Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: xtset question


From   "Nick Cox" <[email protected]>
To   <[email protected]>
Subject   st: RE: xtset question
Date   Tue, 29 Jan 2008 20:59:37 -0000

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/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index