[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: Ordered probit model with censoring
"ERSADO, LIRE" <L.ERSADO@CGIAR.ORG>
I am trying to estimate number of income sources(0, 1, 2,...) a given
household has access to at time t. I have 2 questions (1) although the
dependent variables (i.e.,the number of income sources) can be
ordered, do you think ordered probit is a good model to estimate it?
right censoring would be appropriate since there are only very few
with, say, more than 4 different income sources per household, is
there is a
way to incorporate censoring into ordered probit in STATA?
>>> This sounds a counted scale, and some model which treats it
as such is therefore suggested.
In addition, where's the censoring? The fact that a distribution
falls off rapidly doesn't make it censored.
* For searches and help try: