As far as I know, the Hosmer-Lemeshow test is not a good choice, as it depends on the groups to be formed to compute the statistic. The design and/or hmisch packages by F.Harrell should include some alternative GoF statistics, but I didn't try.
Some weeks ago there has been a similar message about GoF test with sparse date in logistic regression model. You could look for it and contact the author.... best, vito ----- Original Message ----- From: Andi Felber <[EMAIL PROTECTED]> To: <[EMAIL PROTECTED]> Sent: Thursday, July 17, 2003 3:53 PM Subject: [R] Hosmer- Lemeshow test > dear all, > > i made some analysis with logistic regression. do you know if there is > already a function available for the hosmer-lemeshow fitness test? > > thank you very much for the help > > best wishes > > andi > > ______________________________________________ > [EMAIL PROTECTED] mailing list > https://www.stat.math.ethz.ch/mailman/listinfo/r-help ______________________________________________ [EMAIL PROTECTED] mailing list https://www.stat.math.ethz.ch/mailman/listinfo/r-help
