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

Consider $$\zeta_n=\max_{1\le k \le n} X_k ,$$ where $X_k\sim \exp(1)$. Let $\eta_n=\frac{\zeta_n}{\ln(n)}$. What is the limit of $\eta_n$ as $n\to \infty$? (Convergence in distribution).

share|cite|improve this question

closed as off-topic by Davide Giraudo, Silvia Ghinassi, C. Falcon, Math1000, Daniel W. Farlow Jun 30 at 1:32

This question appears to be off-topic. The users who voted to close gave this specific reason:

  • "This question is missing context or other details: Please improve the question by providing additional context, which ideally includes your thoughts on the problem and any attempts you have made to solve it. This information helps others identify where you have difficulties and helps them write answers appropriate to your experience level." – Davide Giraudo, Silvia Ghinassi, C. Falcon, Math1000, Daniel W. Farlow
If this question can be reworded to fit the rules in the help center, please edit the question.

Well, knowing that $\eta_n$ is a maximum, you could try to write the event $[\eta_n\le x]$ for a fixed $x$ using only events $[X_k\le x]$. // Re your approach of MSE, let me venture two guesses: First, the present question is homework. Second, you ask questions here at a pace such that you do not have the time to accept the answers you get, even less to study and understand them. Am I right on both counts? – Did Sep 12 '11 at 9:56
I am thinking of $P[\zeta_n\le \ln(n)x]$, so we will have $F(\ln(n) x)$, where $\zeta_n$ has a distribution of $F(x)=(1-\exp(-\lambda x))^n$, then find the distribution for one n and take the limit. i have $\lim_{n\rightarrow\infty} (1-\exp(-x\ln(n)))^n=\lim_{n\rightarrow\infty}(1-(\exp(-\ln(n))^x)^n=\lim_{n\rig‌​htarrow\infty} (1-(\frac{1}{n})^x)^n$, but i am not even sure if this limit exists. for x>0, we have a limit of 1 x=0,we have a limit of 0. can we just definite x<0 to have a limit of 0 as well? – bear Sep 12 '11 at 14:11
The logarithm of the quantity of interest is $n\log(1-n^{-x})$. Perhaps you know the limit of $n\log(1-an^{-1})$ for every given $a$. Then you might want to show that the limit of $n\log(1-n^{-x})$ is $-\infty$ if $x<1$ and $0$ if $x>1$. Once you know that, what does that tell you on the limit in distribution you are after? – Did Sep 12 '11 at 14:26
we have F(x)=1 for x>1, and F(x)=0, for x<1 – bear Sep 12 '11 at 14:30
Sorry? Do you mean the limits? (At the moment, you consider F(log(n)x) which depends on n and is certainly not 1 for x>1 nor 0 for x<1.) Anyway, how do you prove this? – Did Sep 12 '11 at 14:34