0
$\begingroup$

I have to compute $E(max(X,Y)|min(X,Y)=t)$ where X and Y are independent and have uniform distribution on $[0,1]$.

I made this task:

$E(max(X,Y)|min(X,Y)=t)=f(t)$

$E(max(X,Y)\mathbf{1}_{min(X,Y)>s})=Ef(min(X,Y))\mathbf{1}_{min(X,Y)>s}$

$\int_{s}^{1}x*2xdx=\int_{s}^{1}f(x)*(2-2x)dx$

$2s^2=f(s)*(2-2s)$

$f(s)=(2-2s)/2s^2$

But the answer is diffrent. Where did I mistake? Thanks in advance.

$\endgroup$
3
$\begingroup$

I think you computed your left-hand side incorrectly.

$$E[\max(X,Y) 1_{\min(X,Y) > s}] = \int_s^1 \int_x^1 y \, dy \, dx + \int_s^1 \int_y^1 x \, dx \, dy = \int_s^1 (1-t^2) \, dt$$

Setting the integrand equal to $f(t) 2(1-t)$ yields $f(t) = \frac{1+t}{2}$.


Alternate approach:

$$\begin{align}E[\max(X,Y) \mid \min(X,Y) = t] &= \int_0^\infty P(\max(X,Y) \ge u \mid \min(X,Y) = t) \, du \\ &= \int_0^t 1 \, du + \int_t^1 \frac{2(1-u)}{2(1-t)} \, du \\ &= t + \frac{(1-t)^2 / 2}{1-t} = t + \frac{1-t}{2} = \frac{1 + t}{2}. \end{align}$$

$\endgroup$

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.