Hello,
I am interested in estimating a discrete choice model with the following
nesting structure-respondents in my dataset first choose whether or not to
undertake any training(0 or 1) and then,conditional on this training choice,
choose to be either employed or unemployed. Both the employment and unemployment
categories are populated under either training choice. My understanding is that
the nlogit estimation routine in Stata is inappropriate for this particular
example as it requires the final alternatives(employment and unemployment here)
to be mutually exclusive across the nests, defined here by training category. Is
this correct? If so, is an alternative way of carrying out this estimation? Thanks.
Murali Kuchibhotla
*
* 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/
