Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: R-squared measures proposed by Cameron and Windmeijer (1996) in stata
From
Ronan Conroy <[email protected]>
To
"[email protected]" <[email protected]>
Subject
Re: st: R-squared measures proposed by Cameron and Windmeijer (1996) in stata
Date
Fri, 25 Mar 2011 10:22:03 +0000
On 25 Mar 2011, at 09:21, Nick Cox wrote:
> You need to think about what it would mean in your field to have
> perfect predictions. It means that the data provide a complete
> description _and_ the model captures the generating process exactly.
> Or it would mean that you were using a model with too many parameters.
> In most observational fields even large datasets provide only partial
> data and the model is at best a caricature of the underlying process.
I would second Nick's observations, and add a reminder that you cannot explain 100% of variance if your measurement instruments contain error, which is, of course, what happens in real life and sociology.
The whole notion of R^2 is somehow comforting for those who don't want to have to think too carefully about models. People often say "these factors explained X% of variance". Of course, the statistical model isn't an explanation. Explanations link phenomena by (unobservable) causal links. Statistical models look for shared variation between measurements. Different idea.
Ronán Conroy
[email protected]
Associate Professor
Division of Population Health Sciences
Royal College of Surgeons in Ireland
Beaux Lane House
Dublin 2
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/