Transformation of two dimensional random var

AI Thread Summary
The discussion focuses on transforming a two-dimensional random variable defined by a specific probability density function. The user is seeking clarification on how to determine the integration boundaries for the transformation where z = y - x. They have derived some inequalities but are confused about how to proceed to find the correct limits for y and x. The solution suggests that the correct boundaries for y are from 0 to (2-2z)/3, and for x from y-z to 2-2y. Understanding the graphical representation of the defined region and the transformation process is crucial for deriving the marginal density function.
dannee
Messages
20
Reaction score
0
given the following Probability density function:

f x,y(x,y) = { 0.5, ( 0<y<1, 2y-x<2, 2y+x<2 }
0, else

and i need to find f z(z) while z=y-x

i got really confused while trying to calculate the borders of x and y for the integration.

i would be really thankful for someone explaining me how to come to the borders slow and step by step.

i got to the following equations: y < 2-z , y < (2+z)/3 , x < 2-z, x < (2-2z)/3

but i don't know how to resume from here. it said on the solution that borders are 0 to (2-2z)/3 for y and y-z to 2-2y for x. why?

thanks in advance for any help
 
Physics news on Phys.org
Hi dannee.
The inequalities (0<y<1, 2y-x<2, 2y+x<2) define a region A in the (x,y)-plane.
You can follow these steps, in order to graphically "visualize" the problem:

1) Draw the region A in the (x,y)-plane

2) Introduce the two dimensional transformation u=x , z=y-x

3) Obtain the density function g(u,z) of the two dimensional random variable (U,Z)

4) From the inequalities between (x,y) derive the corresponding inequalities between (u,z).

5) From the inequalities between (u,z) draw the region A' in the (u,z)-plane. A' is the image of A under the transformation introduced in step 2).

6) Finally, from g(u,z) derive the marginal density function g(z)
 

Similar threads

Back
Top