Al Feiveson wrote:
>>Andrea - How many exact zeros do you have? This looks like a mixed
discrete-continuous distribution, requiring something like a
zero-inflated model. Also, do you have repeated measures; i.e. is there
a clustering variable involved here?<<
It could be one of those. I'm (slowly) working on a paper for a zero
inflation model on this. We had issues getting these to fit as it's hard
to tell a degenerate boundary beta mixed with an L-shaped one.
These concentration indices are, if I'm not mistaken, actually somewhat
discrete, too, but I could be wrong. I don't recall the exact formula
for the H-H index.
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/
