Statalist


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

RE: st: xtabond


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: xtabond
Date   Fri, 10 Aug 2007 12:38:54 +0100

I don't think you meant what you wrote. 

sort prov time 
by prov: generate t = _n 
tsset prov t 

or 

bysort prov (time) : gen t = _n 
tsset prov t 

is more likely to be what you meant. 

But while this may make a command 
happy to play with sone data, it ignores
much of the information in those data, 
and creates time equivalences that may not 
match the original (e.g. observations with 
t = 1 may well have different underlying 
-time-s). 

Otherwise put, this is not just a trick to 
get a command to play. It needs justification
in terms of the underlying science (or whatever). 

Nick 
n.j.cox@durham.ac.uk 

Mansour Farahani
 
> You can create another variable for time like
> 
> sort prov
> by prov: generate t=[_n]
> tsset prov t
> 
> where your id is the identification number for each entity in 
> the panel. In this way there should not be any gap then in 
> the data. then you and you can run xtabond.
> 

*
*   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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index