Dear Nick,
I always appreciate your assistance.
Since "year" is only even numbers,
year/2 is integer.
I will try not to make any gap
for any one of the panels.
Thank you.
Yoshiro
--- Nick Cox <[email protected]> からのメッセージ：
> I am not well informed on -xtpcse- but I note that
> your time variable will be an integer only if year
> is an even number. In general, time variables should
> be
> integers.
>
> Nick
> [email protected]
>
> Yoshiro Nagao
>
> > I tried to analyse a biannually
> > sampled data by xtpcse
> > (longitudinal Prais-Winsten model)
> >
> > .gen time_seq = year/2
> > .tsset person time_seq
> > .xtpcse outcome exposure
> >
> > no time periods are common to all panels, cannot
> estimate
> > disturbance
> > covariance matrix using casewise inclusion
> > r(459);
> >
> > Does this error message mean that
> > for each sampled year,
> > even a single missing person
> > is unacceptable?
> *
> * 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/
>
--------------------------------------
GANBARE! NIPPON!
Yahoo! JAPAN JOC OFFICIAL INTERNET PORTAL SITE PARTNER
http://pr.mail.yahoo.co.jp/ganbare-nippon/
*
* 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/
