Re: st: Re: Fixing an -ml model- syntax problem

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

At 05:19 PM 2/20/2007, Clive Nicholas wrote:
> I switched on the -eform- option in -estout- when in fact $\{delta}$ 
> is conventionally measured in log-odds, but it was still -0.20 under 
> -glogit- and 0.20 under -oglm-. However, your solution sorts that 
> out, as well as the fit. Maybe you should incorporate it into the 
> next version of -oglm-, Richard. That leaves my question (2) remaining.
Our messages are crossing here!  Yes, I probably could add Allison's 
delta to the output or include a post-estimation command for 
it.  But, it only applies to a special case, when the 
heteroskedasticity is believed to involve 2 groups.  oglm is much 
more flexible than that.  Personally i think Allison did a great job 
of presenting the problem but his solution has a number of weaknesses 
and limitations in it.

As far as your 2nd question goes, I'm not sure if I've now answered 
it or not.  If not, maybe you could rephrase it, since I'm not quite 
sure what you were asking.


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