[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: RE: help needed for fixed effects Negative Binominal regression model
> -----Original Message-----
> From: firstname.lastname@example.org [mailto:owner-
> email@example.com] On Behalf Of Xiaojin Chen
> Sent: Tuesday, March 07, 2006 12:55 PM
> To: firstname.lastname@example.org
> Subject: st: help needed for fixed effects Negative Binominal regression
> Hello everyone:
> I was just added to this list today and would like get some help about
> effects negative binominal regression model.
> I will be using a longitudinal data such as Adolescent Health. If I want
> to run
> a fixed effects negative binominal regression model, how should the data
> structured? Add Health used the wide format, with each individual having
> line and three variables measuring the same concept at three time points
> Health has three waves so far). I guess I should change it to the long
> that is, each time point should have one line and each individual should
> three lines. Can I re-structure the data in Stata?
Yes. See [D] -reshape- Also, the FAQ by Nick Cox "I am having problems
with the reshape command. Can you give further guidance?" located at
might be helpful.
> Second, I have no experience in STATA. If anyone has experience using
> effects Poisson or negative binominal model, can you share some of the
> with me? I checked the manual and it looks it's not too hard, but any help
> be appreciated.
The basic syntax is:
-xtpoisson y x, i(id) fe-
-xtnbreg y x, i(id) fe-
You might also try the dialog boxes.
* For searches and help try: