On the strong law of large numbers and the law of the logarithm for weighted sums of independent random variables with multidimensional indices (Q1347079)

Versions of the strong law of large numbers and the law of iterated logarithm for weighted sums \(T_{\overline{n}} = \sum_{\overline {k} \leq \overline{n}} a_{\overline {n}, \overline{k}} X_{\overline {k}}\) of a field \(\{X, X_{\overline {n}}, \overline{n} \in N^d\}\) of i.i.d. centered random elements taking values in Banach space and a triangular array \(\{a_{\overline{n}, \overline{k}} : (\overline{n}, \overline {k}) \in N^d \times N^d\}\), where \(N^d\) is \(d\)- dimensional lattice with natural order, are obtained. One of the results (strong law of large numbers) is following. If \(\text{sup }a_{\overline {n}, \overline {k}} < \infty\), \( d \geq 2\) and \({\mathbf E}(|X|^p (\ln |X|)^{d-1}) < \infty\), then \(|\overline {n}|^{-1/p} T_{\overline{n}} \to 0\) a.s. iff \(|\overline n|^{-1/p}T_{\overline n}\to 0\) in probability. The effect of reducing the magnitude of the norming sequence \(|\overline {n} |^{1/p}\), to a sequence type \((|\overline{n}|\ln |\overline {n}|)^{1/2}\) is enquired. The cases of real random variables and random weights are also considered.