Re: st: lrtest again for clarification

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

At 07:24 AM 8/25/2006, Rho YH wrote:
> Thanks for the advice, and I've read the article
> For clarification, please confirm this
> As I specified before,
> logistic A B C D E F G
> est store A
> logistic A B C D E F H I
> lrtest a .
> Model | Obs ll(null) ll(model) df AIC BIC
> -------------+----------------------------------------------------------------
> A | 1695 -556.4883 -311.149 9 640.298 689.2169
> .. | 1695 -556.4883 -318.3306 10 656.6611 711.0155
> ------------------------------------------------------------------------------
> Then . (the latter) is better as A - . < 0 ?
Both the BIC and AIC tests favor model A.

Incidentally, fitstat (part of the spost9 routines) is a handy 
program to have here, as it makes it even easier to do these sorts of 
comparisons and provides additional info.

One other question though - is there any particular reason why G is 
excluded from the 2nd model?  If it were there the models would be nested.

You may also wish to check out the various BIC-related programs 
written by Paul Millar.


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