Re: st: Test of coefficient equality

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

At 10:04 AM 3/5/2004 +0100, Dimova, Ralitza wrote:
> Hi. This is probably a very silly question, but better ask than struggle! 
> Is there any way I can compare the coefficients from Multinomial logits 
> over time? The situation is as follows. I am running a multinomial logit 
> over choices A, B,C and D on data drawn from random samples in year1 and 
> year2. How can I compare whether the effect of variable X on say choice A 
> between years 1 and 2 has changed significantly? Thanks. Ralitza
A draft of what I think is a good way to proceed is currently hiding on the 
web at

http://www.nd.edu/~rwilliam/xsoc593/mlogit_interactions.pdf

It is pretty much the same procedure as I would use with OLS regression, 
except that the more cumbersome but also more appropriate likelihood ratio 
chi square tests are used instead of Wald tests.  Also, mlogit models would 
seem to offer a lot more opportunities for tweaking because of all the 
equations and coefficients involved (but I think you'd want to have a 
pretty good idea of what you were doing, and why, before you started making 
things too complicated.)

Something like this will likely find its way into course notes when I go 
over logit and mlogit models in a few weeks, so if anyone sees any problems 
please let me know.  I'm doing what to me seem like logical extensions but 
that doesn't mean they are optimal or correct.


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