Limit Theorems for Self-Normalized Large Deviation
@article{EJP289,
author = {Qiying Wang},
title = {Limit Theorems for Self-Normalized Large Deviation},
journal = {Electron. J. Probab.},
fjournal = {Electronic Journal of Probability},
volume = {10},
year = {2005},
keywords = {Cram'er large deviation, limit theorem,},
abstract = {Let $X, X_1, X_2, \cdots $ be i.i.d. random variables with zero mean and finite variance $\sigma^2$. It is well known that a finite exponential moment assumption is necessary to study limit theorems for large deviation for the standardized partial sums. In this paper, limit theorems for large deviation for self-normalized sums are derived only under finite moment conditions. In particular, we show that, if $EX^4<\infty$, then
$$\frac {P(S_n /V_n \geq x)}{1-\Phi(x)} = \exp\left\{ -\frac{x^3 EX^3}{3\sqrt{ n}\sigma^3} \right\} \left[ 1 + O\left(\frac{1+x}{\sqrt {n}}\right) \right], $$
for $x\ge 0$ and $x=O(n^{1/6})$, where $S_n=\sum_{i=1}^nX_i$ and $V_n= (\sum_{i=1}^n X_i^2)^{1/2}$.
},
pages = {no. 38, 1260-1285},
issn = {1083-6489},
doi = {10.1214/EJP.v10-289},
url = {http://ejp.ejpecp.org/article/view/289}}