Dear all,
I am estimating a nested multinomial logit.
The first choice is among 5 choices, the second one is among j
alternatives, which are the same among the 5 choices (a-j for choice 1,
a-j for choice 2, etc).
The problem is that when I define the tree structure I have as top level
just the 5th choice and at the bottom level the j alternatives. In sum,
the observations of the whole sample are cumulated in the 5th choice.
Is it possible that Stata does not allow a second choice among equal
alternatives?
If this is the case is there a way to allow it?
If anyone can help me I would really appreciate it.
Many thanks.
Best wishes,
Annalisa
--
*
* 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/
