Re: [Gretl-users] About the output of a Logit model
On Fri, 29 Jan 2010, yinung at Gmail wrote:
I run a logit model and I found that the key output of the logit
model
includes McFadden R-squared. This look great. However, I also saw an
"Adjusted R-squared" reported. I am a little bit confused about it. Does the
"Adjusted R-squared" have any particular definition or meaning for a
(regular) logit?
The "Adjusted R-squared" in this context is just the McFadden
value with the same degrees-of-freedom adjustment that's
commonly applied to the regular R^2. This is given in some other
software (stata? can't remember right now).
Allin