expected value of the esimated covariance matrix

Thread View: sci.math.stat

2 messages

[T] Back to Threads [V] Tree View [G] Group Articles

2 total messages Started by Behzad M Shahshahani Fri, 02 Oct 1992 16:36

#620

Author: Behzad M Shahshahani
Date: Fri, 02 Oct 1992 16:36

9 lines
895 bytes

This message didnot go through:

I forgot to mention that I am interested in the multivariate case.
Therefore here is my question again:
Assuming that the covariance matrix is estimated from a set of
multivariate normal observation, and the usual unbiased estimator S is
used, then what is the expected value of the inverse of matrix S ?

-behzad

Re: expected value of the esimated covariance matrix

#625

Author: Robb Muirhead
Date: Fri, 02 Oct 1992 17:24

6 lines
685 bytes

For a sample of size N from a MVN distribution

E(S**-1)={(N-1)/(N-p-2)}Sigma**-1

(p is the dimension, Sigma is the population cov matrix)

Thread Navigation

This is a paginated view of messages in the thread with full content displayed inline.

Messages are displayed in chronological order, with the original post highlighted in green.

Use pagination controls to navigate through all messages in large threads.

Back to All Threads