problem with cov(X)?

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

problem with cov(X)?

Daniel X. Pape

Hi -

I have been playing around with the statistics functions, and I have
noticed some weird output from cov(X)... Has anyone else noticed this,
or is it just me?

I can send a more detailed message if anyone thinks they want to help...


Dan

--
                 |   Community Architectures for Network Information Systems
Daniel X. Pape   |   National Center for Supercomputing Applications
                 |   University of Illinois - UC


Reply | Threaded
Open this post in threaded view
|

Re: problem with cov(X)?

Daniel X. Pape

John Day writes:

> I don't know if I can help you, but, since I use cov(X) a lot,
> I'd be interested in what you think is wrong with it.

I'm sorry everyone. My mistake.

I thought that the output from cov(X) was off, but that's because I
didn't look at the source before mailing. When I did, I realized that
Octave's cov(X) function was dividing things by (sample size - 1) and
when I was calcuating the covariances manually, I was only dividing by
(sample size).

So, it was not wrong - I was.


Thanks for the help.


Dan