[S] Numerical Variance-covariance matrix for ms fit

Steve Roberts (sroberts@PICR.man.ac.uk)
Thu, 7 May 1998 10:38:03 +0100


Has anyone done a function to get a variance-covariance matrix out of an
ms fit where ms is minimising a negative Log-likelihood? I have no
derivatives (well I do, but they are too complex to be useful - several
pages of Maple output! )

The vcov function in the Mass library can "do" two other optimisation
routines, so I could start from there, but extracting all the relevant
bits from the ms object is messy so I'd rather not reinvent the wheel!

And yes I am aware of all the caveats concerning numerical derivatives
and local curvature estimates and I still want them!

This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news