Re: st: OLS assumptions not met: transformation, gls, or glm as solutions?

["JVerkuilen (Gmail)" <jvverkuilen@gmail.com>]

In addition to what others have suggested you might consider quantile
regression (-qreg-) or using -poisson- with robust standard errors, as
per this blog post:

http://blog.stata.com/tag/log-linear-regression/
*
*   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/