st: Profile analysis in Stata 11

["Liam O'Brien" <lobrien@colby.edu>]

Hi all,

 

I'm trying to perform a simple profile analysis in Stata on a dataset
containing four response variables (y1, y2, y3, y4) and a single grouping
variable (group).   

 

Here are the commands I'm issuing:

 

/* preliminary MANOVA */

manova y1 y2 y3 y4 = group

 

/* test for parallelism */

mat c1 = (1,-1,0,0\0,1,-1,0\0,0,1,-1)

manovatest group, ytrans(c1)

 

/* test for group differences */

mat c2 = (1,1,1,1)

manovatest group, ytrans(c2)

 

These steps work just fine, but when it comes to the test of flatness, I'm
not able to get anything to work.  According to the UCLA ATS help page
(http://www.ats.ucla.edu/stat/stata/faq/profile.htm) something like this
should do the trick:

 

mat xm = (1,0,0)

manovatest, test(xm) trans(c1)

 

However, I'm getting a an error that says the constraint 1.group=0 was
dropped.   I tried replicating the analysis on the UCLA page and got the
same error.  I'm wondering if there might be some change in Stata 11 that is
causing this not to work.  Any suggestions would be very much appreciated.
Thanks!

 

 

Liam



*
*   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/