[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: Validity of R Square with -nl- Regressions

From   Maarten buis <>
Subject   RE: st: Validity of R Square with -nl- Regressions
Date   Fri, 20 Jun 2008 14:57:28 +0100 (BST)

Nice, but just to make sure my main point does not get confused: -nl-
reports Rsquare, not a pseudo-Rsquare. 

-- Maarten

--- Richard Williams <> wrote:
> Continuing this sidelight - I give several formulas for OLS and their
> logical Pseudo R^2 analogs at
> For OLS, the formulas all give the same results, but the Pseudo R^2 
> stats all give different numbers.  Counter to the claims of some, I 
> don't think logic alone clearly favors one Pseudo R^2 measure over
> another.
> One thing I do find kind of remarkable is that the Maximum Likelihood
> R^2 (aka Cox-Snell) isn't just a logical analog - it is the exact 
> same formula for OLS R^2 and Pseudo R^2.  That might be why, as Nick 
> says, some may think this is the "obvious" measure.
> I'm not quite sure why McFadden's Pseudo R^2 came to be blessed by 
> Stata, but I suppose it is as good as any.

Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room Z434

+31 20 5986715

Sent from Yahoo! Mail.
A Smarter Email
*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index