Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Re: Question about xtabond and xtreg


From   "Scott Merryman" <smerryman@kc.rr.com>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: Re: Question about xtabond and xtreg
Date   Thu, 12 Dec 2002 17:45:27 -0600

----- Original Message -----
From: "You-Ta Chuang" <You-Ta.Chuang@mail.atkinson.yorku.ca>
To: <statalist@hsphsun2.harvard.edu>
Sent: Thursday, December 12, 2002 5:21 PM
Subject: st: Question about xtabond and xtreg


> Hi all,
>
> I am playing with the cross-sectional,time series data I have. I couldn't
quite
> figure out something here.
>
> When I used the xtreg, the result table showed
> Number of obs: 4248
> Number of groups: 246
>
> However, when I ran the xtabond (no additional specification, i.e., lag 1
only),
> the result table showed
> Number of obs: 3756
> Number of groups: 240
>
> I am not aware of any missing data in my sample. The time period is 19 yrs.
What
> I don't understand is why the numbers of obs and groups are so different
between
> xtreg and xtabond. I would appreciate it very much if someone could give me
some
> advise here. Thanks!
>
With xtabond, lags(1) the first two observations (per group) are lost due to
lags and differencing.  I imagine that 6 of your groups only had two
observations so they would be lost plus 2 observations per group for the
remaining groups.  Which will leave you with:

4248 - 2*240 - 2*6 =  3756.

Since xtreg does not difference the data or impose lags so no observations are
lost.

Scott


*
*   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