Re: st: Robust OLS and adjusted R-square

[Richard Williams <Richard.A.Williams.5@nd.edu>]

At 02:11 AM 5/6/2004 +0100, Mark Schaffer wrote:
> Ayako,
>
> Quoting Ayako Saiki <smap@brandeis.edu>:
>
> > Dear all,
> >
> > If anybody can help me on this, that'll be great - I ran OLS
> > with robust standard errors (reg.... , robust), but it does not
> > provide adjusted R-square.  Can anybody tell me what is causing
> > this?
> >
>
> It should be there, if you look in the right place:
>
> . sysuse auto, replace
> (1978 Automobile Data)
>
> . regress price mpg, robust
>
> Regression with robust standard errors                 Number of obs
>
> <snip>
>
> . ereturn list r2_a
> scalar e(r2_a)     =  .2087437291901012
Also -- note that the R^2 and adjusted R^2 values are the same regardless 
of whether or not you use robust standard errors.  So, if you also run 
regression without the robust option the value is already reported for you.

This raises an interesting question: when using robust standard errors, is 
adjusted R^2 not reported (a) because there is something that makes it 
inappropriate when using robust standard errors, or (b) because whoever 
wrote the programming code did not bother to include it in the output?

The omission of the Anova table when using robust standard errors is quite 
intentional by the way.  From p. 337, Stata 8 Reference Manual N-R, "When 
you specify robust, regress purposefully suppresses displaying the ANOVA 
table, as it is no longer appropriate in a statistical sense even though, 
mechanically, the numbers would be unchanged. That is, sums of squares 
remain unchanged, but the meaning you might be tempted to give those sums 
is no longer relevant. The F statistic, for instance, is no longer based on 
sums of squares; it becomes a Wald test based on the robustly estimated 
variance matrix."

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/