As you say, data of the form you describe cannot be handled by -xtset-.
You have a choice between modifying your data so that they can be
handled by -xtset- or doing something quite different outside the realm
of -xt-. I doubt that there is a third way.
For example, what do your repeated time values within a panel actually
mean?
Possibly you really have different panels in some sense, so such panels
must be subdivided. Or you could -collapse- those repeated values to
some appropriate summary.
A better answer may require more details than you give.
Nick
[email protected]
-----Original Message-----
From: [email protected]
[mailto:[email protected]] On Behalf Of Sharif S. Aly
Sent: 29 January 2008 20:51
To: [email protected]
Subject: st: xtset question
Dear list,
xtset will not allow declaring more than 2 variables, a panel and a time
variable, the latter has to be understandably not repeated within a
panel. What if there is more than 1 panel variable with repeated time
values within any 1 of those panels but not within all panels. In my
example, a 4 level mixed effects negative binomial model (hence I needed
to xtset my dataset to use xtnbreg) with time as the lowest factor
level.
Any solutions are greatly appreciated?
*
* 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/
