I have found the problem and a solution.

The problem comes from the functions addterm.glm() and dropterm.glm() from MASS package. They use extractAIC() without transmitting the ... argument:

aic <- aic + (extractAIC(object, k = k)[2L] - aic[1L])

I replace the call with:
aic <- aic + (extractAIC(object, k = k, ...)[2L] - aic[1L])

and I create dropterm.glm() and addterm.glm() in global environnement.
I add:
dropterm <- dropterm.lm <- dropterm.glm
addterm <- addterm.lm <- addterm.glm

I copy also stepAIC from MASS package to global environnement.
I detach MASS package, and it works:

> stepAIC(G1, criteria="AICc", steps = 1)
Start:  AIC=70.06
x ~ A + B

       Df Deviance    AIC
- A     1   48.506 68.355
<none>      47.548 70.055
- B     1   57.350 70.867

Now stepAIC can use AICc

Marc

______________________________________________
R-help@r-project.org mailing list -- To UNSUBSCRIBE and more, see
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.

Reply via email to