Re: [Help-gsl] Where oh where did the correlation cofficient go?

[Jack Denman]

On Wed, 2007-05-23 at 10:25 +0100, Brian Gough wrote:
> At Mon, 21 May 2007 23:11:15 -0700,
> Jack Denman wrote:
> > I discovered this great library, but in the linear regression code I
> > cannot find a way to get the correlation coefficient "rho" or r that
> > indicates the degree if fit. Am I missing something?
> 
> Hello,
> 
> We return the covariance matrix which gives more
> information, you can calculate the correlation
> coefficient from it as rho = c11/sqrt(c00*c11).
> 


Thank you for the information and taking the time to write back. I will
look up the significance of the covariance matrix on my own.


-- 
Jack Denman <address@hidden>