| Logistic Regression |
The Summary of Fit table contains summary statistics for the fit of the model including values for Deviance and Pearson's Chi-Squared statistics. These values contrast the fit of your model to that of a saturated model that allows a different fit for each observation. If the data are sparse in the sense that most observations have a different set of explanatory variables, as in this set of data, then the quality of these measures is likely to be poor. Inferences drawn from these measures should be treated cautiously.
Copyright © 2007 by SAS Institute Inc., Cary, NC, USA. All rights reserved.