Nick Cox wrote:
>I'd rather try other two-parameter distributions first, or equivalently
other transformations. For example, fitting a two-parameter gamma is
roughly equivalent in some senses to working on a cube root scale.<
First of all, interesting I didn't know that the gamma lined up with the
cube root scale! You learn something new every day.
Second of all, Stata (and other stats programs) have specialized
software for positive definite data that are skewed. These are better
known as survival analysis programs, e.g. -streg-, which you can happily
use to fit many parametric regression models for completely observed
data simply by not censoring anything. Similarly, -stcox- could be used
for a semi-parametric model. :)
JV
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
