The Question is:

Let ($X_{n})_{n≥1}$ be a sequence of i.i.d. Bernoulli random variables, on the same probability space, with parameter $\frac{1}2$ (P($X_{n}$ = 0) = P($X_{n}$ = 1) = $\frac{1}2$),and let $t_{n}$ be the hitting time of level n by the partial sums, i.e. $t_{n}$ = inf{k | $\mathbf{\Sigma^{k}_{m=1}}X_{m} = n$}. Show that $n^{-1}t_{n}$ converges to 2 almost surely.

My instinct here is to use the Strong Law of Large Numbers and use the

$lim_{n\rightarrow\infty} \ \frac{S_{n}}n = E({X_{1}}) = p\ \ a.s. $

But I am lost here. If we use the above, wouldn't I get that it converges to $\frac{1}2$ almost surely? Please help!


Notice that inter-arrival times $(t_i - t_{i-1})_{i=1}^{\infty}$ form an i.i.d. sequence of geometric random variables of parameter $p = 1/2$. Thus by SLLN,

$$ \frac{t_n}{n} \to \Bbb{E}[t_1] = \frac{1}{p} = 2 \quad \text{a.s.} $$


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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