Re: st: definition of pseudo R^2 for dprobit or probit

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

At 04:21 PM 10/27/2003 -0500, Santiago Bazdresch wrote:
> Hi.
> I cant seem to find the exact definition of the statistic "pseudo R^2" 
> that stata 7 reports with probit or dprobit output.
> anybody know it?
> how should I find it?
> thanks.
Since it has been a few hours and no one more authoritative than me has 
chimed in yet -- In other Stata routines, e.g. Logistic regression, the 
pseudo R^2 used is also known as McFadden's R^2.  I think that is the same 
as what is used in Stata Probit, but I could be wrong.   A good discussion 
of this and other R^2 analogs can be found in Scott Menard's "Applied 
Logistic Regression Analysis 2nd Edition" (paper 106 in the green sage 
series on Quantitative Applications in the Social Sciences).  If you're 
desperate for something quick on this, I have class notes (starting around 
p. 6) that discuss this stat at

http://www.nd.edu/~rwilliam/xsoc593/lectures/l23.pdf

It is hard to write out the formula here, but it is basically the 
chi-square for the current model divided by the chi-square for a model that 
has only the intercept included.

If I am horribly misleading Santiago here, I hope somebody else will step 
in and correct me!  RW


-------------------------------------------
Richard Williams, Associate Professor
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/