scientific article; zbMATH DE number 1336705

From MaRDI portal
Publication:4258783





In this paper the rate of convergence in the law of large numbers for Banach space valued random variables is studied. The author proves the following theorems. Theorem 1. Let \(X_1,X_2,\ldots\) be independent identically distributed \(B\)-valued random variables, \(S_n =X_1 + \cdots +X_n\). Let \(p \in [1,2)\) and suppose that \(E|X_i|^{2p} < \infty\), \(E X_i =0\) and \(B\) is of type \(p.\) Then \[ \sum_{n=1}^\infty P\bigl(|S_n |>n^{1 \over p}\varepsilon \bigl)<\infty\quad\text{for all}\quad\varepsilon >0, \] that is \(S_n/ n^{1\over q}\) converges completely to 0. Let \(\{ X_{nk}, k=1,\ldots,n, n=1,2,\ldots\}\) be a triangular array of row-wise independent \(B\)-valued random variables. This array is weakly mean dominated by the random variable \(X\) if for some \(\gamma >0\) \[ {1 \over n} \sum_{k=1}^n P(|X_{nk}|>x) \leq \gamma P(|X |>x) \] for all \(x>0\) and for all \(n\). Theorem 2. Let \(\{ X_{nk},k=1,\ldots,n, n=1,2\ldots\}\) be an array of row-wise independent \(B\)-valued random variables, \(S_n =\sum_{k=1}^n X_{nk}.\) Suppose that \(\{ X_{nk}\}\) is weakly mean dominated by \(X\) such that \(E|X|^p L(|X |^p)< \infty\), where \(0<p<2\) and \(L\colon [0,\infty) \to [0,\infty)\) is a monotone nondecreasing slowly varying function. Assume that \(B\) is of type \(p+\delta\) for some \(\delta >0\) (\(p+\delta \leq 2\)). Then \[ \sum_{n=1}^\infty n^{-1} L(n) P\bigl(|S_n |>n^{1 \over p}\varepsilon \bigl)< \infty \quad \text{for all}\quad \varepsilon >0. \] Moreover the author finds conditions under which \[ \sum_{n=1}^\infty n^{t\alpha-2} P\bigl(|S_n |> \varepsilon n^\alpha \bigl)< \infty \quad \text{for all}\quad \varepsilon >0. . \] For the entire collection see [Zbl 0896.00028].




Cited in
(26)








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 Q4258783)