RE: st: RE: Changing beta coefficients - w/ logistic regression

[Richard Williams <Richard.A.Williams.5@ND.edu>]

At 05:53 AM 7/22/2005 -0500, Scott Merryman wrote:
> Now -predict prob, p- works, in the sense it doesn't return an error, though
> it does not calculate the probability but rather the fitted values.  The
> probabilities can calculated by using the -invlogit- function, however I am
> not sure why -predict prob, p - is not calculating the probabilities.
Again, most peculiar, and potentially dangerous if you don't happen to 
notice that the p option didn't work.  If you turn trace on, you see that 
eventually a statement like this gets executed:

_predict prob2

That statement works correctly before -foo2- is run but not 
afterwards.  For some reason _predict does not know it should default to 
computing probabilities.  There are other values that could be ereturned, 
so that might be it, but I suspect there is just something "different" 
about -logit- and -mlogit-.

I'm assuming the whole point of this is to use post-estimation commands 
like -predict- with the new beta value.  Ergo, if you want predicted 
probabilities, either do like Scott says and use invlogit, or use a routine 
like -ologit- or -gologit2- that produces equivalent results to logit with 
a dichotomous dv and which works with the original foo.ado.  (I didn't 
consciously plan it that way, but there are a couple of little things 
-gologit2- can do that -logit- can't, and working correctly with foo.ado is 
one of them.)


-------------------------------------------
Richard Williams, Notre Dame Dept of Sociology
OFFICE: (574)631-6668, (574)631-6463
FAX:    (574)288-4373
HOME:   (574)289-5227
EMAIL:  Richard.A.Williams.5@ND.Edu
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/