[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: Endpoint problem with random effects
I am fitting random-effect model with gllamm routine in stata. I am using non-parameteric estimations
with mass-point, the problem is that the standard error of mass-point is so high, probabaly it because
it not certain about location. I am tingking about problem about endpoint(plus minus inf.). Any of you
had expericences same problem before?
How can I include endpoint(mover-stayer) into random effects especially in gllamm routine.
For your informations I am using binomial distributions with logit link, the problem with my data is
high mover-stayer proportions. Any ideas???
* For searches and help try: