st: Ordered/multivariate probit

["Wiemann, Markus" <m.wiemann@fs.de>]

Dear all,

I'd like to run a test to check whether companies that have certain
provisions in contracts perform better over time. Assume the following
setting: 

The dependent variable may take one of three values:
-1 if the company improved its rating over time,
0 if it had a stable rating or 
1 if the company got downgraded.


I have many control variables but actually I am interested in two dummy
variables:
Has_Provision_A which equals 1 if the contract contains some provision
"A" and
Has_Provision_B which equals 1 if the contract contains some provision
"B". 0 in all other cases.
There are contracts that have neither provision "A" nor "B". There are
no contracts that contain "A" and "B" at the same time. 

I tried ordered probit (- oprobit -) regressions. First of all, do you
think this is the best approach? Second, I heard that multivariate
probit might be a more suitable solution. But I am neither sure whether
this is correct nor where to find it in Stata. Do you have any comments
on that?


Thanks and all the best
Markus

*
*   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/