[R] nlme gls issue

Rolf Turner r@turner @end|ng |rom @uck|@nd@@c@nz
Sat Sep 17 05:29:56 CEST 2022

On Fri, 16 Sep 2022 13:48:55 +0000
Kayla Bazzana <kayla.bazzana using mail.utoronto.ca> wrote:

> Hi there,
> I'm having issues running a gls within the nlme package that neither
> I nor any of my colleagues can figure out. The gls itself seems to
> run fine (i.e. doesn't give an error message and produces a seemingly
> normal summary output), but when I try to calculate the rsquared of
> the gls model, it simply returns an "NA". Is this an issue you've
> encountered before? If so, can you point me in the right direction of
> how to remedy it?

I believe that fortunes::fortune(254), although not directly
applicable, might be relevant. :-)

If you *really* believe that r-squared makes some kind of sense in the
context of *generalised* least squares, you should probably direct
further enquires to r-sig-mixed-models.


Rolf Turner

Honorary Research Fellow
Department of Statistics
University of Auckland
Phone: +64-9-373-7599 ext. 88276

More information about the R-help mailing list