Questions regarding nlcom.
I used nlcom to calculate RRR and CI's of an ologit regression using
gllamm for multilevel models.
1) My sample was not a random sample. It oversampled certain
populations. Can I still use nlcom to calculate CI's on predicted
probabilities in multilevel ordered logistic regresssions using gllamm?
2) Assuming the answer is yes...Having used the syntax that M.Ingre so
kindly provided...can I add the effect of another covariate into the
model. i.e.
calculate the predicted probability of levels of income
if female=1 vs if female=0.
now calcuate the predicted probability of levels of income
if female=1 and owns car=1 vs. if female=0
the code works...but is it ok methodology.
thanks.
sanae
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/