On 15/09/2009, at 9:11 AM, Maria Sagot wrote:
> Dear all,
> I'm performing poisson regressions using the glm(family=poisson) and
> zero-inflated regressions using the zeroinfl from the package pscl.
> However
> any of those give an R^2 and I was wondering if there is a way to
> get it.
R^2 is meaningless for such models. They are fitted via maximum
likelihood, not least squares.
cheers,
Rolf Turner
######################################################################
Attention:\ This e-mail message is privileged and confid...{{dropped:9}}