Hi,
I am modeling health care expenditures using glm.
Furthermore, I am using the model to generate
predictions of health care expenditures, so the
variance component is quite important to me. A "Park
test" of my data has suggested that modelling variance
as proportional to the ~1.7th power of the mean might
be most appropriate. Is there a way to specify
variance functions that are intermediate between the
Poisson family (variance proportional to the 1st power
of the mean) and the Gamma family (variance
proportional to the 2nd power of the mean)?
Doug
__________________________________
Do you Yahoo!?
The New Yahoo! Search - Faster. Easier. Bingo.
http://search.yahoo.com
*
* 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/
