st: Re: Additive logistic model

["Joseph Coveney" <jcoveney@bigplanet.com>]

Dorothy Bridges wrote:

I'm wondering whether there's an easy (read: not coding from scratch)
way to estimate an additive logistic model in Stata (like this, for
example: http://gking.harvard.edu/files/abs/trip-abs.shtml; an R
package can be found here http://sekhon.berkeley.edu/robust/).

Essentially, this would require mlogit to read three (or more)
variables representing choice for each observation, rather than just
one categorical variable.  For example: mlogit party1 party2 party3 x,
where each of the party variables records the proportion of votes
obtained by that party in a given district, and x is a vector of
regressors.  (If we had the vote and characteristics of each
individual, of course, this would not be an issue (though I suppose
then there would be bigger issues)).

Many thanks, as always, for your thoughts.

--------------------------------------------------------------------------------

Are you looking for something like -dirifit- or -fmlogit-?  -findit- will lead
you to more about them.

Joseph Coveney


*
*   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/