Computing Conditional Density Functions and Expectations

56 Views Asked by At

Problem: Let the joint density function of $(X,Y)$ be $$f_{X,Y}(x,y)=\begin{cases}\dfrac{1}{y}e^{-x/y}e^{-y}&\text{if }x\in(0,\infty)\text{ and }y\in(0,\infty)\\[0.3em]0&\text{otherwise.}\end{cases}$$ (a) Find $f_Y(y)$ and $f_{X|Y}(x|y)$. Compute $E[Y].$
Attempt: We have for $y\in(0,\infty)$ that $$f_Y(y)=\int_{0}^\infty f_{X,Y}(x,y)\,dx=\int_0^\infty\frac{1}{y}e^{-x/y}e^{-y}\,dx=e^{-y},$$ and $f_y(y)=0$ otherwise. Then for all $x\in(0,\infty)$ and $y\in(0,\infty)$ we have $$f_{X|Y}(x|y)=\frac{f_{X,Y}(x,y)}{f_Y(y)}=\frac{1}{y}e^{-x/y}.$$ Finally, since $Y\thicksim\text{Exp}(1),$ we have that $E[Y]=1$.

(b) Find the conditional expectation $E[X|Y].$
Attempt: We have $$E[X|Y=y]=\int_0^\infty xf_{X|Y}(x|y)\,dx=\int_0^\infty \frac{x}{y}e^{-x/y}\,dx=y,$$ which implies that $E[X|Y]=Y.$

(c) Use (a) and (b) to find $E[X].$
Attempt: We have that $$E[X]=\int_0^\infty E[X|Y=y]f_Y(y)\,dy=\int_0^\infty yf_Y(y)\,dy=E[Y]=1.$$


Could anyone help me check if my approach is correct?
Thank you for your time and appreciate any feedback.

1

There are 1 best solutions below

0
On BEST ANSWER

Everything is correct. By the way, you can see from the density of $X|Y$ that $X|Y \sim \mathrm{Exp}(\mathtt{mean}=y) \therefore \mathsf{E}[X|Y=y] = y \ $ without having to do the integral. But anyway, everything looks good.