Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Let $\{x_n\}_{n=1}^{+\infty}$ independent identically distributed uniformly on $[a,b]$ random variables, $f_n=\max\{x_1, x_2,..x_n\}$, $g_n=\min\{x_1, x_2,..x_n\}$. Prove that for $n\frac{b-f_n}{b-a}$ weakly converges to exponential distribution with parameter $1$.

share|cite|improve this question
see if you can show it for the interval $[0,1]$ first. Also what's the use of defining $g_n$? – tom Oct 16 '12 at 10:02

Set $Y_n := \frac{n(b-f_n)}{b-a}$ and let $c < d$ be given. Then we have \begin{align*} P(c \le Y_n < d) &= P\bigl((b-a)c \le n(b-f_n) < (b-a)d\bigr) \\ &= P\left( b-\frac{(b-a)d}n < f_n \le b-\frac{(b-a)c}n\right)\\ &= P\left(f_n \le b-\frac{(b-a)c}n\right) - P\left(f_n \le b-\frac{(b-a)d}n\right)\\ &= \prod_{i=1}^n P\left(X_n \le b-\frac{(b-a)c}n\right) - \prod_{i=1}^n P\left(X_n \le b-\frac{(b-a)d}n\right)\\ \end{align*} If $c \ge 0$, the first product equals finally (i. e. for large enough $n$) \[ \left[\frac 1{b-a} \left(b-\frac{(b-a)c}n - a\right) \right]^n = \left(1 - \frac cn\right)^n \] and, as this implies $d > 0$, the second one equals in this case $(1 - \frac dn)^n$ So, for $[c,d) \subseteq [0,\infty)$, we have \[ P(c \le Y_n < d) \to \exp(-c) - \exp(-d) = \int_c^d \!\exp(-x)\, dx\] If now $d \le 0$ the second product and the are equal 1, hence \[ P(c \le Y_n < d) \to 0 \] in this case. If finally, $c < 0 < d$, then \[ P(c \le Y_n < d) \to 1 - \exp(-d) = \int_0^d \!\exp(-x)\, dx \] so, in each case \[ P(c \le Y_n < d) \to \int_c^d \chi_{[0,\infty)}(x)\exp(-x)\, dx \] that is $Y_n \to \operatorname{Exp}(1)$ weakly.

share|cite|improve this answer

Your Answer


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.