. V., Normal Approximations: New Results, Methods and Problems, VSP, Utrecht, 1998, 363 pp. | MR | Zbl

[14] Senatov V. V., “Odna otsenka metriki Levi–Prokhorova”, Teoriya veroyatn. i ee primen., 29:1 (1984), 108–113 | MR | Zbl

[15] Senatov V. V., “Neskolko otsenok skorosti skhodimosti v mnogomernoi predelnoi teoreme”, Dokl. AN SSSR, 254:4 (1980), 809–812 | MR