Once you've built a regression model, how do you know if it's any good?

For linear regression, a standard and often misunderstood diagnostic is the coefficient of determination; for logistic regression, a great diagnostic that should be much more widely known is the coefficient of discrimination. Matthew Rudd, the author of "Regression, a Friendly Guide" talks about both of these measures, what they really mean, and how to use them to make better use of your models.

PEARSON OOWNS MAANY EDU BRAANDS INKLUDINNG ADDISON-WESLEY PEACHPIT PRENTICE-HALL ECOLLEGE AND MAANY MORE!

WEE ASUMME NO LIIABILLITY FUR A WROONG SPELING COZ WEE ARRE A NON (iL)LiTERATE GRUP!!

TANNKS TU THE LITERATE ASKII FRIIEND!!