The random variables $X$ and $Y$ have joint density:

$$ f_{X,Y}(x,y) = \begin{cases} 2-x-y,& 0<x,y<1\\ 0,&\text{otherwise}. \end{cases} $$

My question is to find $\operatorname{Var}(X)$. I have tried taking the marginal density so that I can calculate the expected values, but I'm not sure how to deal with the infinities in the limits. I usually deal with infinities when there is an euler function, but I'm not sure how to approach it here. Thanks in advance.

When I arrive at the evaluation of $2y - xy - \frac12 y^2$ from $-\infty$ to $1$, I'm not sure how to proceed.

  • $\begingroup$ The joint density function has not been described well. What values of $x,y$ are permitted? What does "if $0$" mean in this context? Can you explain what "infinities" you encountered? $\endgroup$ – hardmath Mar 29 '15 at 16:49
  • $\begingroup$ Sorry the formating messed up. $\endgroup$ – user2442317 Mar 29 '15 at 16:50
  • 1
    $\begingroup$ For how to format mathematics using LaTeX and MathJax on this site, see this meta post, MathJax basic tutorial and quick reference. $\endgroup$ – hardmath Mar 29 '15 at 16:52

I do your homework just because I would like you to see how to use MathJax.


To answer your question asked in a comment below:

Reason (a): You said so: " 0 < x, y < 1." What else is that supposed to mean?

Reason (b): Also, the double integral over the domain of the joint distribution has to equal $1$. It does:

$$\iint_{[0,1]\times[0,1]} (2-x-y)\ dy\ dx=\int_0^1\left[\int_0^1(2-x-y)\ dy \right]dx=\int_0^1(2-x-\frac{1}{2})\ dx=\left[2x-\frac{1}{2}x^2-\frac{1}{2}x\right]_0^1=2-\frac{1}{2}-\frac{1}{2}=1.$$

Reason (c): I did not want to guess, I used what I was told and looked reasonable...

The original answer:

The joint density is defined as follows: $$f_{X,Y}(x,y)=\begin{cases}2-x-y,&\text{ if }0\le x\le 1 \text{ and }0\le y\le 1\\ 0,& \text{ otherwise.} \end{cases}$$

So, there is no $\infty$ anywhere.

For the marginal density of $X$ we have, for $0\le x\le 1$, $$f_X(x)=\int_0^1f_{X,Y}(x,y)dy=\int_0^1(2-x-y)dy=\left[2y-xy-\frac{1}{2}y^2 \right]_0^1=2-x-\frac{1}{2},$$ and $0$ otherwise. Now, the expectation of $X$ and of $X^2$ are, by definition, $$E[X]=\int_0^1 xf_X(x)=\int_0^1 2x-x^2-\frac{1}{2}xdx=\left[x^2-\frac{1}{3}x^3-\frac{1}{4}x^2\right]_0^1=\frac{5}{12},$$ $$E[X^2]=\int_0^1 x^2f_X(x)=\int_0^12x^2-x^3-\frac{1}{2}x^2 dx=\left[\frac{2}{3}x^3-\frac{1}{4}x^4-\frac{1}{6}x^3\right]_0^1=\frac{1}{4}.$$ The variance is then $$\sigma^2=E[X^2]-E[X]^2=\frac{1}{9}.$$

  • $\begingroup$ How can you just assume those limits? $\endgroup$ – user2442317 Mar 29 '15 at 19:38
  • $\begingroup$ I edited my answer accordingly. $\endgroup$ – zoli Mar 29 '15 at 20:18
  • $\begingroup$ I might be missing some crucial property, but doesn't that mean that x goes from 0 to infinity, and likewise with y (- infinity to 1)? $\endgroup$ – user2442317 Mar 29 '15 at 20:31
  • $\begingroup$ The most important two properties of a density function are (1) its integral over the whole plane equals $1$. (2) The integral of the joint density, over every possible (measurable) planar sets, is positive. These two properties imply my (your) choice. $\endgroup$ – zoli Mar 29 '15 at 20:42
  • $\begingroup$ This makes sense. Thanks a lot! $\endgroup$ – user2442317 Mar 29 '15 at 20:51

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.