Let $X_1,X_2,\ldots,X_n$ be a collection of independent uniformly distributed random variables on the interval from $0$ to $\theta$. The question has three parts.

  1. Find the CDF of $F_{x_n}$(x) of $X_n = \max{\{X_1,\ldots,X_n}\}$ I know that distribution for the uniform distribution is $\frac{1}{b-a}$. In this case $b=\theta$ and $a=0$. So the pdf is $\frac{1}{\theta}$. Then the pdf for $X_n$ should be $\left(\frac 1 \theta \right)^n$.

  2. The PDF is then $n\left(\frac 1 \theta \right)^{n-1}$ .

  3. Part c asks to calculate the mean and variance for $X_n$. I'm confused on how to do this. Since it is a uniform distribution should I just use the uniform distribution pdf to calculate the expectation and variance?


There is some improper notation being used. Presumably something like $X_{(n)}$ is intended for the maximum of the $X_i$. We get rid of the potential confusion, and use the unfairly neglected letter $W$ to denote the maximum of the $X_i$.

Then the maximum of the $X_i$ is $\le W$, that is, $W\le w$ if and only if $X_i\le w$ for all $i$. For $0\lt w \lt \theta$, the probability that $X_i \le w$ is $\frac{w}{\theta}$. Thus by independence the probability that all the $X_i$ are $\le w$ is $\left(\frac{w}{\theta}\right)^n$. We conclude that $$F_W(w)=\Pr(W\le w)=\left(\frac{w}{\theta}\right)^n.$$

Differentiate. We find that the density function $f_W(w)$ is given by $$f_W(w)=\frac{1}{\theta^n} nw^{n-1}$$ (for $0\lt w\lt \theta$).

Now the mean and variance of $w$ can be found. The mean is $$\int_0^\theta w\frac{1}{\theta^n} nw^{n-1}\,dw$$ The integration is straightforward. We get $\frac{n}{n+1}\theta$.

For the variance, note that $\text{Var}(W)=E(W^2)-(E(W))^2$. So we need $E(W^2)$. This is $$\int_0^\theta w^2\frac{1}{\theta^n} nw^{n-1}\,dw.$$

  • $\begingroup$ Thanks. The professors uses the $X_{1}$ and $X_{n}$ to denote the minimum and the maximum which is even more confusing since he's 1's looks like n. $\endgroup$ – Undergradstudent Apr 13 '14 at 0:22
  • $\begingroup$ What if w=theta? How would this proof go? $\endgroup$ – RaviTej310 Sep 22 '18 at 22:45

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.