[R] diagnostic information in glm. How about N of missing observations?

[Frank E Harrell Jr]

On Fri, 19 Dec 2003 12:53:09 +0000
Barry Rowlingson <B.Rowlingson at lancaster.ac.uk> wrote:

> Frank E Harrell Jr wrote:
> 
> > Fitting functions in the Design package tell you how many observations
> > were deleted due to each variable in the model.  They generalize the
> > na.action component stored in the fit object.
> 
>   Surely accessing components of an object is generally a bad idea on 
> O-O design, and there ought to be a _method_ that returns which 
> observations have been omitted. Perhaps an 'omitted' function?

That's perhaps a matter of taste, but anyway I always want to see such
information, so it is always printed by the fitting function's print
method.  The fit object in Design has an element na.action which has an
element nmiss with the details.  You can also set an option for further
details, which then are stored in the element na.detail.response.  This
information provides brief data on the distribution of the response
variable stratified by missingness status of each predictor.

Frank

---
Frank E Harrell Jr   Professor and Chair           School of Medicine
                     Department of Biostatistics   Vanderbilt University