Adjusted McFadden Pseudo R-Squared
Like its ordinary least squares analog, the adjusted McFadden Pseudo \(R^2\) penalizes the McFadden pseudo \(R^2\) as more terms are added to the model,
\[R_{Adj}^2 = 1.0 - \frac{\ln(L)-K}{\ln(L_0)},\]
where \(K\) is the number of estimated parameters in the model. This count includes the intercept and other estimated degrees of freedom in the model.
References
J. S. Long. Regression Models for categorical and limited dependent variables. Sage Publications, Thousand Oaks, CA, 1997.
D. McFadden. Conditional logit analysis of qualitative choice behavior. In P. Zarembka, editor, Frontiers in Econometrics, chapter Four, pages 104–142. Academic Press, New York, 1974.