[R] Logistic regression: At times correlation matrix of coe

[(Ted Harding)]

On 21-Jan-03 Prof Brian D Ripley wrote:
> Does anyone think that the current arrangement (use this scheme
> for more than 4 coefficients) is sensible?

Not particularly, I think.

> Surely the abbreviations are not ("(" for intercept?), and why is
> the diagonal being shown but the top row and last column have been
> omitted?   If the whole matrix was shown, the column labels could
> be omitted.

Agreed; and not only is the whole thing ugly, but it is visually
difficult to interpret.

> I'd much prefer symbolic.cor=FALSE to be the default.

Hear, Hear! I stubbed my toe on this myself a while ago, until some
kind souls on the list directed me to this option (lurking not where
I had been groping for it ... ). When one needs this matrix, it is
usually because one wants to get at the numerical values.

Ted.


--------------------------------------------------------------------
E-Mail: (Ted Harding) <Ted.Harding at nessie.mcc.ac.uk>
Fax-to-email: +44 (0)870 167 1972
Date: 21-Jan-03                                       Time: 09:19:22
------------------------------ XFMail ------------------------------