Conditinal Expectation Problem

  • Thread starter purplebird
  • Start date
  • #1
18
0

Main Question or Discussion Point

Given x,y and z are standard normal distributions with mean 0 and standard deviation 1. x,y and z are also statistically independent.

Find E{x|x+y+z=1}.
 

Answers and Replies

  • #2
111
1
symmetry of x, y, and z. For example, interchanging x and y in E[x| x+y+z=1] to get E[y|y+x+z = 1] doesn't change its value.
 
  • #3
18
0
symmetry of x, y, and z. For example, interchanging x and y in E[x| x+y+z=1] to get E[y|y+x+z = 1] doesn't change its value.
I am not able to understand what you are refering to. Could you please explain. Thanks.
 
  • #4
111
1
Sure thing... If I understand the problem correctly, it doesn't really matter what the exact distributions of x, y, and z are. It only matters that they are identically distributed. Also, note that the condition x+y+z = 1 is unchanged by permuting the letters x,y,z. This, together with their being identically distributed means that

E{x|x+y+z=1} = E{y|x+y+z=1} = E{z|x+y+z=1}


Also, think about what E{x+y+z|x+y+z=1} would be, and remember the additive property of the expected value.

edit: think simple. no hard integrals needed, which was the first thing that came to my mind when I read the problem.
 
Last edited:
  • #5
111
1
I left something out....it is also important that x,y,z are statistically independent. If, for example, x and y were statistically dependent, but x and z were not, then that would create an asymmetry and you could no longer conclude that the expected values of x, y, and z were equal.
 

Related Threads for: Conditinal Expectation Problem

  • Last Post
Replies
3
Views
2K
  • Last Post
Replies
3
Views
1K
Replies
14
Views
13K
Replies
1
Views
17K
  • Last Post
Replies
3
Views
565
  • Last Post
Replies
1
Views
1K
  • Last Post
Replies
11
Views
4K
  • Last Post
Replies
7
Views
3K
Top