|
|  |
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: Obtaining Predicted Probabilities using gologit
At 08:40 AM 6/14/2006, Nick Winter wrote:
my -oprobpr- might be of service here (although it doesn't work, I
think, after gologit. But you could hack into it potentially).
Also check out -adjust-
--Nick
As luck would have it, I had an exchange with Stata tech support a
while back about the use of adjust with the pr option after
gologit2. Here is what I wrote:
***I wrote a customized predict command (gologit2_p) for my gologit2
program so I could compute predicted probabilities. Alas, adjust
still won't let me use the -pr- option after running gologit2. Just
skimming through adjust.ado, it appears that your program needs to be
officially "blessed" (i.e. explicitly mentioned in the adjust.ado
code) for the pr option to work. Is there a way to get your program
blessed, or to otherwise work around
this? Thanks.***
Stata's response:
***The -pr- option of -adjust- is not passed on to -predict-. As
with -exp-, -adjust- does the transform itself; a consequence of this
is that -adjust- will only allow the -pr- options for estimation
results it recognizes.***
So in general, many post-estimation commands will not work or will
have only limited functionality with user-written commands. You
therefore may have to improvise a bit to get what you want. (As a
sidelight, my mfx2 program does many of the same things as
margeff. margeff is often much faster; but mfx2 works with more
programs. I wrote mfx2 partly because margeff did not work with my
programs and some other commands that were added in Stata 9).
-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
FAX: (574)288-4373
HOME: (574)289-5227
EMAIL: [email protected]
WWW (personal): http://www.nd.edu/~rwilliam
WWW (department): http://www.nd.edu/~soc
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
