I have collected the following data before and after a treatment for a
number of sites:
Before TreatmentAfter Treatment
Casesuccess failure success failure
1 nbs1nbf1nas1naf1
2 nbs2nbf2
If the Dependent Variables is Unordered ,why not use Multinomial
Logistic Regression.see
http://gking.harvard.edu/zelig/docs/_TT_mlogit_TT__Multino.html
Hope this helps.
2006/1/11, S.C. Wong [EMAIL PROTECTED]:
I use mnp to run a multinominal probit regression model, but the summary
doesn't contain
I use mnp to run a multinominal probit regression model, but the summary
doesn't contain the model statistics, such as the log-likelihood and degree
of freedom, for the assessment of the goodness-of-fit of the fitted model.
Is there any way that I can generate these statistics for the fitted