[R] Negative intercept in glm poisson model

[Douglas Bates]

Peter Dalgaard wrote:
> Douglas Bates <bates at stat.wisc.edu> writes:
>>Federico Gherardini wrote:
>>>I'm trying to fit a glm model using family=poisson(link =
>>>"identity"). The problem is that the glm function fits a model with
>>>a negative intercept, which sounds like a nonsense to me, being the
>>>response a Poisson variable.
>>
>>Not really.  The negative intercept is on the scale of the linear
>>predictor.  The expected response, which is the exponential of the
>>linear predictor, is always positive.
> 
> 
> Not when the link is "identity"...! However, in that case, sufficient
> extrapolation will always lead to negative values, unless the slope is
> exactly zero. Whether a negative value is a problem for the intercept
> depends on whether it corresponds to a relevant extrapolation of data.
> 

Yes, of course.  Must learn not to post to a public list before the 
second cup of coffee.