scientific article; zbMATH DE number 3942592
[For the entire collection see Zbl 0581.00015.] We show that a (pre-Gaussian) random variable X with values in a uniformly convex Banach space such that \(\lim_{t\to \infty}t^ 2P\{\| X\| >t\}=0\) satisfies the central limit theorem if and only if it satisfies the small balls condition: \[ (*)\quad \forall \epsilon >0,\quad \lim \inf_{n\to \infty}P\{\| S_ n\| /\sqrt{n}<\epsilon \}>0. \] The result improves in these spaces a result of \textit{M. Talagrand} [The Glivenko-Cantelli problem, Ann. Math., II. Ser. (to appear)] asserting that if \(E\| X\|^ 2<\infty\), (*) is necessary and sufficient for the central limit theorem in any Banach space. The proof rests on a new quadratic estimate for sums of independent random variables in smooth normed spaces. The result is expected to be true in any Banach space.
- Un critère sur les petites boules dans le théorème limite central. (A criterion on small balls in the central limit theorem)
- Central limit theorem in a Banach space
- scientific article; zbMATH DE number 3881556
- scientific article; zbMATH DE number 4011523
- Asymptotics of moments in the central limit theorem in Banach spaces
This page was built for publication:
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q3713237)