st: Question about xtabond and lags

[Marcello Pagano <pagano@hsph.harvard.edu>]

For Brindusa:

BRINDUSA ANGHEL wrote:

> > On 10/24/05, BRINDUSA ANGHEL <abrindu@gmail.com> wrote:
> >    
> >
> > > Dear Stata users,
> > >
> > > I am working with a panel dataset for 7 years.
> > > I am trying to estimate with xtabond the following specification:
> > >
> > > xtabond y x1 x2 x3, lags(1) artests(2)
> > >
> > > However, I suspect that my x3 variable is endogeneous, so I try the following:
> > >
> > > xtabond y x1 x2, lags(1) artests(2) pre(x3, endo).
> > > So, when I run this estimation, STATA will use 5 lags (7-2) of x3 as additional instruments, right?
> > >
> > > If I want to use only two lags of the endogeneous variable as instrumens, I understood that I have to write:
> > > xtabond y x1 x2, lags(1) artests(2) pre(x3, lags(.,2) endo).
> > > This means that Stata will use x3 in year 3 and x3 in year 4 as instruments for x3?
> > >
> > > If anyone can help me, I would really appreciate it.
> > > Thank you very much!
> > >
> > > Brindusa Anghel
> > >      
> >    
*
*   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/