On Mon, Dec 17, 2012 at 12:32 PM, Laura R. <laura.roh@googlemail.com> wrote:
> Thank you all for your help. I am still a bit confused, because now I
> read that also with GLM homoscedasticity and normality of residuals
> are assumptions that have to be met. But I will research further on
> that type of models in order to find out whether this works better in
> my case than OLS.
Yes, as I'm stuck teaching a course titled GLM which is for "general
linear model" I always tell students that the terminology, like
everything else in Grad Center, is out of date.
A generalized linear model lets you switch the error distribution to
something like the gamma or inverse Gaussian, which would more
naturally accommodate the skewed errors. This puts the transformation
on the regression structure, not the data themselves.
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/
