[Rd] Tolerances in glm.control

[Thomas Lumley]

On Fri, 9 May 2003, Martin Maechler wrote:

>
> Being on this topic:  What about
> enhancing  summary.glm() and changing print.summary.glm() quite
> a bit such that
>
> o   summary.glm() for back compatibility still computes the Wald tests
> 		  but also does all the "drop1" tests for each coefficient
> and
>
> o  print.summary.glm() would print these tests instead of the
> 		       (too often misleading) Wald ones.
>

I think a bigger problem is confidence intervals.  It's easy enough to get
score tests or (for likelihood-based glms) likelihood ratio tests, but
confidence intervals take more work.  It's true that summary.glm() doesn't
explicitly compute confidence intervals, but it does give the standard
errors, whose only purpose AFAICS is Wald-based confidence intervals.

There is profile.glm() in MASS, which I think needs to be better known,
but it seems trickier to do score-test confidence intervals (I spent one
plane flight trying to do this for the survey package).

	-thomas