The installation instructions for gee were none too specific about the need
to add a .First.lib function calling something to load the shared object
etc.
Without knowing the version of S-PLUS and the OS it is hard to be more
specific, but
.First.lib <- function(...)
dyn.load.lib(..., basename="gee")
should do it in 3.4, 4.x and 2000 (but not later).
On Fri, 27 Jun 2003, Macneill, Stephanie J wrote:
> Hello. I have run into some problems running an old program in Splus and was
> wondering if anyone could help.
>
> A while I downloaded Vincent Carey's GEE program for Splus and used it to
> run a series of analyses. I have recently returned to the data and have
> tried to re-run analyses, but keep on getting error messages. I am trying to
> run a model with unstructured correlation, binomial famimly and logit link.
> Below is a simple version of the regression I am performing with just one
> predictor, but I get the same error no matter how many variables I put in
> the model.
>
> >
> test<-gee(test99[,2]~test99[,4],test99[,1],na.action=na.omit,family=binomial
> ,corstr="unstructured")
>
> [1] "Beginning Cgee S-function, @(#) geeformula.q 4.13 98/01/27"
> [1] "running glm to get initial regression estimate"
> [1] -2.078042029 0.009338763
> Error in .C("Cgee",: "Cgee" is not a symbol in the load table
> Dumped
> Error was while calling subroutine "Cgee"
>
>
>
> Am I doing something obviously wrong here? Is this a problem with GEE? As I
> said, I haven't used the software in a long time, but I cannot even re-run
> models that were run initially without any difficulty.
--
Brian D. Ripley, ripley@stats.ox.ac.uk
Professor of Applied Statistics, http://www.stats.ox.ac.uk/~ripley/
University of Oxford, Tel: +44 1865 272861 (self)
1 South Parks Road, +44 1865 272866 (PA)
Oxford OX1 3TG, UK Fax: +44 1865 272595
|