R help - Nov 2012 - Obtaining R-squared value in Logistic Regression

I do not see an R-squared value after preforming the glm regression.

Is there a separate command for this? 

Thanks



--
View this message in context:
http://r.789695.n4.nabble.com/Obtaining-R-squared-value-in-Logistic-Regression-tp4648954.html
Sent from the R help mailing list archive at Nabble.com.

On Nov 8, 2012, at 1:58 PM, hoguejm <hoguejm at gmail.com> wrote:
> I do not see an R-squared value after preforming the glm regression.
> 
> Is there a separate command for this? 
> 
> Thanks

You might want to look at the lrm() function in Frank's 'rms'
package on CRAN, which will provide Nagelkerke's pseudo R^2 as part of the
model output. However, be sure you understand what it means in the context of
logistic regression. It is not the same as the R^2 in OLS regression.

Regards,

Marc Schwartz

On 09/11/12 08:58, hoguejm wrote:
> I do not see an R-squared value after preforming the glm regression.
>
> Is there a separate command for this?
>
A logistic regression is fitted via maximum likelihood, not by least 
squares.
Unless I am very much mistaken (always a contingency to be considered!)
R-squared makes no sense at all in this context.

Read up on "deviance".

     cheers,

         Rolf Turner

On Nov 8, 2012, at 11:58 AM, hoguejm wrote:
> I do not see an R-squared value after preforming the glm regression.
> 
> Is there a separate command for this? 
Which pseudo-R^2 are you looking for?

-- 

David Winsemius, MD
Alameda, CA, USA