Conditional Expectation : Can't figure it out.

167 Views Asked by At

Given that $X$ and $Y$ are two independent random variables uniform on $[0, 1]$, and having the joint density:

$$f(x,y) = 2\,,\,\,\,\,\,\,0\leq x \leq y $$

Find $E[X\,|\,Y= y]$ and $E[Y\,|\,X = x]$.

Known Solution: $$E[X\,|\,Y= y] = \frac{Y}{2}$$ $$E[Y\,|\,X = x] = \frac{X+1}{2}$$

My Approach (which apparently is wrong):

Given $f(x,y) = 2$, we have:

$$f_X(x) = \int_0^1 2 dy = 2 \,\,\,\, \textrm{and}\,\,\,\, f_Y(y) = \int_0^y 2 dx = 2y$$

From this, we can compute:

$$p_{Y|X}(y|x) = \frac{f(x,y)}{f_X(x)} = \frac{2}{2} = 1 \,\,\,\,\textrm{and}\,\,\,\, p_{X|Y}(x|y) = \frac{f(x,y)}{f_Y(y)} = \frac{2}{2y} = \frac{1}{y}$$

Therefore we have:

$$E[X\,|\,Y = y] = \int_0^y x \cdot p_{X|Y}(x|y) dx = \int_0^y x\cdot \frac{1}{y} dx = \frac{1}{y}\left[\frac{x^2}{2}\right]_0^y = \color{green}{\frac{y}{2} \textrm{(which seems to be right)}}$$

and

$$E[Y\,|\,X = x] = \int_0^1 y \cdot p_{Y|X}(y|x) dy = \int_0^1 y \cdot 1\, dy = \color{red}{\frac{1}{2} \textrm{(WHICH IS WRONG!)}}$$

Can someone please explain to me what am I doing wrong? Thanks in advance!

1

There are 1 best solutions below

2
On BEST ANSWER

The answer to your question is simple: do not forget your indicator functions. More specifically, the marginal density of $X$ is wrong. To see why, just note that the joint density is $$ f(x,y)=2\mathbf{1}_{0<x<y<1} $$ and hence the marginal density of $X$ is given by $$ f_X(x)=\int_{\mathbb{R}} f(x,y)\,\mathrm dy=\int_x^1 2\,\mathrm dy,\quad 0<x<1. $$ The density of $Y$ is correctly specified though.