0
$\begingroup$

Joint PDF of $(X,Y)$ is $f(x,y)=12xy(1-y)$ for $0 < x < 1, 0 < y < 1$ and $0$ elsewhere. Let $Z=XY^2$ and $W=Y$ be a joint transformation of $(X,Y)$.

Find the Jacobian, and hence the joint PDF.

Find the PDF of $Z=XY^2$ from the joint PDF of $(Z,W)$.

When I solve this I keep getting a divergent integral for the marginal pdf of $Z$, am I doing something wrong?

My Working:

The support of $Z$ and $W$ is $0< z <1$ and $0< w <1$

$v_2(z,w)=Y=W, Z=XY^2$ , $v_1(z,w)=X=Z/W^2$

$$J= \begin{pmatrix} dX/dZ & dX/dW\\ dY/dZ & dY/dW\\ \end{pmatrix} = \begin{pmatrix} 1/W^2 & -2Z/W^3\\ 0 & 1\\ \end{pmatrix}=1/W^2 $$

$f(z,w)=|J|.f(v_1(z,w),v_2(z,w))=(1/w^2)[(12z/w^2)(w)(1-w)]=(12z/w^3)(1-w)$

$f_1(z)=\int_0^1f(z,w)\,dw$, which is a divergent integral

$\endgroup$
4
  • 2
    $\begingroup$ It would also help if you show your work, whatever you have done so far. $\endgroup$ May 24, 2018 at 4:49
  • $\begingroup$ If you do not expose your attempts, we cannot distinguish a call for help from an attempt at having one's homework solved for free. $\endgroup$
    – Xi'an
    May 24, 2018 at 7:19
  • $\begingroup$ I added my working out :) $\endgroup$
    – Ricardo723
    May 24, 2018 at 10:36
  • 2
    $\begingroup$ Hint: for starters, check whether your claimed joint density for $X$ and $Y$ is a valid joint density. After fixing the error, re-do your work and check if the corrected joint density that you find for $Z$ and $W$ is a valid joint density. For example, where do make a note of the fact that if $Y = W= 0.5$, (say), then $Z$ cannot possibly exceed $0.25$, and so your expression for $f_{Z,W}$ cannot be valid for all $z, w \in (0,1)$? This restriction will affect your marginal pdf calculation. $\endgroup$ May 24, 2018 at 14:38

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.