[R] Force coefficients in glm()

[Prof Brian Ripley]

On Mon, 15 May 2006, Dan Bebber wrote:

> Hello,
>
> I have a model
> glm(Y ~ X, family = quasipoisson(link = "identity"))
>
> I would like to vary the coefficient for X and observe the effect on the
> deviance.
>
> Is this possible?

library(MASS)
?profile
?profile.glm

(since this is just a profile likelihood).

-- 
Brian D. Ripley,                  ripley at 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