Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

Let $U_1$, $U_2$ .. $U_n$ be identical and independent random variables distributed Uniform(0, 1). How can I find the cumulative distribution function of the conditional distribution of $U_{(n-1)}$ given $U_{(n)} = c$? Here, $U_{(n-1)}$ refers to the second largest of the aforementioned uniform random variables.

I know that I can find the unconditional distribution of $U_{(n-1)}$: It's just Beta(2, $n - 2 + 1$) or Beta(2, $n-1$) because the ith order statistic of uniforms is distributed Beta(i, $n - i + 1$). However, how do I find the conditional CDF of $U_{(n-1)}$ given that $U_{(n)} = c$??

share|improve this question

1 Answer 1

up vote 4 down vote accepted

To condition on $A=[U_{(n)}=c]$ is to condition on the event that one value in the random sample $(U_k)_{1\leqslant k\leqslant n}$ is $c$ and the $n-1$ others are in $(0,c)$. Thus, conditionally on $A$, the rest of the sample is i.i.d. uniform on $(0,c)$. In particular $U_{(n-1)}\lt x$ means that the $n-1$ values are in $(0,x)$, which happens with probability $(x/c)^{n-1}$. The conditional density of $U_{(n-1)}$ is $$ f_{U_{(n-1)}\mid A}(x)=(n-1)c^{-(n-1)}x^{n-2}\mathbf 1_{0\lt x\lt c}. $$ Thus, conditionally on $U_{(n)}$, $U_{(n-1)}$ is distributed as $\bar U_{(n-1)}\cdot U_{(n)}$, where $\bar U_{(n-1)}$ is independent of $U_{(n)}$ and distributed as $U_{(n-1)}$.

share|improve this answer

Your Answer

 
discard

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.