English

Weak convergence of self-normalized partial sums processes

Probability 2013-06-21 v2

Abstract

Let {X,Xn,n1}\{X, X_n, n\geq 1\} be a sequence of independent identically distributed non-degenerate random variables. Put S0=0,Sn=i=1nXiS_0=0, S_n = \sum^n_{i=1} X_i and Vn2=i=1nXi2,n1.V_n^2=\sum^n_{i=1} X_i^2, n\ge 1. A weak convergence theorem is established for the self-normalized partial sums processes {S[nt]/Vn,0t1}\{S_{[nt]}/V_n, 0\le t\le 1\} when XX belongs to the domain of attraction of a stable law with index α(0,2]\alpha \in (0,2]. The respective limiting distributions of the random variables max1inXi/Sn{\max_{1\le i\le n}|X_i|}/{S_n} and max1inXi/Vn{\max_{1\le i\le n}|X_i|}/{V_n} are also obtained under the same condition.

Keywords

Cite

@article{arxiv.1204.2074,
  title  = {Weak convergence of self-normalized partial sums processes},
  author = {Miklós Csörgő and Zhishui Hu},
  journal= {arXiv preprint arXiv:1204.2074},
  year   = {2013}
}
R2 v1 2026-06-21T20:47:10.086Z