0
$\begingroup$

Suppose that $\left(X_n\right)_{n\in\mathbb{N}}$ are independent identically distributed random variables each following uniform distribution on $\left[0,1\right]$. Apply Central Limit Theorem to determine the limit $$\lim_{n\to\infty}\mathbb{P}\left(\left(X_1\cdot X_2\cdot\cdots\cdot X_n\right)^{n^{-\frac{1}{2}}}e^{n^{\frac{1}{2}}}\in\left[a,b\right]\right),\quad a,b\in\mathbb{R}_+\,.$$

Thanks !

$\endgroup$
0
$\begingroup$

Taking logs and denoting by $S_n$ the sums $S_n := \sum_{i=1}^n \log X_i$ we must find the limit of

$$P\left (\frac{S_n}{n^{1/2}} + n^{1/2} \in [\log a, \log b] \right).$$

Now let $Y:= \log X$. We have

$$E(Y) = \int_0^1 \log x dx = -1 $$

and

$$E(Y^2) = \int_0^1 (\log x)^2 dx = 2$$

whence

$$\mathrm{var} (Y) = 1$$

so that in the usual notation $\sigma(Y)=1$. Hence

$$P\left (\frac{S_n}{n^{1/2}} + n^{1/2} \in [\log a, \log b] \right)= P\left (\frac{S_n - (-n)}{1 \cdot n^{1/2}} \in [\log a, \log b] \right).$$

I have written this in the usual form required for the CLT. The variances being finite, and the variables in the sum iid we conclude that the limit is

$$\frac{1}{\sqrt{2\pi}}\int_{\log a}^{\log b} e^{-t^2/2}dt. $$

$\endgroup$

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.