[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: invoking -test- for nested nbreg models with robust standard errors
I am running nested nbreg models, clustered on the id of the respondent. I would like to test whether my full model improves satisfactorily upon my constrained model. I realize that I cannot use -lrtest- to do this, and I have read the Stata FAQ on using the -test- command. My question is one of logistics - what is the syntax I use to achieve this task? The way I would do it with lrtest is the
nbreg y x1 x2 x3 x4, cluster(id)
est store A
nbreg y x1 x2 x3, cluster(id)
which we know is incorrect and doesn't work. Can someone tell me the equivilant commands using the -test- command? I can't figure out how to take into account multiple models with the test command, as it appears to work off of only one model.
This seems so frustratingly simple that I fully expect to hit my head with my hand when I read the answer! But I would really appreciate any help you can provide, because as of yet I have been unable to figure out how to do it.
* For searches and help try: