Hallo
I need some help with the output provided by gam.check after a gam fit (using
the package mgcv).
To give a brief description of my data, I have
claims: a vector of values, which include NA's and one large negative value
- otherwise all positive (55 values in total that are not NA).
origin: a factor with 10 levels
j : taking the values 1, 2, ...., 10
I have fitted a gam, with:
> library(mgcv)
> modelg <- gam(claims ~ origin + s(log(col)), family = quasipoisson(),
+ subset=!is.na(claims))
(I have obtained a work-around of the quasipoisson family to deal with the
negative value.)
After this I tried the following:
> gam.check(modelg)
This gives a plot in the usual way (that looks fine, except for the
"outlier" which was a negative value), but returns the following
message:
fit method: deviance based outer iter. - newton, exact hessian.
step failed after 1 iteration.
gradient range [9.136047,9.136047] (score 994.1401 & scale 733.0711).
Hessian positive definite, eigenvalue range [42.98903,42.98903].
What does this mean? Is there a problem?
Any help is much appreciated.
Thanks.
Jacob L van Wyk
Dept of Statistics
University of Johannesburg
P O Box 524
Auckland Park 2006
South Africa
Tel: +27-11-559-3080
Fax: +27-11-559-2832
Cell: +27-82-859-2031
[[alternative HTML version deleted]]
