Does anyone have experience programming a maximum likelihood estimator
for nested logit with non-mutually exclusive groups at the second level
alternative? I have data where people decide on a calling plan (A,B, or
C) based on their estimated number of calls in a month (and other
unobserved characteristics). With this data, it is common for two
customers with identical usage to choose different calling plans.
Unfortunately, the nlogit command in stata requires that the first level
alternatives (number of calls) provide for mutually exclusive groups in
the 2nd level (plan choice). Hence, I need to program the ml estimator
to give me some additional flexibilty.
Thanks for your help.
Eric
Eric D. Wolff
Graduate School of Industrial Administration, Room 313
Carnegie-Mellon University
Tech and Frew
Pittsburgh, PA 15213
(412) 268-2939 (office)
(412) 268-6837 (fax)
email: [email protected]
*
* 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/
