Re: st: RE: Re: Lags and panel data

[Michael Hanson <mshanson@mac.com>]

Jenniffer,

Can you type -xtset- in the Stata Command window and then copy and  
paste the results into an e-mail reply?  My conjecture is that you  
have non-consecutive years, and the command -tsset ... , yearly- that  
you issued tells Stata that it should expect observations in  
consecutive years for the purposes of lagging, etc.

If you want to treat, for example, data in 2003 as the first lag of  
data in 2005, then you should use -xtset [i variable] [t variable],  
delta(n)-, where n = the number of years that separate the two years  
in your sample.  (Note: "n" in the -xtset- command should be an actual  
number, not the letter "n"!  In the example I gave, n = 2.)

Hope this helps,
Mike

On Jun 15, 2009, at 2:03 PM, Solorzano Mosquera, Jenniffer wrote:

> Yes. It does, but actually when I try without any additional  
> calculation
> or generation of variables... There Is no observations...
>
> xi: xtabond res2 capacita SFTPW USFTPW FTNPW TFTW ptemporales inno
> gradolic i.ISIC
>
>
> xi: xtabond res2 l(0/1).capacita l(0/1).ptemporales, lags(1)
> no observations
> r(2000);
>
> All variables are for two periods (years)
>
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu
> [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Martin  
> Weiss
> Sent: Monday, June 15, 2009 1:47 PM
> To: statalist@hsphsun2.harvard.edu
> Subject: st: Re: Lags and panel data
>
> <>
>
> Does -xtabond- not calculate the lags needed for its purposes on the
> fly?
>
> ***
> webuse abdata, clear
> keep n w k ys yr1980 yr1981 yr1982 yr1983 yr1984 year id xtabond n
> l(0/1).w l(0/2).(k ys) yr1980-yr1984, lags(2)
> ***
>
>
> HTH
> Martin
> _______________________
> ----- Original Message -----
> From: "Solorzano Mosquera, Jenniffer"  
> <JENNIFFERS@Contractual.iadb.org>
> To: <statalist@hsphsun2.harvard.edu>
> Sent: Monday, June 15, 2009 7:40 PM
> Subject: st: Lags and panel data
>
>
> > Hi!
> >
> > I've been trying to set my dataset as panel data with
> >
> > tsset idstd_2003 year , yearly
> >
> > And then try to calculate lags for different variables an use the
> > xtabond estimation method. But when I try to generate the lag for the
> > dummy variable "capacita"...
> >
> > g lagcapacita = l.capacita
> >
> > ...I just get missing values.
> >
> > I have a panel of 1173  firms for 2 years.
> > Could anyone tell me what it could be  possibly wronged? Thanks!
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/