I am trying to use the glm.nb proceedure in the new MASS library. I have
count data (of snails) taken from a number of habitats (artesian springs).
I want to model the abundances, including an effect of the organic content
of the sediment. When I try to fit a model like:
nb.model.p <- glm.nb(abundance~Organics.per.wt)
I get the error message:
Error: no valid set of coefficients has been found:please supply starting
values
I looked in the help for the glm proceedure and found the "start()"
statement for specifying coefficients. However, I could not find much
documentation on what should be in start. I searched the R website and also
didn't find much detail on start(). On a lark I tried the following, just
to see if I had the syntax correct:
nb.model.p <- glm.nb(abundance~Organics.per.wt,start=(1,1))
which I did not. I had a look in Venables and Ripley, but could not find an
example there either. So if anyone can point me in the direction of the
proper documentation or provide any helpful input I would greatly appreciate
it.
Thanks,
Chris Wilcox
Dr. Chris Wilcox
Research Fellow
Department of Zoology and Entomology
University of Queensland
St. Lucia 4072 QLD
Australia
email: c.wilcox at uq.edu.au
tel: 61.7.3365.1686
fax: 61.7.3365.1655
website: www.uq.edu.au/~uqcwilco
"In God we trust all others must bring data" W. Edwards Deming
