1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Homework Help: LaPlacian joint probability density function.

  1. May 7, 2013 #1
    A joint pdf is given as pxy(x,y)=(1/4)^2 exp[-1/2 (|x| + |y|)] for x and y between minus and plus infinity.

    Find the joint pdf W=XY and Z=Y/X.

    f(w,z)=∫∫f(x,y)=∫∫(1/4)^2*e^(-(|x|+|y|)/2)dxdy -∞<x,y<∞
    Someone told me I can not use Jacobian because of the absolute value. Is that true?
    So far this is what I have but I feel like I am not going anywhere.


    =(1/4)^2[∫[e^(-x/2)+∫e^(x/2))] * [∫[e^(-y/2)+∫e^(y/2))] the limits from -∞<x,y<0 and 0<x,y<∞

    =[(1/4)^2 ]*4*[ ∫ [e^(x/2)dx] + ∫ [e^(y/2)dy] ] 0<x,y<∞

    I have problems with transforming the limits for the new functions.
    Last edited: May 8, 2013
  2. jcsd
  3. May 8, 2013 #2

    Ray Vickson

    User Avatar
    Science Advisor
    Homework Helper

    Of course you can use the Jacobian, but you need to be careful to ensure that the variables therein do not cross zero; that is, |x| is perfectly well differentiable as long as x > 0 or x < 0. So, for positive w and z, what is the WY-region of {w < W < w+dw, z < Z < z+dz}? You ought to be able to find it as the union of two disjoint subregions, and in each subregion you can certainly use a Jacobian.

    Now you need to look at the other cases where w> 0 and z > 0 do not both hold.
  4. May 8, 2013 #3
    That is the part where I got stuck.
    What is the best way to solve this problem? should I use Jacobian or should I use the distribution properties?

    But my biggest question and where I need help is with the boundaries.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted