Scribd
Upload
171 views

Jacobian

1. The document describes computing the Jacobian and inverse transformations to find the joint probability density function fz,w(Z,W) for new random variables z=x+y and w=x/(x+y) given the original joint pdf fx,y(X,Y)=6e^-(3X+2Y). 2. The inverse transformations are found to be x=zw and y=z(1-w). The Jacobian is computed to be 1/(zw+z(1-w)). 3. Substituting into the original pdf and Jacobian gives fz,w(Z,W)=6Ze^-(Z(W+2)). 4. The marginal pdfs are found

Uploaded by

das1717
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
171 views

Jacobian

1. The document describes computing the Jacobian and inverse transformations to find the joint probability density function fz,w(Z,W) for new random variables z=x+y and w=x/(x+y) given the original joint pdf fx,y(X,Y)=6e^-(3X+2Y). 2. The inverse transformations are found to be x=zw and y=z(1-w). The Jacobian is computed to be 1/(zw+z(1-w)). 3. Substituting into the original pdf and Jacobian gives fz,w(Z,W)=6Ze^-(Z(W+2)). 4. The marginal pdfs are found

Uploaded by

das1717
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

EECS 501

2-D EXAMPLE OF JACOBIAN METHOD

Fall 2001

Given: fx,y (X, Y ) = 6e(3X +2Y ) for X, Y 0; 0 otherwise (2-D exponential). Goal: Compute fz,w (Z, W ) for transformation z = x + y and w = x/(x + y ). 1. Compute the inverse transformation of the given one: z = z (x, y ) = x + y w = w(x, y ) = x/(x + y ) Inverse x = x(z, w) = zw y = y (z, w) = z (1 w)

since w = x/(x + y ) = x/z x = zw and y = z x = z zw = z (1 w). 2. Compute the Jacobian=determinant of the Jacobian matrix J: |J | = |det
z x w x z y w y

| = |det

y (x+y )2

1
x (x+y )2

|=|

1 x+y |

1 x+y .

since x, y 0 for this particular fx,y (X, Y ). 3. Substitute inverse transformation into fx,y (X, Y )/|J (X, Y )|: fz,w (Z, W ) = =
fx,y (X,Y ) |J (X,Y )| |X =x(Z,W ),Y =y (Z,W )

as dened above

fx,y (ZW,Z (1W )) 1/(ZW +Z (1W ))

= 6Ze(3ZW +2Z (1W )) = 6ZeZ (W +2)

for Z 0 and 0 W 1 since x, y 0 0 w 1. 4. If desired, compute marginal pdfs for z and/or w: fz (Z ) = 6Ze2Z fw ( W ) =
0 1 ZW e dW 0

= 6(e2Z e3Z ) for Z 0.


6 (W +2)2

6ZeZ (W +2) dZ =

for 0 W 1.

Check: Both marginal pdfs integrate to 1. Compare to exponential fx (X ), fy (Y ). Given: fx,y (X, Y ) = 9e(3X +3Y ) for X, Y 0; 0 otherwise (2-D exponential). Goal: Compute fz,w (Z, W ) for transformation z = x + y and w = x/(x + y ). Now get fz,w (Z, W ) = 9Ze3Z for Z 0 and 0 W 1. This is a 2nd -order Erlang or Gamma pdf in z ; a uniform pdf in w. Note that z and w are now independent random variables, unlike before.

You might also like