Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: multi-level nlogit model
From
Stefan Bernhard <[email protected]>
To
statalist <[email protected]>
Subject
st: multi-level nlogit model
Date
Sun, 11 Aug 2013 11:14:19 +0200
dear statalisters,
(Stata12)
it seems to me that the documentation for -nlogit- lacks detailed
information on how to specify multi-level models and I can't find the
right words to feed Google with to find the answers I'm looking for
(there are a few statalist entries on multilevel nlogit models but
none show any syntax on how to do this).
The syntax for -nlogit- includes ... [|| lev1_equation [||
lev2_equation ...]] ....
However, the part for -nlogitgen- says nothing about multiple levels.
How do I proceed to specify a multi-level model? The Example in the
[R] is quite nice, but only shows a two-level model. Do I need to set
the multiple levels with successive -nlogitgen-? Do I need to set all
levels in one -nlogitgen-? Or do I only need to set the bottom level
with -nlogitgen- and all other levels are then set in -nlogit-?
Furthermore, can I use -nlogitgen- to set the branches and multiple
levels based on already existing categorical variables?
E.g. On one level I have 3 types of internet tariffs: cable, dsl and
mobile, which are identified in a categorical variable. Instead of
having to specify all of the 150 different tariffs manually, can't I
use the categorical variable that already contains exactly the same
information?
I.e. suppose in the example
"nlogitgen type = restaurant(fast: Freebirds | MamasPizza, family:
CafeEccell | LosNortenos | WingsNmore, fancy: Christophers | MadCows)"
the variable "type" already exists and I just need to tell -nlogitgen-
to set the branch using that one.
hope you can - as always - help me
best, stefan bernhard,
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/