English

Cluster sets for partial sums and partial sum processes

Probability 2014-03-28 v1

Abstract

Let X,X1,X2,X,X_1,X_2,\ldots be i.i.d. mean zero random vectors with values in a separable Banach space BB, Sn=X1++XnS_n=X_1+\cdots+X_n for n1n\ge1, and assume {cn:n1}\{c_n:n\ge1\} is a suitably regular sequence of constants. Furthermore, let S(n)(t),0t1S_{(n)}(t),0\le t\le1 be the corresponding linearly interpolated partial sum processes. We study the cluster sets A=C({Sn/cn})A=C(\{S_n/c_n\}) and A=C({S(n)()/cn})\mathcal{A}=C(\{S_{(n)}(\cdot)/c_n\}). In particular, AA and A\mathcal{A} are shown to be nonrandom, and we derive criteria when elements in BB and continuous functions f:[0,1]Bf:[0,1]\to B belong to AA and A\mathcal{A}, respectively. When B=RdB=\mathbb{R}^d we refine our clustering criteria to show both AA and A\mathcal{A} are compact, symmetric, and star-like, and also obtain both upper and lower bound sets for A\mathcal{A}. When the coordinates of XX in Rd\mathbb{R}^d are independent random variables, we are able to represent A\mathcal {A} in terms of AA and the classical Strassen set K\mathcal{K}, and, except for degenerate cases, show A\mathcal{A} is strictly larger than the lower bound set whenever d2d\ge2. In addition, we show that for any compact, symmetric, star-like subset AA of Rd\mathbb{R}^d, there exists an XX such that the corresponding functional cluster set A\mathcal{A} is always the lower bound subset. If d=2d=2, then additional refinements identify A\mathcal{A} as a subset of {(x1g1,x2g2):(x1,x2)A,g1,g2K}\{(x_1g_1,x_2g_2):(x_1,x_2)\in A,g_1,g_2\in\mathcal{K}\}, which is the functional cluster set obtained when the coordinates are assumed to be independent.

Keywords

Cite

@article{arxiv.1403.6971,
  title  = {Cluster sets for partial sums and partial sum processes},
  author = {Uwe Einmahl and Jim Kuelbs},
  journal= {arXiv preprint arXiv:1403.6971},
  year   = {2014}
}

Comments

Published in at http://dx.doi.org/10.1214/12-AOP827 the Annals of Probability (http://www.imstat.org/aop/) by the Institute of Mathematical Statistics (http://www.imstat.org)

R2 v1 2026-06-22T03:35:50.387Z