Kono,
I think that Austin is right. Maybe it is better for your understanding:
-gen newtime=(periodstart-1966)/4- and -tsset ccode newtime-.
Rodrigo.
----- Original Message -----
From: "Austin Nichols" <[email protected]>
To: <[email protected]>
Sent: Monday, August 28, 2006 10:38 AM
Subject: Re: st: xtabond2: all variables are dropped due to collinearity
kono--
I'm sure David Roodman will reply, with actual expertise to offer, but
I will offer an untested hypothesis: Your -tsset- shows gaps--is it
possible that the lagged values don't exist (i.e. the last year of
data is not in the data, only a value from four years before)? In
that case, create a new time variable, e.g.
gen tetrad=periodstart/4
tsset ccode tetrad
or somesuch.
On 8/28/06, kono hisaki <[email protected]> wrote:
> I have tried to estimate Blundell-Bond estimator by xtabond2,
> . tsset ccode periodstart
> panel variable: ccode, 10 to 226
> time variable: periodstart, 1970 to 1990, but with gaps
> Thanks for anticipation.
> Hisaki
*
* 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/
*
* 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/
