Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
| From | Nick Cox <n.j.cox@durham.ac.uk> |
| To | "'statalist@hsphsun2.harvard.edu'" <statalist@hsphsun2.harvard.edu> |
| Subject | st: RE: comparing different means using ttest |
| Date | Thu, 16 Dec 2010 18:02:16 +0000 |
Maarten pointed out two misconceptions here. I think there are several others. If you are using -regress- as here then with standard procedures the mean residual is automatically zero. In any case the mean residual tells you nothing whatsoever about homo- or heteroscedasticity, any more than the mean of anything tells you about its variability.