Re: st: BICREG: BIC-based tests for individual effects

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

At 05:45 PM 2/23/2005 -0700, Paul Millar wrote:
> In R (and Splus) there is a routine written by Raftery that uses a BIC 
> based probability test for each individual slope for OLS and logistic 
> models.  Is such a thing also available in Stata?  Is there anyone else 
> interested in such a facility?
I'm not aware of any existing routines that do this, but it would be 
nice.  Zhiqiang Wang's -lrdrop1- routine does AIC tests for logistic 
regression models.  I suspect it wouldn't be too hard to modify it to do 
BIC tests.

Long and Freese's -fitstat- routine (part of -spostado-) will do BIC 
comparisons of models (both OLS and logit), but you have to specify each 
model yourself.  The -mlogtest- routine that also comes with -spostado- 
does chi-square tests of each coefficient in an -mlogit- model - again I 
suspect it wouldn't be too hard to modify it if someone wanted to.


-------------------------------------------
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/