Thanks for the off list plot. I don't see anything untoward in the
residual plots here - a good way of getting a feel for what residual
plots should look like if the model is ok, is to simulate some data from
your fitted model using rnbinom and the fitted values, refit the model
to the simulated data, and take a look at the resulting residual plots.
Doing this a few times gives you a feel for the range of plots to expect
if everything is ok.
best,
Simon
On 06/12/12 06:21, Tania Mendo Aguilar wrote:
Dear All,
I am fitting scallop count data to negative binomial GAMs. I have two
significant parameters that explain 43%of the deviance. The adjusted r square
is 0.25. The gam.check function gives me the figure attached. In the graph of
linear predictor vs. residuals there seems to be more negative residual values
than positive. Is that telling me that the fit is underestimating the response?
Can I accept this model?
Tania Mendo
PhD Candidate | Institute for Marine and Antarctic Studies (IMAS)
University of Tasmania
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
--
Simon Wood, Mathematical Science, University of Bath BA2 7AY UK
+44 (0)1225 386603 http://people.bath.ac.uk/sw283
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
