Re: st: xtabond2- number of observations per group

[Nick Cox <njcoxstata@gmail.com>]

This is econometric hieroglyphics to me, but at a wild guess, if you
are using lag 2, then no lagged values are available for the first and
second observations in a panel. That would account for the minimum of
1. I don't know about the 8. There is likely to be some
misunderstanding there.
Nick
njcoxstata@gmail.com


On 6 September 2013 12:30, Filipa Alexandra Da Silva Fernandes
<f.fernandes.1@research.gla.ac.uk> wrote:
> Dear Statalist,
>
> I am using the command -xtabond2- in stata 12.1 to perform a one step system GMM.
> However, I cannot understand the number of observations per group that I am obtaining in my output.
> My dataset has  a minimum of 3 observations per group and a maximum of 9. However, in the below output, Stata  gives me a minimum of 1 and a maximum of 8 observations.
> Can you help me? Is there a way to understand why this is happening?
>
>
> Thanks,
>
> Filipa
>
>
>
>  xi: xtabond2 e el  k w land sl sgrowth i.year i.ind i.countrynew, gmm(el k w, lag(2 .
>> )) gmm(land sl sgrowth, lag (2 .)) iv(year* ind* countrynew*) small r nomata noconstant
>>
> i.year            _Iyeara2003-2005    (naturally coded; _Iyeara2003 omitted)
> i.ind             _Iind_1-2           (naturally coded; _Iind_1 omitted)
> i.country      _Icountry_1-2    (naturally coded; _Icountryne_1 omitted)
>
> Building GMM instruments.......
> Estimating.
> Performing specification tests.
>
> Dynamic panel-data estimation, one-step system GMM
> ------------------------------------------------------------------------------
> Group variable: id                              Number of obs      =    402175
> Time variable : year                            Number of groups   =     95740
> Number of instruments = 159                     Obs per group: min =         1
> F(30, 95740)  =  51855.03                                      avg =      4.20
> Prob > F      =     0.000                                      max =         8
> --------------------------------------------------------------------------------
>                |               Robust
>            e |      Coef.   Std. Err.      t    P>|t|     [95% Conf. Interval]
> ---------------+----------------------------------------------------------------
>           el |   .6566588   .0246946    26.59   0.000     .6082575      .70506
>              k |    .021753   .0169495     1.28   0.199    -.0114679    .0549739
>             w |  -.0153329   .0707296    -0.22   0.828    -.1539622    .1232964
>        land |  -.3344615   .0781881    -4.28   0.000    -.4877093   -.1812136
>       sl |  -.0459528    .024158    -1.90   0.057    -.0933021    .0013965
>        sgrowth |   .0262927   .0547286     0.48   0.631    -.0809748    .1335602
>    _Iyeara2004 |   3.232151   2.186755     1.48   0.139    -1.053865    7.518167
>    _Iyeara2005 |   3.244041   2.186569     1.48   0.138     -1.04161    7.529692
>        _Iind_2 |  -.2841066   .5094497    -0.56   0.577    -1.282622     .714409
>  _Icountryne_2 |  -1.939647   2.667379    -0.73   0.467    -7.167681    3.288387
> --------------------------------------------------------------------------------
> Instruments for first differences equation
>   Standard
>     D.(year ind country)
>   GMM-type (missing=0, separate instruments for each period unless collapsed)
>     L(3/.).(el sl)
>     L(2/.).(k land sgrowth w)
> Instruments for levels equation
>   Standard
>     year ind countrynew
>   GMM-type (missing=0, separate instruments for each period unless collapsed)
>     DL2.(el sl)
>     DL.(k land sgrowth w)
> ------------------------------------------------------------------------------
> Arellano-Bond test for AR(1) in first differences: z = -15.09  Pr > z =  0.000
> Arellano-Bond test for AR(2) in first differences: z =  -1.34  Pr > z =  0.182
> ------------------------------------------------------------------------------
> Sargan test of overid. restrictions: chi2(128)  =5906.37  Prob > chi2 =  0.000
>   (Not robust, but not weakened by many instruments.)
> Hansen test of overid. restrictions: chi2(128)  =1080.95  Prob > chi2 =  0.000
>   (Robust, but weakened by many instruments.)
>
> *
> *   For searches and help try:
> *   http://www.stata.com/help.cgi?search
> *   http://www.stata.com/support/faqs/resources/statalist-faq/
> *   http://www.ats.ucla.edu/stat/stata/
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/