Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Let $X_k,k\in N$ be independent identically distributed random variables, and suppose that for some $p\in (0;2)$ the sequence $\{\frac{1}{n^{1/p}}\sum_{k\leq n} X_k\}_{n\ge 1}$ is a.s. bounded. Prove that $X_1$ is in $L^p$.

The question cannot be that hard, but I got lost at the moment :(

share|improve this question

1 Answer 1

up vote 2 down vote accepted

Hints: Let $S_n=\sum\limits_{k\leqslant n}X_k$.

(1.) The event $[n^{-1/p}S_n\ \text{bounded}]$ is included in the event $[n^{-1/p}X_n\ \text{bounded}]$.

(2.) For each $x\gt0$, the series $\sum\limits_n\mathrm P(|X_1|\gt xn^{1/p})$ converges if and only if $X_1$ is in $L^p$.

(3.) By Borel-Cantelli, if the series $\sum\limits_n\mathrm P(|X_n|\gt xn^{1/p})$ diverges, then $\limsup\limits_nn^{-1/p}|X_n|\geqslant x$ almost surely.

(4.) Assume that $X_1$ is not in $L^p$, then use the contrapositive of (2.), then (3.), and finally (1.) to conclude that $n^{-1/p}S_n$ is not almost surely bounded.

share|improve this answer
    
+1 for nice argument! For (1), $$ \frac{X_n}{n^{1/p}} = \frac{S_n}{n^{1/p}} - \left(1-\frac{1}{n}\right)^{1/p}\frac{S_{n-1}}{(n-1)^{1/p}},$$ which shows that boundedness of $S_n/n^{1/p}$ indeed implies the boundedness of $X_n / n^{1/p}$. –  sos440 Aug 24 '12 at 15:49

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.