Variance of a vector product/sum combination

  • Thread starter nikozm
  • Start date
  • #1
50
0
Hi,

i am trying to find the variance of the following: y*(s+n), where y is a m \times 1 vector following a chi-squared distribution with 2k degrees of freedom, s is a m \times 1 vector following a Gaussian distribution with zero mean and unit-variance, and n is a m \times 1 vector following a Gaussian distribution with zero mean and variance z.

Any help would be useful
 

Answers and Replies

  • #2
statdad
Homework Helper
1,495
35
If y is [itex] m \times 1 [/itex] and both s, n are also [itex] m \times 1 [/itex], the product [itex] y \cdot (s + n) [/itex] is not defined: what are you trying to do?
 
  • #3
50
0
Sorry for this typo. Let y be a 1 \times m vector instead.

Thanks
 
  • #4
statdad
Homework Helper
1,495
35
Are they independent? and do you mean the variance/covariance matrix of [itex] s [/itex] is the identity matrix and that of [itex] n [/itex] is [itex] z [/itex] times the identity matrix?
 
  • #5
50
0
yes. and they are mutually independent random vectors. Do you have any clue ?
 

Related Threads on Variance of a vector product/sum combination

  • Last Post
Replies
2
Views
737
  • Last Post
Replies
12
Views
7K
Replies
10
Views
2K
Replies
4
Views
1K
  • Last Post
Replies
3
Views
1K
Replies
3
Views
2K
Top