I have a program I am (almost) ready to unleash on the world, but since it
is fairly complex (roughly comparable to -mlogit-) I wouldn't mind getting
a few more beta-testers first. I'm of course interested in bug reports,
but I'm also interested in comments on ease of use and clarity of
documentation (like, can anybody besides me understand what the program
does?) I'll also be presenting on this at the Boston meetings in July.
The program is called -gologit2-. If you are familiar with the original
gologit program, ologit, the proportional odds assumption made by ologit,
and/or the idea of partial proportional odds models, this program may be of
interest to you.
Those brave enough to try it can locate and install the program by typing
the following from within Stata:
Besides the program-related files, there is a manual (of sorts) entitled
gologit2.pdf; you can either get it as an ancillary file or download it
directly from
Here is the description of the program. Again, any feedback would be
appreciated. RW
gologit2 estimates generalized logistic regression models for
ordinal dependent variables. A major strength of gologit2 is that it
can also estimate two special cases of the generalized model: the
proportional odds model and the partial proportional odds model. Hence,
gologit2 can estimate models that are less restrictive than the
proportional odds/parallel lines models estimated by ologit (whose
assumptions are often violated) but more parsimonious and interpretable
than those estimated by a non-ordinal method, such as multinomial logistic
regression (i.e. mlogit). Other key strengths of gologit2 include options
for linear constraints, alternative model parameterizations, automated
model fitting, Stata 8.2 survey data (svy) estimation, and the computation
of estimated probabilities via the predict command. gologit2 is inspired by
Vincent Fu's gologit program and is backward compatible with it but offers
several additional powerful options.
gologit2 is still in the beta testing stage. Comments are welcome.
