scientific article; zbMATH DE number 3998854

From MaRDI portal
Publication:4725401





Let \(H\) be a separable Hilbert space and \(X_ 1,X_ 2,\ldots\) i.i.d. random elements of \(H\), with symmetric distribution. Denote the sum \(X_ 1+X_ 2+\ldots+X_ n\) by \(S_ n\). \textit{Yu. V. Prokhorov} [Teor. Veroyatn. Primen. 28, 354--358 (1983; Zbl 0509.60011)] has shown an example where \[ \| S_ n\|^ 2b_ n^{-1}\to 1\quad\text{ in probability, as }n\to \infty, \tag{1} \] with an appropriate norming sequence \(b_ n\). This cannot occur in the finite dimensional case. In the present paper sufficient and (under further restrictions) necessary conditions are given for (1). Besides, the author proves that in (1) convergence in probability can never be strengthened to a.s. convergence, answering thus a question of Prokhorov in the negative.











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 Q4725401)